RN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4.2 Implementação do modelo baseado em Rede Neural"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from keras import optimizers\n",
    "import pickle\n",
    "from sklearn import model_selection\n",
    "from tensorflow.keras.layers import Dense, Input\n",
    "from tensorflow.keras.models import Sequential\n",
    "from sklearn.model_selection import train_test_split\n",
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score, accuracy_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/proposicoes_ord_2013-2023.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['idProposicao'] = df['idProposicao'].astype('str')\n",
    "colunas = [col for col in df.columns if col != 'VirouLei'] + ['VirouLei']\n",
    "df = df.reindex(columns=colunas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.select_dtypes(include=['int64', 'float64'])\n",
    "y = X['VirouLei']\n",
    "X = X.drop(columns=['Ano','VirouLei'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Administração e Políticas Públicas</th>\n",
       "      <th>Desenvolvimento Social e Urbano</th>\n",
       "      <th>Direitos e Justiça</th>\n",
       "      <th>Economia e Desenvolvimento Sustentável</th>\n",
       "      <th>Infraestrutura e Gestão de Recursos</th>\n",
       "      <th>Influencia</th>\n",
       "      <th>Aliado</th>\n",
       "      <th>Oposição</th>\n",
       "      <th>PPC_igual_partidoAutor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.077973</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.024691</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.025341</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.148148</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.033138</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7936</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.080078</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7937</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.029297</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7938</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.132812</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7939</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.007812</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7940</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.132812</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7941 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Administração e Políticas Públicas  Desenvolvimento Social e Urbano  \\\n",
       "0                                      0                                1   \n",
       "1                                      1                                1   \n",
       "2                                      0                                1   \n",
       "3                                      0                                0   \n",
       "4                                      0                                0   \n",
       "...                                  ...                              ...   \n",
       "7936                                   1                                1   \n",
       "7937                                   0                                1   \n",
       "7938                                   0                                0   \n",
       "7939                                   1                                0   \n",
       "7940                                   0                                1   \n",
       "\n",
       "      Direitos e Justiça  Economia e Desenvolvimento Sustentável  \\\n",
       "0                      0                                       0   \n",
       "1                      0                                       0   \n",
       "2                      0                                       0   \n",
       "3                      0                                       1   \n",
       "4                      0                                       1   \n",
       "...                  ...                                     ...   \n",
       "7936                   0                                       1   \n",
       "7937                   0                                       0   \n",
       "7938                   1                                       0   \n",
       "7939                   0                                       0   \n",
       "7940                   1                                       0   \n",
       "\n",
       "      Infraestrutura e Gestão de Recursos  Influencia  Aliado  Oposição  \\\n",
       "0                                       0    0.077973       1         0   \n",
       "1                                       1    0.024691       1         0   \n",
       "2                                       0    0.025341       1         0   \n",
       "3                                       0    0.148148       1         0   \n",
       "4                                       0    0.033138       1         0   \n",
       "...                                   ...         ...     ...       ...   \n",
       "7936                                    0    0.080078       0         1   \n",
       "7937                                    0    0.029297       1         0   \n",
       "7938                                    0    0.132812       1         0   \n",
       "7939                                    0    0.007812       1         0   \n",
       "7940                                    0    0.132812       1         0   \n",
       "\n",
       "      PPC_igual_partidoAutor  \n",
       "0                          0  \n",
       "1                          0  \n",
       "2                          0  \n",
       "3                          0  \n",
       "4                          0  \n",
       "...                      ...  \n",
       "7936                       0  \n",
       "7937                       0  \n",
       "7938                       0  \n",
       "7939                       0  \n",
       "7940                       0  \n",
       "\n",
       "[7941 rows x 9 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Undersampling\n",
    "\n",
    "Os nossos dados estão desbalanceados.\n",
    "Por esse motivo fizemos undersampling para igualar a quantidade de exemplos da classe minoritária (1) e da classe majoritária (0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Administração e Políticas Públicas</th>\n",
       "      <th>Desenvolvimento Social e Urbano</th>\n",
       "      <th>Direitos e Justiça</th>\n",
       "      <th>Economia e Desenvolvimento Sustentável</th>\n",
       "      <th>Infraestrutura e Gestão de Recursos</th>\n",
       "      <th>Influencia</th>\n",
       "      <th>Aliado</th>\n",
       "      <th>Oposição</th>\n",
       "      <th>PPC_igual_partidoAutor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7159</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.107212</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5122</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.080235</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7771</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.115234</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5433</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.066277</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>468</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.048733</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7934</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.132812</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7935</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.115234</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7936</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.080078</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7937</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.029297</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7938</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.132812</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1514 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Administração e Políticas Públicas  Desenvolvimento Social e Urbano  \\\n",
       "7159                                   1                                0   \n",
       "5122                                   1                                0   \n",
       "7771                                   1                                0   \n",
       "5433                                   0                                0   \n",
       "468                                    0                                1   \n",
       "...                                  ...                              ...   \n",
       "7934                                   1                                1   \n",
       "7935                                   0                                1   \n",
       "7936                                   1                                1   \n",
       "7937                                   0                                1   \n",
       "7938                                   0                                0   \n",
       "\n",
       "      Direitos e Justiça  Economia e Desenvolvimento Sustentável  \\\n",
       "7159                   0                                       1   \n",
       "5122                   1                                       0   \n",
       "7771                   1                                       1   \n",
       "5433                   1                                       0   \n",
       "468                    0                                       0   \n",
       "...                  ...                                     ...   \n",
       "7934                   0                                       1   \n",
       "7935                   1                                       0   \n",
       "7936                   0                                       1   \n",
       "7937                   0                                       0   \n",
       "7938                   1                                       0   \n",
       "\n",
       "      Infraestrutura e Gestão de Recursos  Influencia  Aliado  Oposição  \\\n",
       "7159                                    0    0.107212       0         1   \n",
       "5122                                    0    0.080235       0         1   \n",
       "7771                                    0    0.115234       1         0   \n",
       "5433                                    0    0.066277       0         1   \n",
       "468                                     0    0.048733       1         0   \n",
       "...                                   ...         ...     ...       ...   \n",
       "7934                                    0    0.132812       1         0   \n",
       "7935                                    0    0.115234       1         0   \n",
       "7936                                    0    0.080078       0         1   \n",
       "7937                                    0    0.029297       1         0   \n",
       "7938                                    0    0.132812       1         0   \n",
       "\n",
       "      PPC_igual_partidoAutor  \n",
       "7159                       0  \n",
       "5122                       1  \n",
       "7771                       0  \n",
       "5433                       0  \n",
       "468                        1  \n",
       "...                      ...  \n",
       "7934                       1  \n",
       "7935                       0  \n",
       "7936                       0  \n",
       "7937                       0  \n",
       "7938                       0  \n",
       "\n",
       "[1514 rows x 9 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "undersample = RandomUnderSampler()\n",
    "X_resampled, y_resampled = undersample.fit_resample(X, y)\n",
    "X_resampled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Chamando o modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "RANDOM_STATE=2\n",
    "BATCH_SIZE=10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Neurônios por camada\n",
    "\n",
    " $n\\Large \\le \\lfloor \\frac{\\lvert DS \\rvert - 10}{10\\times (d+2)} \\rfloor$ "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número maximo de neurônios na camada oculta:  13.0\n"
     ]
    }
   ],
   "source": [
    "n = np.floor((1514 - 10) / (10*(9+2)))\n",
    "print(\"Número maximo de neurônios na camada oculta: \", n)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Separando em treino, teste e validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = model_selection.train_test_split(X_resampled, y_resampled, random_state=RANDOM_STATE, test_size=0.2)\n",
    "X_train, X_val, y_train, y_val = model_selection.train_test_split(X_train, y_train, random_state=RANDOM_STATE, test_size=0.2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/adriel/anaconda3/lib/python3.11/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(10, input_dim = 9, kernel_initializer='normal', activation='relu'))\n",
    "model.add(Dense(5, kernel_initializer='normal', activation='relu'))\n",
    "model.add(Dense(1, kernel_initializer='normal', activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgd = optimizers.SGD(learning_rate=0.01)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">100</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>)              │            <span style=\"color: #00af00; text-decoration-color: #00af00\">55</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │             <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │           \u001b[38;5;34m100\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m)              │            \u001b[38;5;34m55\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │             \u001b[38;5;34m6\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">161</span> (644.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m161\u001b[0m (644.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">161</span> (644.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m161\u001b[0m (644.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "None\n"
     ]
    }
   ],
   "source": [
    "model.compile(loss='binary_crossentropy', optimizer=\"adam\", metrics=['accuracy'])\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 4ms/step - accuracy: 0.5055 - loss: 0.6931 - val_accuracy: 0.6790 - val_loss: 0.6915\n",
      "Epoch 2/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6674 - loss: 0.6891 - val_accuracy: 0.7243 - val_loss: 0.6747\n",
      "Epoch 3/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6536 - loss: 0.6694 - val_accuracy: 0.7243 - val_loss: 0.6373\n",
      "Epoch 4/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6763 - loss: 0.6431 - val_accuracy: 0.7243 - val_loss: 0.6099\n",
      "Epoch 5/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6665 - loss: 0.6245 - val_accuracy: 0.7243 - val_loss: 0.5943\n",
      "Epoch 6/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6839 - loss: 0.6008 - val_accuracy: 0.7119 - val_loss: 0.5856\n",
      "Epoch 7/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7157 - loss: 0.5840 - val_accuracy: 0.7037 - val_loss: 0.5788\n",
      "Epoch 8/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7146 - loss: 0.5850 - val_accuracy: 0.7037 - val_loss: 0.5733\n",
      "Epoch 9/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7070 - loss: 0.5686 - val_accuracy: 0.7037 - val_loss: 0.5730\n",
      "Epoch 10/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7137 - loss: 0.5770 - val_accuracy: 0.7078 - val_loss: 0.5677\n",
      "Epoch 11/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6809 - loss: 0.5830 - val_accuracy: 0.7078 - val_loss: 0.5634\n",
      "Epoch 12/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6991 - loss: 0.5745 - val_accuracy: 0.7078 - val_loss: 0.5588\n",
      "Epoch 13/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7257 - loss: 0.5497 - val_accuracy: 0.7037 - val_loss: 0.5595\n",
      "Epoch 14/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7170 - loss: 0.5723 - val_accuracy: 0.7119 - val_loss: 0.5529\n",
      "Epoch 15/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.7055 - loss: 0.5723 - val_accuracy: 0.7078 - val_loss: 0.5534\n",
      "Epoch 16/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6962 - loss: 0.5512 - val_accuracy: 0.7078 - val_loss: 0.5523\n",
      "Epoch 17/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7033 - loss: 0.5434 - val_accuracy: 0.7078 - val_loss: 0.5496\n",
      "Epoch 18/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7148 - loss: 0.5397 - val_accuracy: 0.7078 - val_loss: 0.5530\n",
      "Epoch 19/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7133 - loss: 0.5448 - val_accuracy: 0.7078 - val_loss: 0.5470\n",
      "Epoch 20/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6982 - loss: 0.5509 - val_accuracy: 0.7078 - val_loss: 0.5430\n",
      "Epoch 21/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7030 - loss: 0.5498 - val_accuracy: 0.7160 - val_loss: 0.5417\n",
      "Epoch 22/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6716 - loss: 0.5647 - val_accuracy: 0.7078 - val_loss: 0.5410\n",
      "Epoch 23/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7056 - loss: 0.5459 - val_accuracy: 0.7078 - val_loss: 0.5416\n",
      "Epoch 24/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6932 - loss: 0.5472 - val_accuracy: 0.7078 - val_loss: 0.5394\n",
      "Epoch 25/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7011 - loss: 0.5562 - val_accuracy: 0.7078 - val_loss: 0.5387\n",
      "Epoch 26/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7102 - loss: 0.5388 - val_accuracy: 0.7202 - val_loss: 0.5364\n",
      "Epoch 27/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7026 - loss: 0.5516 - val_accuracy: 0.7160 - val_loss: 0.5359\n",
      "Epoch 28/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7091 - loss: 0.5333 - val_accuracy: 0.7078 - val_loss: 0.5372\n",
      "Epoch 29/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6726 - loss: 0.5586 - val_accuracy: 0.7160 - val_loss: 0.5351\n",
      "Epoch 30/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7202 - loss: 0.5257 - val_accuracy: 0.7202 - val_loss: 0.5331\n",
      "Epoch 31/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6983 - loss: 0.5396 - val_accuracy: 0.7078 - val_loss: 0.5362\n",
      "Epoch 32/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7124 - loss: 0.5243 - val_accuracy: 0.7202 - val_loss: 0.5312\n",
      "Epoch 33/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6924 - loss: 0.5464 - val_accuracy: 0.7160 - val_loss: 0.5321\n",
      "Epoch 34/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6883 - loss: 0.5540 - val_accuracy: 0.7160 - val_loss: 0.5323\n",
      "Epoch 35/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6877 - loss: 0.5454 - val_accuracy: 0.7160 - val_loss: 0.5369\n",
      "Epoch 36/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7364 - loss: 0.5217 - val_accuracy: 0.7202 - val_loss: 0.5299\n",
      "Epoch 37/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6809 - loss: 0.5508 - val_accuracy: 0.7078 - val_loss: 0.5335\n",
      "Epoch 38/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7341 - loss: 0.5065 - val_accuracy: 0.7202 - val_loss: 0.5294\n",
      "Epoch 39/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6959 - loss: 0.5521 - val_accuracy: 0.7160 - val_loss: 0.5300\n",
      "Epoch 40/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6886 - loss: 0.5353 - val_accuracy: 0.7202 - val_loss: 0.5290\n",
      "Epoch 41/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7000 - loss: 0.5280 - val_accuracy: 0.7078 - val_loss: 0.5338\n",
      "Epoch 42/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6931 - loss: 0.5267 - val_accuracy: 0.7202 - val_loss: 0.5279\n",
      "Epoch 43/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6862 - loss: 0.5470 - val_accuracy: 0.7160 - val_loss: 0.5295\n",
      "Epoch 44/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7210 - loss: 0.5239 - val_accuracy: 0.7160 - val_loss: 0.5288\n",
      "Epoch 45/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7224 - loss: 0.5244 - val_accuracy: 0.7160 - val_loss: 0.5293\n",
      "Epoch 46/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6896 - loss: 0.5431 - val_accuracy: 0.7119 - val_loss: 0.5271\n",
      "Epoch 47/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7079 - loss: 0.5435 - val_accuracy: 0.7160 - val_loss: 0.5292\n",
      "Epoch 48/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7035 - loss: 0.5339 - val_accuracy: 0.7202 - val_loss: 0.5288\n",
      "Epoch 49/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7253 - loss: 0.5175 - val_accuracy: 0.7202 - val_loss: 0.5276\n",
      "Epoch 50/50\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7187 - loss: 0.5334 - val_accuracy: 0.7202 - val_loss: 0.5262\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs=50, batch_size=BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m31/31\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n",
      "\u001b[1m8/8\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--> Acuracia (treino): 0.7159\n",
      "--> Acuracia (validacao): 0.7202\n",
      "--> acc_train - acc_val = -0.0043\n",
      "--> E_val - E_in = -0.0019\n"
     ]
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Metrica de erro')\n",
    "plt.ylabel('Erro')\n",
    "plt.xlabel('Epoca')\n",
    "plt.legend(['Treinamento', 'Validacao'])\n",
    "plt.show()\n",
    "\n",
    "pred = np.round(model.predict(X_train))\n",
    "acc_train = accuracy_score(y_train, pred)\n",
    "\n",
    "pred = np.round(model.predict(X_val))\n",
    "acc_val = accuracy_score(y_val, pred)\n",
    "\n",
    "print(f'\\n--> Acuracia (treino): {acc_train:.4f}')\n",
    "print(f'--> Acuracia (validacao): {acc_val:.4f}')\n",
    "print(f\"--> acc_train - acc_val = {acc_train - acc_val:.4f}\")\n",
    "print(f\"--> E_val - E_in = {(history.history['val_loss'][-1] - history.history['loss'][-1]):.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Matriz de Confusão"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_prob = model.predict(X_test)\n",
    "y_pred = (y_pred_prob > 0.5).astype(int)\n",
    "\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=['Class 0', 'Class 1'], yticklabels=['Class 0', 'Class 1'])\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('Actual')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precisão: 0.7649947130635394\n",
      "Recall: 0.7194719471947195\n",
      "F1 Score: 0.7125119085249974\n",
      "\n",
      "Relatório de Classificação:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.64      0.91      0.75       141\n",
      "           1       0.87      0.56      0.68       162\n",
      "\n",
      "    accuracy                           0.72       303\n",
      "   macro avg       0.76      0.73      0.71       303\n",
      "weighted avg       0.76      0.72      0.71       303\n",
      "\n"
     ]
    }
   ],
   "source": [
    "precision = precision_score(y_test, y_pred, average='weighted')\n",
    "recall = recall_score(y_test, y_pred, average='weighted')\n",
    "f1 = f1_score(y_test, y_pred, average='weighted')\n",
    "\n",
    "print('Precisão:', precision)\n",
    "print('Recall:', recall)\n",
    "print('F1 Score:', f1)\n",
    "\n",
    "print(\"\\nRelatório de Classificação:\\n\", classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Regularização"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/adriel/anaconda3/lib/python3.11/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_1\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential_1\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">100</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │            <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │           \u001b[38;5;34m100\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │            \u001b[38;5;34m11\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">111</span> (444.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m111\u001b[0m (444.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">111</span> (444.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m111\u001b[0m (444.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "None\n",
      "Epoch 1/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.5859 - loss: 0.6937 - val_accuracy: 0.7037 - val_loss: 0.6869\n",
      "Epoch 2/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6815 - loss: 0.6834 - val_accuracy: 0.6708 - val_loss: 0.6712\n",
      "Epoch 3/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6502 - loss: 0.6694 - val_accuracy: 0.6708 - val_loss: 0.6519\n",
      "Epoch 4/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6687 - loss: 0.6486 - val_accuracy: 0.6708 - val_loss: 0.6354\n",
      "Epoch 5/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6547 - loss: 0.6372 - val_accuracy: 0.6626 - val_loss: 0.6271\n",
      "Epoch 6/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6733 - loss: 0.6256 - val_accuracy: 0.6708 - val_loss: 0.6144\n",
      "Epoch 7/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7004 - loss: 0.6050 - val_accuracy: 0.6626 - val_loss: 0.6080\n",
      "Epoch 8/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7028 - loss: 0.5962 - val_accuracy: 0.6626 - val_loss: 0.6034\n",
      "Epoch 9/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6750 - loss: 0.6155 - val_accuracy: 0.6626 - val_loss: 0.6049\n",
      "Epoch 10/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6905 - loss: 0.5978 - val_accuracy: 0.6626 - val_loss: 0.6022\n",
      "Epoch 11/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6971 - loss: 0.5986 - val_accuracy: 0.7037 - val_loss: 0.5947\n",
      "Epoch 12/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7122 - loss: 0.5865 - val_accuracy: 0.7037 - val_loss: 0.5956\n",
      "Epoch 13/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7187 - loss: 0.5947 - val_accuracy: 0.7037 - val_loss: 0.5913\n",
      "Epoch 14/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7325 - loss: 0.5695 - val_accuracy: 0.7037 - val_loss: 0.5871\n",
      "Epoch 15/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7286 - loss: 0.5819 - val_accuracy: 0.7037 - val_loss: 0.5840\n",
      "Epoch 16/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6793 - loss: 0.5968 - val_accuracy: 0.7037 - val_loss: 0.5906\n",
      "Epoch 17/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7142 - loss: 0.5848 - val_accuracy: 0.7037 - val_loss: 0.5820\n",
      "Epoch 18/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7239 - loss: 0.5711 - val_accuracy: 0.7037 - val_loss: 0.5801\n",
      "Epoch 19/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6993 - loss: 0.5965 - val_accuracy: 0.7037 - val_loss: 0.5786\n",
      "Epoch 20/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7037 - loss: 0.5796 - val_accuracy: 0.7037 - val_loss: 0.5813\n",
      "Epoch 21/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7103 - loss: 0.5769 - val_accuracy: 0.7037 - val_loss: 0.5839\n",
      "Epoch 22/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7051 - loss: 0.5863 - val_accuracy: 0.7037 - val_loss: 0.5750\n",
      "Epoch 23/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7117 - loss: 0.5824 - val_accuracy: 0.7037 - val_loss: 0.5727\n",
      "Epoch 24/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7176 - loss: 0.5710 - val_accuracy: 0.7037 - val_loss: 0.5748\n",
      "Epoch 25/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7145 - loss: 0.5738 - val_accuracy: 0.7037 - val_loss: 0.5724\n",
      "Epoch 26/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6969 - loss: 0.5814 - val_accuracy: 0.7037 - val_loss: 0.5718\n",
      "Epoch 27/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7102 - loss: 0.5678 - val_accuracy: 0.7037 - val_loss: 0.5682\n",
      "Epoch 28/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7179 - loss: 0.5753 - val_accuracy: 0.7037 - val_loss: 0.5690\n",
      "Epoch 29/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7252 - loss: 0.5595 - val_accuracy: 0.7037 - val_loss: 0.5670\n",
      "Epoch 30/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7279 - loss: 0.5642 - val_accuracy: 0.7037 - val_loss: 0.5657\n",
      "Epoch 31/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7076 - loss: 0.5698 - val_accuracy: 0.7037 - val_loss: 0.5683\n",
      "Epoch 32/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7146 - loss: 0.5610 - val_accuracy: 0.7037 - val_loss: 0.5710\n",
      "Epoch 33/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7219 - loss: 0.5582 - val_accuracy: 0.7037 - val_loss: 0.5631\n",
      "Epoch 34/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7072 - loss: 0.5643 - val_accuracy: 0.7037 - val_loss: 0.5716\n",
      "Epoch 35/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7115 - loss: 0.5721 - val_accuracy: 0.7037 - val_loss: 0.5695\n",
      "Epoch 36/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7164 - loss: 0.5612 - val_accuracy: 0.7037 - val_loss: 0.5630\n",
      "Epoch 37/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7182 - loss: 0.5483 - val_accuracy: 0.7037 - val_loss: 0.5676\n",
      "Epoch 38/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7142 - loss: 0.5643 - val_accuracy: 0.7037 - val_loss: 0.5635\n",
      "Epoch 39/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7110 - loss: 0.5508 - val_accuracy: 0.7037 - val_loss: 0.5669\n",
      "Epoch 40/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7068 - loss: 0.5675 - val_accuracy: 0.7037 - val_loss: 0.5623\n",
      "Epoch 41/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7375 - loss: 0.5508 - val_accuracy: 0.7037 - val_loss: 0.5581\n",
      "Epoch 42/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7376 - loss: 0.5507 - val_accuracy: 0.7037 - val_loss: 0.5583\n",
      "Epoch 43/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7043 - loss: 0.5689 - val_accuracy: 0.7037 - val_loss: 0.5558\n",
      "Epoch 44/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7241 - loss: 0.5499 - val_accuracy: 0.7037 - val_loss: 0.5602\n",
      "Epoch 45/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7103 - loss: 0.5628 - val_accuracy: 0.7037 - val_loss: 0.5561\n",
      "Epoch 46/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6791 - loss: 0.5767 - val_accuracy: 0.7037 - val_loss: 0.5655\n",
      "Epoch 47/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7087 - loss: 0.5571 - val_accuracy: 0.7037 - val_loss: 0.5574\n",
      "Epoch 48/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7191 - loss: 0.5535 - val_accuracy: 0.7037 - val_loss: 0.5610\n",
      "Epoch 49/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7321 - loss: 0.5413 - val_accuracy: 0.7078 - val_loss: 0.5653\n",
      "Epoch 50/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7097 - loss: 0.5613 - val_accuracy: 0.7037 - val_loss: 0.5552\n",
      "Epoch 51/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6875 - loss: 0.5746 - val_accuracy: 0.7119 - val_loss: 0.5608\n",
      "Epoch 52/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6946 - loss: 0.5718 - val_accuracy: 0.7037 - val_loss: 0.5525\n",
      "Epoch 53/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7331 - loss: 0.5428 - val_accuracy: 0.7037 - val_loss: 0.5541\n",
      "Epoch 54/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6813 - loss: 0.5743 - val_accuracy: 0.7119 - val_loss: 0.5584\n",
      "Epoch 55/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7207 - loss: 0.5577 - val_accuracy: 0.7037 - val_loss: 0.5522\n",
      "Epoch 56/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7364 - loss: 0.5492 - val_accuracy: 0.7037 - val_loss: 0.5608\n",
      "Epoch 57/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6989 - loss: 0.5645 - val_accuracy: 0.7037 - val_loss: 0.5522\n",
      "Epoch 58/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7206 - loss: 0.5537 - val_accuracy: 0.7037 - val_loss: 0.5536\n",
      "Epoch 59/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7221 - loss: 0.5558 - val_accuracy: 0.7037 - val_loss: 0.5561\n",
      "Epoch 60/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7031 - loss: 0.5661 - val_accuracy: 0.7037 - val_loss: 0.5541\n",
      "Epoch 61/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7270 - loss: 0.5434 - val_accuracy: 0.7037 - val_loss: 0.5506\n",
      "Epoch 62/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6935 - loss: 0.5678 - val_accuracy: 0.7078 - val_loss: 0.5622\n",
      "Epoch 63/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7168 - loss: 0.5495 - val_accuracy: 0.7037 - val_loss: 0.5487\n",
      "Epoch 64/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7131 - loss: 0.5614 - val_accuracy: 0.7119 - val_loss: 0.5584\n",
      "Epoch 65/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7334 - loss: 0.5306 - val_accuracy: 0.7078 - val_loss: 0.5477\n",
      "Epoch 66/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7037 - loss: 0.5635 - val_accuracy: 0.7037 - val_loss: 0.5513\n",
      "Epoch 67/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7107 - loss: 0.5561 - val_accuracy: 0.7037 - val_loss: 0.5489\n",
      "Epoch 68/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6790 - loss: 0.5715 - val_accuracy: 0.7037 - val_loss: 0.5508\n",
      "Epoch 69/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7004 - loss: 0.5658 - val_accuracy: 0.7037 - val_loss: 0.5478\n",
      "Epoch 70/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7183 - loss: 0.5502 - val_accuracy: 0.7037 - val_loss: 0.5478\n",
      "Epoch 71/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6997 - loss: 0.5605 - val_accuracy: 0.7037 - val_loss: 0.5495\n",
      "Epoch 72/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7275 - loss: 0.5541 - val_accuracy: 0.7037 - val_loss: 0.5518\n",
      "Epoch 73/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7154 - loss: 0.5525 - val_accuracy: 0.7037 - val_loss: 0.5498\n",
      "Epoch 74/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7118 - loss: 0.5469 - val_accuracy: 0.7037 - val_loss: 0.5500\n",
      "Epoch 75/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6958 - loss: 0.5694 - val_accuracy: 0.7119 - val_loss: 0.5557\n",
      "Epoch 76/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7300 - loss: 0.5396 - val_accuracy: 0.7037 - val_loss: 0.5464\n",
      "Epoch 77/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7129 - loss: 0.5530 - val_accuracy: 0.7037 - val_loss: 0.5462\n",
      "Epoch 78/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7214 - loss: 0.5555 - val_accuracy: 0.7078 - val_loss: 0.5609\n",
      "Epoch 79/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.6874 - loss: 0.5628 - val_accuracy: 0.7037 - val_loss: 0.5464\n",
      "Epoch 80/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7186 - loss: 0.5403 - val_accuracy: 0.7037 - val_loss: 0.5470\n",
      "Epoch 81/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6819 - loss: 0.5747 - val_accuracy: 0.7078 - val_loss: 0.5585\n",
      "Epoch 82/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7214 - loss: 0.5529 - val_accuracy: 0.7037 - val_loss: 0.5450\n",
      "Epoch 83/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7026 - loss: 0.5611 - val_accuracy: 0.7078 - val_loss: 0.5569\n",
      "Epoch 84/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.6931 - loss: 0.5696 - val_accuracy: 0.7078 - val_loss: 0.5586\n",
      "Epoch 85/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6958 - loss: 0.5603 - val_accuracy: 0.7037 - val_loss: 0.5458\n",
      "Epoch 86/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7193 - loss: 0.5467 - val_accuracy: 0.7037 - val_loss: 0.5446\n",
      "Epoch 87/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7080 - loss: 0.5594 - val_accuracy: 0.7078 - val_loss: 0.5505\n",
      "Epoch 88/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7407 - loss: 0.5247 - val_accuracy: 0.7078 - val_loss: 0.5518\n",
      "Epoch 89/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7112 - loss: 0.5508 - val_accuracy: 0.7119 - val_loss: 0.5523\n",
      "Epoch 90/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7304 - loss: 0.5410 - val_accuracy: 0.7037 - val_loss: 0.5509\n",
      "Epoch 91/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.6967 - loss: 0.5766 - val_accuracy: 0.7037 - val_loss: 0.5444\n",
      "Epoch 92/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7349 - loss: 0.5328 - val_accuracy: 0.7037 - val_loss: 0.5443\n",
      "Epoch 93/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7296 - loss: 0.5529 - val_accuracy: 0.7037 - val_loss: 0.5500\n",
      "Epoch 94/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7317 - loss: 0.5362 - val_accuracy: 0.7037 - val_loss: 0.5435\n",
      "Epoch 95/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7159 - loss: 0.5499 - val_accuracy: 0.7037 - val_loss: 0.5485\n",
      "Epoch 96/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7023 - loss: 0.5619 - val_accuracy: 0.7037 - val_loss: 0.5457\n",
      "Epoch 97/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7113 - loss: 0.5564 - val_accuracy: 0.7037 - val_loss: 0.5501\n",
      "Epoch 98/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7282 - loss: 0.5259 - val_accuracy: 0.7037 - val_loss: 0.5431\n",
      "Epoch 99/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7015 - loss: 0.5567 - val_accuracy: 0.7078 - val_loss: 0.5537\n",
      "Epoch 100/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7343 - loss: 0.5495 - val_accuracy: 0.7119 - val_loss: 0.5479\n",
      "Epoch 101/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7354 - loss: 0.5338 - val_accuracy: 0.7037 - val_loss: 0.5430\n",
      "Epoch 102/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7308 - loss: 0.5450 - val_accuracy: 0.7160 - val_loss: 0.5437\n",
      "Epoch 103/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7111 - loss: 0.5644 - val_accuracy: 0.7119 - val_loss: 0.5511\n",
      "Epoch 104/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7130 - loss: 0.5477 - val_accuracy: 0.7078 - val_loss: 0.5566\n",
      "Epoch 105/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7097 - loss: 0.5499 - val_accuracy: 0.7037 - val_loss: 0.5447\n",
      "Epoch 106/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7026 - loss: 0.5558 - val_accuracy: 0.7037 - val_loss: 0.5446\n",
      "Epoch 107/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7004 - loss: 0.5694 - val_accuracy: 0.7119 - val_loss: 0.5478\n",
      "Epoch 108/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7114 - loss: 0.5531 - val_accuracy: 0.7037 - val_loss: 0.5437\n",
      "Epoch 109/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7268 - loss: 0.5509 - val_accuracy: 0.7037 - val_loss: 0.5420\n",
      "Epoch 110/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7137 - loss: 0.5362 - val_accuracy: 0.7037 - val_loss: 0.5451\n",
      "Epoch 111/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7046 - loss: 0.5559 - val_accuracy: 0.7037 - val_loss: 0.5432\n",
      "Epoch 112/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7069 - loss: 0.5704 - val_accuracy: 0.7119 - val_loss: 0.5463\n",
      "Epoch 113/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7070 - loss: 0.5533 - val_accuracy: 0.7037 - val_loss: 0.5414\n",
      "Epoch 114/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7232 - loss: 0.5389 - val_accuracy: 0.7037 - val_loss: 0.5463\n",
      "Epoch 115/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7214 - loss: 0.5453 - val_accuracy: 0.7037 - val_loss: 0.5478\n",
      "Epoch 116/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7081 - loss: 0.5464 - val_accuracy: 0.7037 - val_loss: 0.5486\n",
      "Epoch 117/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7061 - loss: 0.5503 - val_accuracy: 0.7037 - val_loss: 0.5407\n",
      "Epoch 118/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7091 - loss: 0.5644 - val_accuracy: 0.7119 - val_loss: 0.5457\n",
      "Epoch 119/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7158 - loss: 0.5508 - val_accuracy: 0.7078 - val_loss: 0.5465\n",
      "Epoch 120/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7104 - loss: 0.5406 - val_accuracy: 0.7037 - val_loss: 0.5456\n",
      "Epoch 121/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7029 - loss: 0.5537 - val_accuracy: 0.7078 - val_loss: 0.5464\n",
      "Epoch 122/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7128 - loss: 0.5454 - val_accuracy: 0.7160 - val_loss: 0.5567\n",
      "Epoch 123/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7308 - loss: 0.5300 - val_accuracy: 0.7037 - val_loss: 0.5400\n",
      "Epoch 124/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7233 - loss: 0.5479 - val_accuracy: 0.7160 - val_loss: 0.5413\n",
      "Epoch 125/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7107 - loss: 0.5596 - val_accuracy: 0.7037 - val_loss: 0.5389\n",
      "Epoch 126/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7222 - loss: 0.5306 - val_accuracy: 0.7037 - val_loss: 0.5429\n",
      "Epoch 127/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6896 - loss: 0.5609 - val_accuracy: 0.7037 - val_loss: 0.5409\n",
      "Epoch 128/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7043 - loss: 0.5539 - val_accuracy: 0.7037 - val_loss: 0.5426\n",
      "Epoch 129/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7216 - loss: 0.5404 - val_accuracy: 0.7037 - val_loss: 0.5400\n",
      "Epoch 130/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7045 - loss: 0.5474 - val_accuracy: 0.7160 - val_loss: 0.5495\n",
      "Epoch 131/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6953 - loss: 0.5650 - val_accuracy: 0.7037 - val_loss: 0.5381\n",
      "Epoch 132/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6933 - loss: 0.5570 - val_accuracy: 0.7078 - val_loss: 0.5432\n",
      "Epoch 133/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7220 - loss: 0.5368 - val_accuracy: 0.7037 - val_loss: 0.5462\n",
      "Epoch 134/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6868 - loss: 0.5566 - val_accuracy: 0.7078 - val_loss: 0.5396\n",
      "Epoch 135/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6813 - loss: 0.5675 - val_accuracy: 0.7078 - val_loss: 0.5402\n",
      "Epoch 136/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7101 - loss: 0.5499 - val_accuracy: 0.7037 - val_loss: 0.5355\n",
      "Epoch 137/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7246 - loss: 0.5493 - val_accuracy: 0.7160 - val_loss: 0.5351\n",
      "Epoch 138/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7112 - loss: 0.5475 - val_accuracy: 0.7037 - val_loss: 0.5349\n",
      "Epoch 139/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6970 - loss: 0.5644 - val_accuracy: 0.7119 - val_loss: 0.5387\n",
      "Epoch 140/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6844 - loss: 0.5629 - val_accuracy: 0.7037 - val_loss: 0.5364\n",
      "Epoch 141/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7161 - loss: 0.5448 - val_accuracy: 0.7078 - val_loss: 0.5385\n",
      "Epoch 142/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7037 - loss: 0.5557 - val_accuracy: 0.7078 - val_loss: 0.5410\n",
      "Epoch 143/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7109 - loss: 0.5445 - val_accuracy: 0.7078 - val_loss: 0.5361\n",
      "Epoch 144/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7206 - loss: 0.5330 - val_accuracy: 0.7037 - val_loss: 0.5349\n",
      "Epoch 145/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7091 - loss: 0.5497 - val_accuracy: 0.7078 - val_loss: 0.5346\n",
      "Epoch 146/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7170 - loss: 0.5346 - val_accuracy: 0.7119 - val_loss: 0.5346\n",
      "Epoch 147/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6847 - loss: 0.5703 - val_accuracy: 0.7078 - val_loss: 0.5424\n",
      "Epoch 148/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7013 - loss: 0.5537 - val_accuracy: 0.7078 - val_loss: 0.5360\n",
      "Epoch 149/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7229 - loss: 0.5417 - val_accuracy: 0.7078 - val_loss: 0.5352\n",
      "Epoch 150/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.7097 - loss: 0.5521 - val_accuracy: 0.7160 - val_loss: 0.5319\n",
      "Epoch 151/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7088 - loss: 0.5478 - val_accuracy: 0.7078 - val_loss: 0.5355\n",
      "Epoch 152/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7074 - loss: 0.5476 - val_accuracy: 0.7078 - val_loss: 0.5424\n",
      "Epoch 153/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7046 - loss: 0.5550 - val_accuracy: 0.7160 - val_loss: 0.5459\n",
      "Epoch 154/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6785 - loss: 0.5636 - val_accuracy: 0.7078 - val_loss: 0.5354\n",
      "Epoch 155/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7179 - loss: 0.5342 - val_accuracy: 0.7078 - val_loss: 0.5321\n",
      "Epoch 156/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6938 - loss: 0.5631 - val_accuracy: 0.7037 - val_loss: 0.5332\n",
      "Epoch 157/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7076 - loss: 0.5629 - val_accuracy: 0.7078 - val_loss: 0.5387\n",
      "Epoch 158/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7247 - loss: 0.5274 - val_accuracy: 0.7078 - val_loss: 0.5331\n",
      "Epoch 159/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6935 - loss: 0.5516 - val_accuracy: 0.7078 - val_loss: 0.5371\n",
      "Epoch 160/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6795 - loss: 0.5611 - val_accuracy: 0.7202 - val_loss: 0.5321\n",
      "Epoch 161/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6962 - loss: 0.5461 - val_accuracy: 0.7078 - val_loss: 0.5358\n",
      "Epoch 162/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7145 - loss: 0.5321 - val_accuracy: 0.7037 - val_loss: 0.5340\n",
      "Epoch 163/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7127 - loss: 0.5463 - val_accuracy: 0.7160 - val_loss: 0.5308\n",
      "Epoch 164/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7123 - loss: 0.5510 - val_accuracy: 0.7078 - val_loss: 0.5359\n",
      "Epoch 165/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.7156 - loss: 0.5337 - val_accuracy: 0.7078 - val_loss: 0.5333\n",
      "Epoch 166/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6960 - loss: 0.5532 - val_accuracy: 0.7078 - val_loss: 0.5343\n",
      "Epoch 167/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7202 - loss: 0.5253 - val_accuracy: 0.7078 - val_loss: 0.5345\n",
      "Epoch 168/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7084 - loss: 0.5464 - val_accuracy: 0.7160 - val_loss: 0.5388\n",
      "Epoch 169/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7117 - loss: 0.5420 - val_accuracy: 0.7160 - val_loss: 0.5404\n",
      "Epoch 170/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6947 - loss: 0.5555 - val_accuracy: 0.7160 - val_loss: 0.5399\n",
      "Epoch 171/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7206 - loss: 0.5376 - val_accuracy: 0.7078 - val_loss: 0.5384\n",
      "Epoch 172/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 10ms/step - accuracy: 0.7016 - loss: 0.5364 - val_accuracy: 0.7078 - val_loss: 0.5315\n",
      "Epoch 173/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7228 - loss: 0.5339 - val_accuracy: 0.7160 - val_loss: 0.5300\n",
      "Epoch 174/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7193 - loss: 0.5475 - val_accuracy: 0.7160 - val_loss: 0.5298\n",
      "Epoch 175/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6902 - loss: 0.5585 - val_accuracy: 0.7078 - val_loss: 0.5375\n",
      "Epoch 176/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6940 - loss: 0.5631 - val_accuracy: 0.7160 - val_loss: 0.5419\n",
      "Epoch 177/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6904 - loss: 0.5599 - val_accuracy: 0.7160 - val_loss: 0.5300\n",
      "Epoch 178/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7208 - loss: 0.5390 - val_accuracy: 0.7160 - val_loss: 0.5349\n",
      "Epoch 179/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7090 - loss: 0.5512 - val_accuracy: 0.7160 - val_loss: 0.5364\n",
      "Epoch 180/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6983 - loss: 0.5468 - val_accuracy: 0.7078 - val_loss: 0.5327\n",
      "Epoch 181/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7197 - loss: 0.5468 - val_accuracy: 0.7078 - val_loss: 0.5322\n",
      "Epoch 182/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7092 - loss: 0.5447 - val_accuracy: 0.7160 - val_loss: 0.5299\n",
      "Epoch 183/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7093 - loss: 0.5474 - val_accuracy: 0.7160 - val_loss: 0.5297\n",
      "Epoch 184/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7074 - loss: 0.5473 - val_accuracy: 0.7078 - val_loss: 0.5340\n",
      "Epoch 185/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7247 - loss: 0.5379 - val_accuracy: 0.7078 - val_loss: 0.5338\n",
      "Epoch 186/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7032 - loss: 0.5479 - val_accuracy: 0.7202 - val_loss: 0.5300\n",
      "Epoch 187/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7013 - loss: 0.5435 - val_accuracy: 0.7160 - val_loss: 0.5409\n",
      "Epoch 188/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7173 - loss: 0.5331 - val_accuracy: 0.7202 - val_loss: 0.5284\n",
      "Epoch 189/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6849 - loss: 0.5444 - val_accuracy: 0.7160 - val_loss: 0.5430\n",
      "Epoch 190/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7286 - loss: 0.5317 - val_accuracy: 0.7078 - val_loss: 0.5315\n",
      "Epoch 191/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6982 - loss: 0.5451 - val_accuracy: 0.7078 - val_loss: 0.5320\n",
      "Epoch 192/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7031 - loss: 0.5439 - val_accuracy: 0.7160 - val_loss: 0.5388\n",
      "Epoch 193/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7009 - loss: 0.5485 - val_accuracy: 0.7078 - val_loss: 0.5330\n",
      "Epoch 194/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6846 - loss: 0.5539 - val_accuracy: 0.7202 - val_loss: 0.5299\n",
      "Epoch 195/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7300 - loss: 0.5502 - val_accuracy: 0.7078 - val_loss: 0.5301\n",
      "Epoch 196/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6924 - loss: 0.5589 - val_accuracy: 0.7160 - val_loss: 0.5282\n",
      "Epoch 197/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7248 - loss: 0.5301 - val_accuracy: 0.7202 - val_loss: 0.5289\n",
      "Epoch 198/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7179 - loss: 0.5375 - val_accuracy: 0.7078 - val_loss: 0.5300\n",
      "Epoch 199/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.7058 - loss: 0.5518 - val_accuracy: 0.7202 - val_loss: 0.5294\n",
      "Epoch 200/200\n",
      "\u001b[1m97/97\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7327 - loss: 0.5239 - val_accuracy: 0.7160 - val_loss: 0.5281\n"
     ]
    }
   ],
   "source": [
    "from keras import regularizers\n",
    "from keras.callbacks import EarlyStopping\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "# Após o teste anterior, testem a rede neural abaixo, treinando-a com o Early Stopping\n",
    "model.add(Dense(10, input_dim=9, kernel_initializer='normal', kernel_regularizer=regularizers.l2(0.01), activation='relu'))\n",
    "model.add(Dense(1, kernel_initializer='normal', activation='sigmoid'))\n",
    "\"\"\"model.add(Dropout(0.2))\n",
    "model.add(Dense(200, kernel_initializer='normal', kernel_regularizer=regularizers.l2(0.01), activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(1, kernel_initializer='normal', activation='sigmoid'))\n",
    "\"\"\"\n",
    "# Early Stopping\n",
    "#es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=20)\n",
    "\n",
    "# Compilar o modelo\n",
    "model.compile(loss='binary_crossentropy', optimizer=\"adam\", metrics=['accuracy'])\n",
    "\n",
    "print(model.summary())\n",
    "\n",
    "# Treinamento SEM Early Stopping\n",
    "history = model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs=200, batch_size=BATCH_SIZE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m31/31\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n",
      "\u001b[1m8/8\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "\n",
      "--> Acuracia (treino): 0.7190\n",
      "--> Acuracia (validacao): 0.7160\n",
      "--> acc_train - acc_val = 0.0030\n",
      "--> E_val - E_in = -0.0154\n"
     ]
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Metrica de erro')\n",
    "plt.ylabel('Erro')\n",
    "plt.xlabel('Epoca')\n",
    "plt.legend(['Treinamento', 'Validacao'])\n",
    "plt.show()\n",
    "\n",
    "pred = np.round(model.predict(X_train))\n",
    "acc_train = accuracy_score(y_train, pred)\n",
    "\n",
    "pred = np.round(model.predict(X_val))\n",
    "acc_val = accuracy_score(y_val, pred)\n",
    "\n",
    "print(f'\\n--> Acuracia (treino): {acc_train:.4f}')\n",
    "print(f'--> Acuracia (validacao): {acc_val:.4f}')\n",
    "print(f\"--> acc_train - acc_val = {acc_train - acc_val:.4f}\")\n",
    "print(f\"--> E_val - E_in = {(history.history['val_loss'][-1] - history.history['loss'][-1]):.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Utilizando K-Fold Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/adriel/anaconda3/lib/python3.11/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5083\n",
      "--> Acuracia (validacao): 0.5103\n",
      "--> E_val - E_in = -0.0002\n",
      "--> acc_in - acc_val = -0.0020\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5088\n",
      "--> Acuracia (validacao): 0.5083\n",
      "--> E_val - E_in = -0.0000\n",
      "--> acc_in - acc_val = 0.0005\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5088\n",
      "--> Acuracia (validacao): 0.5083\n",
      "--> E_val - E_in = -0.0000\n",
      "--> acc_in - acc_val = 0.0005\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5088\n",
      "--> Acuracia (validacao): 0.5083\n",
      "--> E_val - E_in = -0.0001\n",
      "--> acc_in - acc_val = 0.0005\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlIAAAHFCAYAAAA5VBcVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB/i0lEQVR4nO3dd1xT1/sH8M8NIQkbRWUqIqC4Bzhwj7pr0Q61WrXa9ld3ra2r2lqtrdZWW20dtXVVrdp+nW0Vi9a9F4riRgWZggqIrCT390fIhcgQEBJJP+/XKy/IvTc3J5dAHs55znMEURRFEBEREVGJyUzdACIiIqKKioEUERERUSkxkCIiIiIqJQZSRERERKXEQIqIiIiolBhIEREREZUSAykiIiKiUmIgRURERFRKDKSIiIiISomBFFEFtmbNGgiCAEEQcODAgXz7RVGEj48PBEFAx44dS/UcS5cuxZo1a0r0mAMHDhTaJlMQBAGff/65qZvxwunYsaP0/nn6dunSpRKfqzjvsTt37kAQhGe+p/THFXTbtGlTidpGVJ7kpm4AET0/Ozs7rFy5Mt8H2cGDB3Hr1i3Y2dmV+txLly5FlSpV8Pbbbxf7Mc2aNcPx48dRr169Uj8vGUetWrWwYcOGfNu9vb1N0Jr8xo0bh0GDBhls8/X1NVFriPJjIEVkBgYMGIANGzZgyZIlsLe3l7avXLkSgYGBSElJMUo7srOzIQgC7O3t0apVK6M8Z0Xz5MkTWFtb59uu0WigVquhVCqN2h4rK6sX+mdVo0aNF7p9RBzaIzIDb775JgBg48aN0rbk5GRs2bIFI0aMKPAxWVlZmDNnDvz8/KBUKlG1alUMHz4c9+/fl46pWbMmLl++jIMHD0rDKjVr1gSQO3y3bt06fPTRR3B3d4dSqcTNmzcLHdo7efIk+vTpAycnJ6hUKnh7e2PChAnS/ps3b2L48OHw9fWFtbU13N3d0adPH4SFhRXrOqSkpOC9996Dk5MTbG1t0aNHD1y/fr3AY2/cuIFBgwahWrVqUCqVqFu3LpYsWVKs5xFFEUuXLkWTJk1gZWWFSpUq4fXXX0dERITBcR07dkSDBg1w6NAhtG7dGtbW1hgxYoQ0bDV//nzMmTMHXl5eUCqV2L9/PwBg586dCAwMhLW1Nezs7NC1a1ccP368WG0raxkZGZg2bRq8vLygUCjg7u6OMWPG4NGjR898bExMDPr37w87Ozs4ODhgwIABiIuLK/9GExkRAykiM2Bvb4/XX38dq1atkrZt3LgRMpkMAwYMyHe8VqtFUFAQ5s2bh0GDBuHvv//GvHnzEBISgo4dOyI9PR0AsG3bNtSqVQtNmzbF8ePHcfz4cWzbts3gXNOmTUNkZCSWL1+OP//8E9WqVSuwjXv27EG7du0QGRmJhQsXYvfu3ZgxYwbi4+OlY2JiYuDk5IR58+YhODgYS5YsgVwuR8uWLXHt2rUir4Eoiujbt68U2G3btg2tWrVCz5498x0bHh6O5s2b49KlS1iwYAH++usv9O7dG+PHj8esWbOKfB4AeP/99zFhwgS89NJL2L59O5YuXYrLly+jdevWBq8HAGJjY/HWW29h0KBB2LVrF0aPHi3tW7x4Mf799198++232L17N/z8/PDbb78hKCgI9vb22LhxI1auXImHDx+iY8eOOHLkyDPbVhpqtdrgptVqAeRe02+//RZDhgzB33//jYkTJ2Lt2rXo3LkzMjMzCz1neno6XnrpJfzzzz+YO3cu/vjjD7i4uBT4fizKvHnzoFAoYG1tjbZt22Lnzp3P9VqJypxIRBXW6tWrRQDi6dOnxf3794sAxEuXLomiKIrNmzcX3377bVEURbF+/fpihw4dpMdt3LhRBCBu2bLF4HynT58WAYhLly6Vtj39WD3987Vv377Qffv375e2eXt7i97e3mJ6enqxX59arRazsrJEX19f8cMPPyzy2N27d4sAxEWLFhls//LLL0UA4syZM6Vt3bt3Fz08PMTk5GSDY8eOHSuqVCrxwYMHhT7P8ePHRQDiggULDLZHRUWJVlZW4uTJk6VtHTp0EAGI+/btMzj29u3bIgDR29tbzMrKkrZrNBrRzc1NbNiwoajRaKTtqampYrVq1cTWrVsXeQ1KSt++p2+DBw8WRVEUg4ODRQDi/PnzDR63efNmEYC4YsUKg3PlfZ8sW7ZMBCDu2LHD4LHvvfeeCEBcvXp1kW2LiYkR33vvPfH3338XDx8+LG7YsEFs1aqVCED8+eefn++FE5Uh9kgRmYkOHTrA29sbq1atQlhYGE6fPl3osN5ff/0FR0dH9OnTx6AnokmTJnBxcSnRbLvXXnvtmcdcv34dt27dwjvvvAOVSlXocWq1Gl999RXq1asHhUIBuVwOhUKBGzdu4MqVK0U+h35YbPDgwQbbn05UzsjIwL59+9CvXz9YW1sbvP5evXohIyMDJ06cKPR5/vrrLwiCgLfeesvgsS4uLmjcuHG+a1epUiV07ty5wHO98sorsLS0lO5fu3YNMTExGDJkCGSy3D/Ptra2eO2113DixAk8efKk0Lbp86ye7lkqire3N06fPm1w++KLLwAA//77LwDkm2jwxhtvwMbGBvv27Sv0vPv374ednR1eeeUVg+1P/zwK4+rqihUrVuCNN95A27ZtMWjQIBw6dAhNmzbF1KlToVari3UeovLGZHMiMyEIAoYPH47FixcjIyMDtWvXRrt27Qo8Nj4+Ho8ePYJCoShwf2JiYrGf19XV9ZnH6POuPDw8ijxu4sSJWLJkCaZMmYIOHTqgUqVKkMlkePfdd6XhxsIkJSVBLpfDycnJYLuLi0u+49RqNX744Qf88MMPBZ6rqNcfHx8PURTh7Oxc4P5atWoZ3C/q+jy9LykpqdDHuLm5QavV4uHDhwUmqwO6oOju3bvS/ZkzZz6z7INKpUJAQECB+/TXtGrVqgbbBUGAi4uL1N7CHlvQNXr651ESlpaWGDBgAKZOnYobN26gbt26pT4XUVlhIEVkRt5++2189tlnWL58Ob788stCj6tSpQqcnJwQHBxc4P6SlEsQBOGZx+g/iO/du1fkcevXr8fQoUPx1VdfGWxPTEyEo6NjkY91cnKCWq1GUlKSQTD1dHJzpUqVYGFhgSFDhmDMmDEFnsvLy6vQ56lSpQoEQcDhw4cLnGH39Lairs/T+/Ttjo2NzXdsTEwMZDIZKlWqVOj5/vzzT4O8JTc3t0KPLQ79Nb1//75BMCWKIuLi4tC8efMiH3vq1Kl825832VwURQAw6LEjMiW+E4nMiLu7OyZNmoQ+ffpg2LBhhR738ssvIykpCRqNBgEBAfluderUkY5VKpXP7A16ltq1a0vDjkUlKAuCkC8Q+fvvvxEdHf3M5+jUqRMA5KuJ9Ntvvxnct7a2RqdOnXD+/Hk0atSowNf/dK9WXi+//DJEUUR0dHSBj23YsOEz21qYOnXqwN3dHb/99psUMABAWloatmzZIs3kK0zDhg0N2vK8gVSXLl0A6ALcvLZs2YK0tDRpf0E6deqE1NTUfMnhT/88SiI7OxubN29GlSpV4OPjU+rzEJUl9kgRmZl58+Y985iBAwdiw4YN6NWrFz744AO0aNEClpaWuHfvHvbv34+goCD069cPgO7DedOmTdi8eTNq1aoFlUpVqmBhyZIl6NOnD1q1aoUPP/wQNWrUQGRkJPbs2SMFPy+//DLWrFkDPz8/NGrUCGfPnsU333zzzCFBAOjWrRvat2+PyZMnIy0tDQEBATh69CjWrVuX79hFixahbdu2aNeuHUaNGoWaNWsiNTUVN2/exJ9//inlBhWkTZs2+L//+z8MHz4cZ86cQfv27WFjY4PY2FgcOXIEDRs2xKhRo0p8fQBdL8v8+fMxePBgvPzyy3j//feRmZmJb775Bo8ePSrWz7Ysde3aFd27d8eUKVOQkpKCNm3a4OLFi5g5cyaaNm2KIUOGFPrYoUOH4rvvvsPQoUPx5ZdfwtfXF7t27cKePXuK9dwTJ05EdnY22rRpAxcXF0RFReGHH35AaGgoVq9eDQsLi7J6mUTPx6Sp7kT0XPLO2itKQTPvsrOzxW+//VZs3LixqFKpRFtbW9HPz098//33xRs3bkjH3blzR+zWrZtoZ2cnAhA9PT1FUcydmffHH3/ke76CZu2Jom7GW8+ePUUHBwdRqVSK3t7eBrPxHj58KL7zzjtitWrVRGtra7Ft27bi4cOH880IK8yjR4/EESNGiI6OjqK1tbXYtWtX8erVq/lm7YmibubciBEjRHd3d9HS0lKsWrWq2Lp1a3HOnDnPfB5RFMVVq1aJLVu2FG1sbEQrKyvR29tbHDp0qHjmzBnpmA4dOoj169fP91j9rL1vvvmmwHNv375dbNmypahSqUQbGxuxS5cu4tGjR4vVrpIorH15paeni1OmTBE9PT1FS0tL0dXVVRw1apT48OHDfOd6+md079498bXXXhNtbW1FOzs78bXXXhOPHTtWrFl7K1euFFu0aCFWrlxZlMvlYqVKlcTu3buLe/bsKcUrJSo/gijm6T8mIiIiomJjjhQRERFRKTGQIiIiIiolBlJEREREpcRAioiIiKiUGEgRERERlRIDKSIiIqJSYkHOcqTVahETEwM7O7tiLaNBREREpieKIlJTU+Hm5vbM5YgYSJWjmJgYVK9e3dTNICIiolKIiop65soKDKTKkX7h16ioKNjb25u4NURERFQcKSkpqF69erEWcGcgVY70w3n29vYMpIiIiCqY4qTlMNmciIiIqJQYSBERERGVEgMpIiIiolJijhQREZk1jUaD7OxsUzeDXiCWlpawsLAok3MxkCIiIrMkiiLi4uLw6NEjUzeFXkCOjo5wcXF57jqPDKSIiMgs6YOoatWqwdramoWRCYAuwH7y5AkSEhIAAK6urs91PgZSRERkdjQajRREOTk5mbo59IKxsrICACQkJKBatWrPNczHZHMiIjI7+pwoa2trE7eEXlT698bz5s8xkCIiIrPF4TwqTFm9NxhIEREREZUSAykiIiIzdufOHQiCgNDQUFM3xSwxkCIiInpBCIJQ5O3tt98u8TmrV6+O2NhYNGjQoOwbbAIvWmDIWXsV0ONMNR49yYLK0gJVbJWmbg4REZWR2NhY6fvNmzfjs88+w7Vr16Rt+tlmetnZ2bC0tCzynBYWFnBxcSnbhpKEPVIV0Oojt9H26/1Y8M+1Zx9MREQVhouLi3RzcHCAIAjS/YyMDDg6OuL3339Hx44doVKpsH79egDA6tWrUbduXahUKvj5+WHp0qXSOZ/uwTlw4AAEQcC+ffsQEBAAa2trtG7d2iBgu3XrFoKCguDs7AxbW1s0b94ce/fuNWhrzZo1MWfOHAwdOhS2trbw9PTEjh07cP/+fQQFBcHW1hYNGzbEmTNnDB537NgxtG/fHlZWVqhevTrGjx+PtLQ0g/N+9dVXGDFiBOzs7FCjRg2sWLFC2u/l5QUAaNq0KQRBQMeOHQEAWq0Ws2fPhoeHB5RKJZo0aYLg4ODn/6E8AwOpCkhpqfuxZWZrTdwSIqKKQxRFPMlSm+QmimKZvY4pU6Zg/PjxuHLlCrp3746ff/4Z06dPx5dffokrV67gq6++wqeffoq1a9cWeZ7p06djwYIFOHPmDORyOUaMGCHte/z4MXr16oW9e/fi/Pnz6N69O/r06YPIyEiDc3z33Xdo06YNzp8/j969e2PIkCEYOnQo3nrrLZw7dw4+Pj4YOnSo9PrDwsLQvXt3vPrqq7h48SI2b96MI0eOYOzYsQbnXbBgAQICAnD+/HmMHj0ao0aNwtWrVwEAp06dAgDs3bsXsbGx2Lp1KwBg0aJFWLBgAb799ltcvHgR3bt3xyuvvIIbN2483wV/Bg7tVUAqS13hsEw1AykiouJKz9ag3md7TPLc4bO7w1pRNh+5EyZMwKuvvird/+KLL7BgwQJpm5eXF8LDw/HTTz9h2LBhhZ7nyy+/RIcOHQAAU6dORe/evZGRkQGVSoXGjRujcePG0rFz5szBtm3bsHPnToOgp1evXnj//fcBAJ999hmWLVuG5s2b44033gCgC/oCAwMRHx8PFxcXfPPNNxg0aBAmTJgAAPD19cXixYvRoUMHLFu2DCqVSjrv6NGjpXN89913OHDgAPz8/FC1alUAgJOTk8GQ5bfffospU6Zg4MCBAICvv/4a+/fvx/fff48lS5aU4koXDwOpCkgpz+mRUmtM3BIiIjK2gIAA6fv79+8jKioK77zzDt577z1pu1qthoODQ5HnadSokfS9fpmUhIQE1KhRA2lpaZg1axb++usvxMTEQK1WIz09PV+PVN5zODs7AwAaNmyYb1tCQgJcXFxw9uxZ3Lx5Exs2bJCOEUURWq0Wt2/fRt26dfOdVz+8qV/SpSApKSmIiYlBmzZtDLa3adMGFy5cKPI6PC8GUhWQUq7rkcrg0B4RUbFZWVogfHZ3kz13WbGxsZG+12p1nwM///wzWrZsaXDcs5Y9yZukri9OqT/fpEmTsGfPHnz77bfw8fGBlZUVXn/9dWRlZT3zHEWdV6vV4v3338f48ePztadGjRoFnld/Hv05ivJ0kU1RFMu9KCsDqQqIPVJERCUnCEKZDa+9KJydneHu7o6IiAgMHjy4zM57+PBhvP322+jXrx8AXc7UnTt3nvu8zZo1w+XLl+Hj41PqcygUCgC69RT17O3t4ebmhiNHjqB9+/bS9mPHjqFFixalb3AxmNc76j9CSjZnjhQR0X/e559/jvHjx8Pe3h49e/ZEZmYmzpw5g4cPH2LixImlOqePjw+2bt2KPn36QBAEfPrpp8XqEXqWKVOmoFWrVhgzZgzee+892NjY4MqVKwgJCcEPP/xQrHNUq1YNVlZWCA4OhoeHB1QqFRwcHDBp0iTMnDkT3t7eaNKkCVavXo3Q0FCDYcTywFl7FZB+aI+z9oiI6N1338Uvv/yCNWvWoGHDhujQoQPWrFkjlQkoje+++w6VKlVC69at0adPH3Tv3h3NmjV77rY2atQIBw8exI0bN9CuXTs0bdoUn376qZSjVRxyuRyLFy/GTz/9BDc3NwQFBQEAxo8fj48++ggfffQRGjZsiODgYOzcuRO+vr7P3e6iCGJZzskkAykpKXBwcEBycjLs7e3L7Lxn7z7Aa8uOo6aTNQ5M6lRm5yUiMhcZGRm4ffs2vLy8pJlgRHkV9R4pyec3e6QqIKlHikN7REREJsVAqgLSJ5tnZDPZnIiIyJQYSFVA7JEiIiJ6MTCQqoA4a4+IiOjFwECqAtIP7Wm0ItQaBlNERIXhfCoqTFm9NxhIVUCqPBVy2StFRJSfvjL2kydPTNwSelHp3xtPV1EvKRbkrIAUFrnxb0a2BjZK/hiJiPKysLCAo6OjtD6btbV1uS8VQhWDKIp48uQJEhIS4Ojo+MyldJ7F5J/AS5cuxTfffIPY2FjUr18f33//Pdq1a1fo8ZmZmZg9ezbWr1+PuLg4eHh4YPr06RgxYgQAIDs7G3PnzsXatWsRHR2NOnXq4Ouvv0aPHj0KPN/cuXPxySef4IMPPsD3338vbRdFEbNmzcKKFSvw8OFDtGzZEkuWLEH9+vXL9PWXhkwmQGEhQ5ZGyx4pIqJCuLi4AECRi93Sf5ejo6P0HnkeJg2kNm/ejAkTJmDp0qVo06YNfvrpJ/Ts2RPh4eEGixfm1b9/f8THx2PlypXw8fFBQkIC1Gq1tH/GjBlYv349fv75Z/j5+WHPnj3o168fjh07hqZNmxqc6/Tp01ixYoXBKtN68+fPx8KFC7FmzRrUrl0bc+bMQdeuXXHt2jXY2dmV7YUoBaWcgRQRUVEEQYCrqyuqVauG7OxsUzeHXiCWlpbP3RMlEU2oRYsW4siRIw22+fn5iVOnTi3w+N27d4sODg5iUlJSoed0dXUVf/zxR4NtQUFB4uDBgw22paamir6+vmJISIjYoUMH8YMPPpD2abVa0cXFRZw3b560LSMjQ3RwcBCXL19e3JcnJicniwDE5OTkYj+muPy/+Ef0nPKXeCW27M9NRET0X1aSz2+TJZtnZWXh7Nmz6Natm8H2bt264dixYwU+ZufOnQgICMD8+fPh7u6O2rVr4+OPP0Z6erp0TGZmZr5S71ZWVjhy5IjBtjFjxqB379546aWX8j3P7du3ERcXZ9A2pVKJDh06FNo2/XOnpKQY3MoL19sjIiIyPZMN7SUmJkKj0cDZ2dlgu7OzM+Li4gp8TEREBI4cOQKVSoVt27YhMTERo0ePxoMHD7Bq1SoAQPfu3bFw4UK0b98e3t7e2LdvH3bs2AGNJrcK+KZNm3Du3DmcPn26wOfRP39Bbbt7926hr2nu3LmYNWvWs198GWAtKSIiItMzefmDp2dRiKJY6MwKrVYLQRCwYcMGtGjRAr169ZLymPS9UosWLYKvry/8/PygUCgwduxYDB8+XBoLjYqKwgcffID169c/cyHLkrQNAKZNm4bk5GTpFhUV9czXX1r6HikuE0NERGQ6JgukqlSpAgsLi3y9TwkJCfl6gvRcXV3h7u4OBwcHaVvdunUhiiLu3bsHAKhatSq2b9+OtLQ03L17F1evXoWtrS28vLwAAGfPnkVCQgL8/f0hl8shl8tx8OBBLF68GHK5HBqNRsriL0nbAN3wn729vcGtvOiLcrJHioiIyHRMFkgpFAr4+/sjJCTEYHtISAhat25d4GPatGmDmJgYPH78WNp2/fp1yGQyeHh4GByrUqng7u4OtVqNLVu2ICgoCADQpUsXhIWFITQ0VLoFBARg8ODBCA0NhYWFBby8vODi4mLQtqysLBw8eLDQthlbbiDFHikiIiJTMWn5g4kTJ2LIkCEICAhAYGAgVqxYgcjISIwcORKAbqgsOjoav/76KwBg0KBB+OKLLzB8+HDMmjULiYmJmDRpEkaMGAErKysAwMmTJxEdHY0mTZogOjoan3/+ObRaLSZPngwAsLOzQ4MGDQzaYWNjAycnJ2m7IAiYMGECvvrqK/j6+sLX1xdfffUVrK2tMWjQIGNdniIpLZlsTkREZGomDaQGDBiApKQkzJ49G7GxsWjQoAF27doFT09PAEBsbCwiIyOl421tbRESEoJx48YhICAATk5O6N+/P+bMmSMdk5GRgRkzZiAiIgK2trbo1asX1q1bB0dHxxK1bfLkyUhPT8fo0aOlgpz//PPPC1FDCgBUHNojIiIyOUEUuaJjeUlJSYGDgwOSk5PLPF9q3Mbz+PNCDGb2qYfhbbzK9NxERET/ZSX5/Db5rD0qHX2OVAaH9oiIiEyGgVQFxWRzIiIi02MgVUFJlc2ZI0VERGQyDKQqKKmyOYf2iIiITIaBVAWlknqkOLRHRERkKgykKih9jxSTzYmIiEyHgVQFxWRzIiIi02MgVUEx2ZyIiMj0GEhVUFy0mIiIyPQYSFVQKmmtPQ7tERERmQoDqQqKPVJERESmx0CqgsqdtcceKSIiIlNhIFVB6ZPNs9gjRUREZDIMpCooDu0RERGZHgOpCkpaIoZ1pIiIiEyGgVQFJS0Rw8rmREREJsNAqoLK7ZFiIEVERGQqDKQqKCnZXKOFRiuauDVERET/TQykKih9sjnAmXtERESmwkCqgsobSDHhnIiIyDQYSFVQcgsZLGQCAOZJERERmQoDqQpMpa8lxZl7REREJsFAqgJT6hcu5tAeERGRSTCQqsD0eVIZ7JEiIiIyCQZSFVjuMjHskSIiIjIFBlIVmL6WFJPNiYiITIOBVAXG9faIiIhMi4FUBcb19oiIiEyLgVQFpu+RymCPFBERkUkwkKrAlKwjRUREZFIMpCowJpsTERGZFgOpCozlD4iIiEyLgVQFJlU259AeERGRSTCQqsBye6QYSBEREZkCA6kKTJq1l82hPSIiIlNgIFWBMdmciIjItBhIVWBMNiciIjItBlIVGHOkiIiITIuBVAWm4qw9IiIik2IgVYFxaI+IiMi0GEhVYPo6UhnskSIiIjIJBlIVGHukiIiITIuBVAXGZHMiIiLTYiBVgbGOFBERkWkxkKrAVJYc2iMiIjIlBlIVmL5HisnmREREpsFAqgLTr7WXybX2iIiITIKBVAXGZHMiIiLTYiBVgeVNNhdF0cStISIi+u9hIFWB6ZPNASBLw14pIiIiY2MgVYHpe6QADu8RERGZAgOpCszSQoAg6L7PYMI5ERGR0TGQqsAEQchNOGcJBCIiIqMzeSC1dOlSeHl5QaVSwd/fH4cPHy7y+MzMTEyfPh2enp5QKpXw9vbGqlWrpP3Z2dmYPXs2vL29oVKp0LhxYwQHBxucY9myZWjUqBHs7e1hb2+PwMBA7N692+CYx48fY+zYsfDw8ICVlRXq1q2LZcuWld0LLyOsbk5ERGQ6clM++ebNmzFhwgQsXboUbdq0wU8//YSePXsiPDwcNWrUKPAx/fv3R3x8PFauXAkfHx8kJCRArVZL+2fMmIH169fj559/hp+fH/bs2YN+/frh2LFjaNq0KQDAw8MD8+bNg4+PDwBg7dq1CAoKwvnz51G/fn0AwIcffoj9+/dj/fr1qFmzJv755x+MHj0abm5uCAoKKucrU3xcuJiIiMh0BNGE8+ZbtmyJZs2aGfT01K1bF3379sXcuXPzHR8cHIyBAwciIiIClStXLvCcbm5umD59OsaMGSNt69u3L2xtbbF+/fpC21K5cmV88803eOeddwAADRo0wIABA/Dpp59Kx/j7+6NXr1744osvivX6UlJS4ODggOTkZNjb2xfrMSXVfv5+RD54gq2jW6NZjUrl8hxERET/JSX5/DbZ0F5WVhbOnj2Lbt26GWzv1q0bjh07VuBjdu7ciYCAAMyfPx/u7u6oXbs2Pv74Y6Snp0vHZGZmQqVSGTzOysoKR44cKfCcGo0GmzZtQlpaGgIDA6Xtbdu2xc6dOxEdHQ1RFLF//35cv34d3bt3L/Q1ZWZmIiUlxeBW3pgjRUREZDomG9pLTEyERqOBs7OzwXZnZ2fExcUV+JiIiAgcOXIEKpUK27ZtQ2JiIkaPHo0HDx5IeVLdu3fHwoUL0b59e3h7e2Pfvn3YsWMHNBrDoa+wsDAEBgYiIyMDtra22LZtG+rVqyftX7x4Md577z14eHhALpdDJpPhl19+Qdu2bQt9TXPnzsWsWbNKe0lKRb9MTAaH9oiIiIzO5Mnmgn7+fg5RFPNt09NqtRAEARs2bECLFi3Qq1cvLFy4EGvWrJF6pRYtWgRfX1/4+flBoVBg7NixGD58OCwsLAzOVadOHYSGhuLEiRMYNWoUhg0bhvDwcGn/4sWLceLECezcuRNnz57FggULMHr0aOzdu7fQ1zJt2jQkJydLt6ioqNJelmKTks3ZI0VERGR0JuuRqlKlCiwsLPL1PiUkJOTrpdJzdXWFu7s7HBwcpG1169aFKIq4d+8efH19UbVqVWzfvh0ZGRlISkqCm5sbpk6dCi8vL4NzKRQKKdk8ICAAp0+fxqJFi/DTTz8hPT0dn3zyCbZt24bevXsDABo1aoTQ0FB8++23eOmllwpsn1KphFKpLPU1KQ0mmxMREZmOyXqkFAoF/P39ERISYrA9JCQErVu3LvAxbdq0QUxMDB4/fixtu379OmQyGTw8PAyOValUcHd3h1qtxpYtW545004URWRmZgLQlVDIzs6GTGZ4eSwsLKDVvlg9P1y4mIiIyHRMWv5g4sSJGDJkCAICAhAYGIgVK1YgMjISI0eOBKAbKouOjsavv/4KABg0aBC++OILDB8+HLNmzUJiYiImTZqEESNGwMrKCgBw8uRJREdHo0mTJoiOjsbnn38OrVaLyZMnS8/7ySefoGfPnqhevTpSU1OxadMmHDhwQKo3ZW9vjw4dOmDSpEmwsrKCp6cnDh48iF9//RULFy408lUqmsqSdaSIiIhMxaSB1IABA5CUlITZs2cjNjYWDRo0wK5du+Dp6QkAiI2NRWRkpHS8ra0tQkJCMG7cOAQEBMDJyQn9+/fHnDlzpGMyMjIwY8YMREREwNbWFr169cK6devg6OgoHRMfH48hQ4YgNjYWDg4OaNSoEYKDg9G1a1fpmE2bNmHatGkYPHgwHjx4AE9PT3z55ZdSkPeiyJ21x6E9IiIiYzNpHSlzZ4w6UlP+dxGbz0RhUvc6GNPJp1yeg4iI6L+kQtSRorKhL3/AHikiIiLjYyBVwTHZnIiIyHQYSFVwXLSYiIjIdBhIVXAqS9aRIiIiMhUGUhWcvkcqg5XNiYiIjI6BVAWnZI8UERGRyTCQquBy60ixR4qIiMjYGEhVcEw2JyIiMh0GUhUck82JiIhMh4FUBcceKSIiItNhIFXB6XOkMljZnIiIyOgYSFVwubP22CNFRERkbAykKjhpaI+z9oiIiIyOgVQFl7vWHof2iIiIjI2BVAWnsmSyORERkakwkKrgcnukGEgREREZGwOpCk6fI6XRisjWMJgiIiIyJgZSFZx+1h7AXikiIiJjYyBVwSks8gRSrCVFRERkVAykKjiZTJCCKfZIERERGRcDKTPAopxERESmwUDKDOgTzrlMDBERkXExkDIDLIFARERkGgykzIA0tMceKSIiIqNiIGUGpPX22CNFRERkVAykzICKyeZEREQmwUDKDHDhYiIiItNgIGUGcmftsUeKiIjImBhImQH2SBEREZkGAykzoLTMSTZnjxQREZFRMZAyA6wjRUREZBoMpMxA7qw9Du0REREZEwMpM8A6UkRERKbBQMoM6If2uNYeERGRcTGQMgPskSIiIjINBlJmIHetPQZSRERExsRAygywjhQREZFpMJAyAypLDu0RERGZAgMpM8A6UkRERKbBQMoM5K61x6E9IiIiY2IgZQbYI0VERGQaDKTMQO6sPfZIERERGRMDKTOgTzbPYo8UERGRUTGQMgMc2iMiIjINBlJmgMnmREREpsFAygywR4qIiMg0GEiZASnZnJXNiYiIjIqBlBnQD+1la0RotKKJW0NERPTfwUDKDKgsc3+MnLlHRERkPAykzIDCIvfHyOE9IiIi42EgZQbkFjLIZQIAICObPVJERETGwkDKTOTO3GOPFBERkbEwkDITypzq5iyBQEREZDwmD6SWLl0KLy8vqFQq+Pv74/Dhw0Uen5mZienTp8PT0xNKpRLe3t5YtWqVtD87OxuzZ8+Gt7c3VCoVGjdujODgYINzLFu2DI0aNYK9vT3s7e0RGBiI3bt353uuK1eu4JVXXoGDgwPs7OzQqlUrREZGls0LL2NSjxSH9oiIiIxGbson37x5MyZMmIClS5eiTZs2+Omnn9CzZ0+Eh4ejRo0aBT6mf//+iI+Px8qVK+Hj44OEhASo1Wpp/4wZM7B+/Xr8/PPP8PPzw549e9CvXz8cO3YMTZs2BQB4eHhg3rx58PHxAQCsXbsWQUFBOH/+POrXrw8AuHXrFtq2bYt33nkHs2bNgoODA65cuQKVSlXOV6V0VFKPFIf2iIiIjEUQRdFkhYdatmyJZs2aYdmyZdK2unXrom/fvpg7d26+44ODgzFw4EBERESgcuXKBZ7Tzc0N06dPx5gxY6Rtffv2ha2tLdavX19oWypXroxvvvkG77zzDgBg4MCBsLS0xLp160r78pCSkgIHBwckJyfD3t6+1Ocpjh7fH8LVuFRseLcl2vhUKdfnIiIiMmcl+fw22dBeVlYWzp49i27duhls79atG44dO1bgY3bu3ImAgADMnz8f7u7uqF27Nj7++GOkp6dLx2RmZubrNbKyssKRI0cKPKdGo8GmTZuQlpaGwMBAAIBWq8Xff/+N2rVro3v37qhWrRpatmyJ7du3F/maMjMzkZKSYnAzFv3QHtfbIyIiMh6TBVKJiYnQaDRwdnY22O7s7Iy4uLgCHxMREYEjR47g0qVL2LZtG77//nv873//M+h96t69OxYuXIgbN25Aq9UiJCQEO3bsQGxsrMG5wsLCYGtrC6VSiZEjR2Lbtm2oV68eACAhIQGPHz/GvHnz0KNHD/zzzz/o168fXn31VRw8eLDQ1zR37lw4ODhIt+rVq5f28pSYvro5k82JiIiMx+TJ5oIgGNwXRTHfNj2tVgtBELBhwwa0aNECvXr1wsKFC7FmzRqpV2rRokXw9fWFn58fFAoFxo4di+HDh8PCwsLgXHXq1EFoaChOnDiBUaNGYdiwYQgPD5eeBwCCgoLw4YcfokmTJpg6dSpefvllLF++vNDXMm3aNCQnJ0u3qKioUl+XkuJ6e0RERMZnskCqSpUqsLCwyNf7lJCQkK+XSs/V1RXu7u5wcHCQttWtWxeiKOLevXsAgKpVq2L79u1IS0vD3bt3cfXqVdja2sLLy8vgXAqFAj4+PggICMDcuXPRuHFjLFq0SGqbXC6XeqjyPldRs/aUSqU0E1B/MxbO2iMiIjI+kwVSCoUC/v7+CAkJMdgeEhKC1q1bF/iYNm3aICYmBo8fP5a2Xb9+HTKZDB4eHgbHqlQquLu7Q61WY8uWLQgKCiqyPaIoIjMzU2pb8+bNce3aNYNjrl+/Dk9Pz2K/RmNiHSkiIiLjM2n5g4kTJ2LIkCEICAhAYGAgVqxYgcjISIwcORKAbqgsOjoav/76KwBg0KBB+OKLLzB8+HDMmjULiYmJmDRpEkaMGAErKysAwMmTJxEdHY0mTZogOjoan3/+ObRaLSZPniw97yeffIKePXuievXqSE1NxaZNm3DgwAGDelOTJk3CgAED0L59e3Tq1AnBwcH4888/ceDAAeNdoBJgsjkREZHxmTSQGjBgAJKSkjB79mzExsaiQYMG2LVrl9TrExsbazCUZmtri5CQEIwbNw4BAQFwcnJC//79MWfOHOmYjIwMzJgxAxEREbC1tUWvXr2wbt06ODo6SsfEx8djyJAhiI2NhYODAxo1aoTg4GB07dpVOqZfv35Yvnw55s6di/Hjx6NOnTrYsmUL2rZtW/4XphSYbE5ERGR8Jq0jZe6MWUfq852XsebYHYzp5I1J3f3K9bmIiIjMWYWoI0VlS5q1x2RzIiIio2EgZSZUHNojIiIyOgZSZoJ1pIiIiIzvuZLNz549iytXrkAQBNStWxfNmjUrq3ZRCemTzTM4tEdERGQ0pQqkEhISMHDgQBw4cACOjo4QRRHJycno1KkTNm3ahKpVq5Z1O+kZpIKc7JEiIiIymlIN7Y0bNw4pKSm4fPkyHjx4gIcPH+LSpUtISUnB+PHjy7qNVAy5gRR7pIiIiIylVD1SwcHB2Lt3L+rWrSttq1evHpYsWYJu3bqVWeOo+KTK5hzaIyIiMppS9UhptVpYWlrm225paSkt+EvGpeLQHhERkdGVKpDq3LkzPvjgA8TExEjboqOj8eGHH6JLly5l1jgqPq61R0REZHylCqR+/PFHpKamombNmvD29oaPjw+8vLyQmpqKH374oazbSMXAtfaIiIiMr1Q5UtWrV8e5c+cQEhKCq1evQhRF1KtXDy+99FJZt4+KicnmRERExlfiQEqtVkOlUiE0NBRdu3Y1WOiXTIeLFhMRERlfiYf25HI5PD09odFwCOlFkrvWHn8uRERExlKqHKkZM2Zg2rRpePDgQVm3h0pJxWRzIiIioytVjtTixYtx8+ZNuLm5wdPTEzY2Ngb7z507VyaNo+LLmyMliiIEQTBxi4iIiMxfqQKpvn37lnEz6HnpAylAF0zpe6iIiIio/JQq2RwARowYgerVq5d5g6h09MnmAAMpIiIiYylVsvm3337LZPMXjKWFAP1oHqubExERGUepks27dOmCAwcOlHFT6HkIggCVnOvtERERGVOpcqR69uyJadOm4dKlS/D398+XbP7KK6+USeOoZJSWMqRnazhzj4iIyEhKFUiNGjUKALBw4cJ8+wRB4LCfiXCZGCIiIuMqVSCl1bLH40XE6uZERETGVaIcqV69eiE5OVm6/+WXX+LRo0fS/aSkJNSrV6/MGkclk1tLij1SRERExlCiQGrPnj3IzMyU7n/99dcG1c3VajWuXbtWdq2jEpGWiWGPFBERkVGUKJASRbHI+2RanLVHRERkXKUqf0AvptweKQ7tERERGUOJAilBEPKt4cY13V4cSvZIERERGVWJZu2Jooi3334bSqUSAJCRkYGRI0dKdaTy5k+R8THZnIiIyLhKFEgNGzbM4P5bb72V75ihQ4c+X4uo1HIDKfZIERERGUOJAqnVq1eXVzuoDLCOFBERkXEx2dyMqPTJ5qxsTkREZBQMpMyI0pI9UkRERMbEQMqMcK09IiIi42IgZUaYbE5ERGRcDKTMCJPNiYiIjIuBlBlhZXMiIiLjYiBlRrjWHhERkXExkDIj+h6pDPZIERERGQUDKTMiJZuzR4qIiMgoGEiZESabExERGRcDKTPCRYuJiIiMi4GUGWFlcyIiIuNiIGVGmCNFRERkXAykzIiKs/aIiIiMioGUGVGyjhQREZFRMZAyI3mTzUVRNHFriIiIzB8DKTOi75HSioBay0CKiIiovDGQMiP6yuYAZ+4REREZAwMpM6If2gOAzGwmnBMREZU3BlJmRBAEKOT6mXvskSIiIipvDKTMTG4tKfZIERERlTcGUmaG6+0REREZDwMpM5NbAoGBFBERUXkzeSC1dOlSeHl5QaVSwd/fH4cPHy7y+MzMTEyfPh2enp5QKpXw9vbGqlWrpP3Z2dmYPXs2vL29oVKp0LhxYwQHBxucY9myZWjUqBHs7e1hb2+PwMBA7N69u9DnfP/99yEIAr7//vvneq3GoK9uzqE9IiKi8ic35ZNv3rwZEyZMwNKlS9GmTRv89NNP6NmzJ8LDw1GjRo0CH9O/f3/Ex8dj5cqV8PHxQUJCAtRqtbR/xowZWL9+PX7++Wf4+flhz5496NevH44dO4amTZsCADw8PDBv3jz4+PgAANauXYugoCCcP38e9evXN3i+7du34+TJk3Bzcyunq1C2OLRHRERkRKIJtWjRQhw5cqTBNj8/P3Hq1KkFHr97927RwcFBTEpKKvScrq6u4o8//miwLSgoSBw8eHCRbalUqZL4yy+/GGy7d++e6O7uLl66dEn09PQUv/vuuyLP8bTk5GQRgJicnFyixz2PvkuOiJ5T/hL3XIo12nMSERGZk5J8fptsaC8rKwtnz55Ft27dDLZ369YNx44dK/AxO3fuREBAAObPnw93d3fUrl0bH3/8MdLT06VjMjMzoVKpDB5nZWWFI0eOFHhOjUaDTZs2IS0tDYGBgdJ2rVaLIUOGYNKkSfl6qQqTmZmJlJQUg5uxMUeKiIjIeEw2tJeYmAiNRgNnZ2eD7c7OzoiLiyvwMREREThy5AhUKhW2bduGxMREjB49Gg8ePJDypLp3746FCxeiffv28Pb2xr59+7Bjxw5oNIY5Q2FhYQgMDERGRgZsbW2xbds21KtXT9r/9ddfQy6XY/z48cV+TXPnzsWsWbOKfXx54NAeERGR8Zg82VwQBIP7oijm26an1WohCAI2bNiAFi1aoFevXli4cCHWrFkj9UotWrQIvr6+8PPzg0KhwNixYzF8+HBYWFgYnKtOnToIDQ3FiRMnMGrUKAwbNgzh4eEAgLNnz2LRokVYs2ZNoW0pyLRp05CcnCzdoqKiSnIpyoSUbK5msjkREVF5M1kgVaVKFVhYWOTrfUpISMjXS6Xn6uoKd3d3ODg4SNvq1q0LURRx7949AEDVqlWxfft2pKWl4e7du7h69SpsbW3h5eVlcC6FQgEfHx8EBARg7ty5aNy4MRYtWgQAOHz4MBISElCjRg3I5XLI5XLcvXsXH330EWrWrFnoa1IqldJMQP3N2KQeqWz2SBEREZU3kwVSCoUC/v7+CAkJMdgeEhKC1q1bF/iYNm3aICYmBo8fP5a2Xb9+HTKZDB4eHgbHqlQquLu7Q61WY8uWLQgKCiqyPaIoIjMzEwAwZMgQXLx4EaGhodLNzc0NkyZNwp49e0rzco1GKS0Rwx4pIiKi8mbS8gcTJ07EkCFDEBAQgMDAQKxYsQKRkZEYOXIkAN1QWXR0NH799VcAwKBBg/DFF19g+PDhmDVrFhITEzFp0iSMGDECVlZWAICTJ08iOjoaTZo0QXR0ND7//HNotVpMnjxZet5PPvkEPXv2RPXq1ZGamopNmzbhwIEDUr0pJycnODk5GbTV0tISLi4uqFOnjjEuTakppTpS7JEiIiIqbyYNpAYMGICkpCTMnj0bsbGxaNCgAXbt2gVPT08AQGxsLCIjI6XjbW1tERISgnHjxiEgIABOTk7o378/5syZIx2TkZGBGTNmICIiAra2tujVqxfWrVsHR0dH6Zj4+HgMGTIEsbGxcHBwQKNGjRAcHIyuXbsa7bWXFyabExERGY8giqJo6kaYq5SUFDg4OCA5Odlo+VLzg69i6YFbGN6mJmb2KV7ZBiIiIspVks9vk8/ao7KlsmSPFBERkbEwkDIzUkFO5kgRERGVOwZSZoaz9oiIiIyHgZSZUVqyjhQREZGxMJAyM7lr7bFHioiIqLwxkDIzLH9ARERkPAykzEzuWnsMpIiIiMobAykzk7vWHof2iIiIyhsDKTOjZI8UERGR0TCQMjO5daTYI0VERFTeGEiZGSabExERGQ8DKTOTW/6AgRQREVF5YyBlZnLX2uPQHhERUXljIGVm9D1S2RoRGq1o4tYQERGZNwZSZkY/aw9grxQREVF5YyBlZhQWeQIprrdHRERUrhhImRm5hQxymQCACedERETljYGUGWLCORERkXEwkDJDLIFARERkHAykzJA+kMpgdXMiIqJyxUDKDCktWd2ciIjIGBhImaHc9fYYSBEREZUnBlJmKDdHikN7RERE5YmBlBni0B4REZFxMJAyQ+yRIiIiMg4GUmZIKdf1SGUwR4qIiKhcMZAyQ/r19jJZ/oCIiKhcMZAyQyzISUREZBwMpMyQfmiPgRQREVH5YiBlhlSWTDYnIiIyBgZSZkjqkWKyORERUbliIGWGpLX22CNFRERUrhhImaHcWXvskSIiIipPDKTMEJPNiYiIjIOBlBlisjkREZFxMJAyQ+yRIiIiMg4GUmZISjZnZXMiIqJyxUDKDLGyORERkXEwkDJDSkvWkSIiIjIGBlJmKLdHikN7RERE5YmBlBlSWTLZnIiIyBgYSJkh5kgREREZBwMpM8RZe0RERMbBQMoMKTm0R0REZBQMpMyQvkcqS62FKIombg0REZH5YiBlhvSBFMBeKSIiovLEQMoM6WftAQykiIiIyhMDKTMklwmQCbrvWUuKiIio/DCQMkOCIOQuXMzq5kREROWGgZSZUlqyujkREVF5YyBlpnJrSbFHioiIqLwwkKqIrgUDW94FQn8r9BAuE0NERFT+GEhVRPGXgLA/gBshhR7ChYuJiIjKn8kDqaVLl8LLywsqlQr+/v44fPhwkcdnZmZi+vTp8PT0hFKphLe3N1atWiXtz87OxuzZs+Ht7Q2VSoXGjRsjODjY4BzLli1Do0aNYG9vD3t7ewQGBmL37t0G55gyZQoaNmwIGxsbuLm5YejQoYiJiSnbF19a7v66rzHnCj2EyeZERETlz6SB1ObNmzFhwgRMnz4d58+fR7t27dCzZ09ERkYW+pj+/ftj3759WLlyJa5du4aNGzfCz89P2j9jxgz89NNP+OGHHxAeHo6RI0eiX79+OH/+vHSMh4cH5s2bhzNnzuDMmTPo3LkzgoKCcPnyZQDAkydPcO7cOXz66ac4d+4ctm7diuvXr+OVV14pv4tREm5NdV8f3gGePCjwEPZIERERlT9BNOEaIi1btkSzZs2wbNkyaVvdunXRt29fzJ07N9/xwcHBGDhwICIiIlC5cuUCz+nm5obp06djzJgx0ra+ffvC1tYW69evL7QtlStXxjfffIN33nmnwP2nT59GixYtcPfuXdSoUaNYry8lJQUODg5ITk6Gvb19sR5TbD/4A0k3gcFbAN+X8u0e/MsJHL2ZhEUDmyCoiXvZPjcREZEZK8nnt8l6pLKysnD27Fl069bNYHu3bt1w7NixAh+zc+dOBAQEYP78+XB3d0ft2rXx8ccfIz09XTomMzMTKpXK4HFWVlY4cuRIgefUaDTYtGkT0tLSEBgYWGh7k5OTIQgCHB0dCz0mMzMTKSkpBrdy49ZM97WQ4T0O7REREZU/uameODExERqNBs7OzgbbnZ2dERcXV+BjIiIicOTIEahUKmzbtg2JiYkYPXo0Hjx4IOVJde/eHQsXLkT79u3h7e2Nffv2YceOHdBoDIe4wsLCEBgYiIyMDNja2mLbtm2oV69egc+bkZGBqVOnYtCgQUVGpnPnzsWsWbNKchlKz70ZEPY7EH22wN0q1pEiIiIqdyZPNhcEweC+KIr5tulptVoIgoANGzagRYsW6NWrFxYuXIg1a9ZIvVKLFi2Cr68v/Pz8oFAoMHbsWAwfPhwWFhYG56pTpw5CQ0Nx4sQJjBo1CsOGDUN4eHi+58zOzsbAgQOh1WqxdOnSIl/LtGnTkJycLN2ioqJKcilKRp9wHn0OKGB0VuqRYvkDIiKicmOyQKpKlSqwsLDI1/uUkJCQr5dKz9XVFe7u7nBwcJC21a1bF6Io4t69ewCAqlWrYvv27UhLS8Pdu3dx9epV2NrawsvLy+BcCoUCPj4+CAgIwNy5c9G4cWMsWrTI4Jjs7Gz0798ft2/fRkhIyDPHSZVKpTQTUH8rNy4NAZkcSEsAku/lb4tUkJM9UkREROXFZIGUQqGAv78/QkIMayGFhISgdevWBT6mTZs2iImJwePHj6Vt169fh0wmg4eHh8GxKpUK7u7uUKvV2LJlC4KCgopsjyiKyMzMlO7rg6gbN25g7969cHJyKulLLF+WVkC1nKHIAvKkcmftsUeKiIiovJh0aG/ixIn45ZdfsGrVKly5cgUffvghIiMjMXLkSAC6obKhQ4dKxw8aNAhOTk4YPnw4wsPDcejQIUyaNAkjRoyAlZUVAODkyZPYunUrIiIicPjwYfTo0QNarRaTJ0+WzvPJJ5/g8OHDuHPnDsLCwjB9+nQcOHAAgwcPBgCo1Wq8/vrrOHPmDDZs2ACNRoO4uDjExcUhKyvLiFfoGdxzEs4LyJNSsrI5ERFRuTNZsjkADBgwAElJSZg9ezZiY2PRoEED7Nq1C56engCA2NhYg5pStra2CAkJwbhx4xAQEAAnJyf0798fc+bMkY7JyMjAjBkzEBERAVtbW/Tq1Qvr1q0zmG0XHx+PIUOGIDY2Fg4ODmjUqBGCg4PRtWtXAMC9e/ewc+dOAECTJk0M2rx//3507NixfC5ISbn7A2fX6PKkniL1SHFoj4iIqNyYtI6UuSvXOlIAEHcJWN4GUNgBUyMBWW4H45L9N/HNnmsY2Lw65r3WqOyfm4iIyExViDpSVAaq+gGW1kBWKpB0w2AXc6SIiIjKHwOpisxCDrg21n3/VJ4UZ+0RERGVPwZSFZ2+wvlTeVKsI0VEL6LLMclYd/wOtFpmlZB5MGmyOZWBQmbuKVnZnIheQJP/dxGXY1LgZKtEr4aupm4O0XNjj1RFpw+k4i8B6tzSDLmz9tgjRUQvhidZalyJ1a1BuvdKvIlbQ1Q2GEhVdJW8AKtKgCZLF0zl0NeRSstijxQRvRjCY1KgH9E7eO0+h/fILDCQqugEIU+eVO7wXq0qNhAE4EpsCi5FJ5uocUREuS7ey/1blJSWhTD+bSIzwEDKHOgXMI45L23ydLLBK43dAAAL/rlmilYRERm4eO8RAN3/fwCw/1qC6RpDVEYYSJmDQhLOP3ypNixkAvZfu4/Tdx6YoGFERLku5vRA9WqgSzLff5WBFFV8DKTMgX5o7/41IDNV2lyzig36B+gWc/5mzzWwiD1R+bmfmomPfr+As3f5T0tBUjOyEXE/DQAwtrMPAODCvWTcT80s6mFELzwGUubAzhmw9wAgAjGhBrvGdfaFQi7DqdsPcPhGokmaR/RfsGjfdWw5dw/zdl81dVNeSJeidbP13B2tUNfVHvXddMtuHLp+35TNInpuDKTMhXtT3dcYw8Kcbo5WGNJKtwg0e6WIykdKRja2nosGAJyLfITUjGwTt+jFExb9CADQyMMBANDZrxoA4F/mSVEFx0DKXOgTzp/KkwKA0R29YaOwQFh0MvZcjjNyw4jM39az9/Akp9SIRivi+K0kE7foxaOfsdcwJ5DqWEcXSB26fh9qDevdUcXFQMpcSCUQzufb5WSrxIi2XgCAb/+5Dg1rtxCVGa1WxK/H7wIAqtgqAIDD6AXQlzpo5O4IAGhS3RGO1pZIzVDjXOQj0zWM6DkxkDIXbk0ACEByJPA4f87Bu+1qwcHKEjcTHmNHaHSZP33wpTj0X34ctxPTyvzcRC+yo7cSEZGYBlulHDN61wMAHL7BvJ+8kp9k427SEwBAQ3ddj5SFTECH2lUBsAwCVWwMpMyFygGo4qv7/qk8KQBwsLLEyA7eAIDv9l5HVhkuZvwwLQtTtlzEqTsP8OO/N8vsvEQVgb436rVm7nipnjPkMgF3kp4gMidwIOBiTn5UTSdrOFhbStv1eVIsg0AVGQMpc1JEnhQADGvtiap2SkQ9SMfmM1Fl9rTf772O5HRdcu1fF2OQ/KT0ibbZGi0WhlzHQc7koQog6sET7MtZM25IYE3YKuVo5lkJAHD4Jt/Dern5UY4G29v7VoUgAFfjUhHzKN0ELSN6fgykzImUJ5W/RwoArBVyjO2kq9/yw74bSC+Ddfiux6di/clIAICTjQKZai22nb9X6vP9fiYKi/fdwPiN55GRzXUC6cW24WQktCLQxscJPtVsAQDtfasAAA5fZ56UXtg9fX6Ug8H2SjYKNK3uCAA4cI2BJ1VMDKTMib7Cecw5oJAyBwNbVIe7oxUSUjOx7sSd53o6URTxxV/h0GhFdK3njA9e0g0t/nYqslRlFrRaESuP3AYAJKdn488LMc/VPqLylJGtwebTun8ihgbWlLa389Xl/Ry9lcjZaDn0ieb6GXt5ScN7zJOiCoqBlDlxbgDILIEnScCjuwUeopRbYEJOwLP0wK3nqnez/1oCDt9IhKWFgOm96iKoiTtUljJcj3+Mc5EPS3y+g9fvS5WPAWDdiYJfA9GL4K+LsXj4JBvujlbokhMMAEADdwdpNtqFnLXl/ssSH2ci+lE6BAFSEc689GUQjt5MRKaavdBU8TCQMieWKsC5vu77Qob3AKBfU3fUqmqDR0+y8cvh26V6qiy1FnP+ugIAGNHGCzWr2MDByhJ9GukWSt6QM9xXEr8ciQAAvNbMAwq5DBfvJeNC1KNStY+oPImiiLXH7gAABreqAblF7p9SC5mANj664b1DJh7e02pFxKdkmLQN+t6oWlVsYKeyzLe/vps9qtkp8SRLg1O3ubwOVTwMpMzNMxLOAUBuIcNHXesAAH45HIEHaVklfpp1J+4iIjENVWwV0rpZAPBmyxoAgL8vxpYo6fxKbAqO3kyCTAA+7OqLlxu6Ss9D9KIJjXqEsOhkKCxkGBBQPd9+KU/KxGUQftx/Ey2/2oe5u6+YbFUDKT/qqURzPUEQ0LFOThmEq8yTooqHgZS5kfKk8hfmzKtnAxfUd7NHWpYGyw6UrGTBg7QsLNp7HQDwcbc6Bv9lNq3uCD8XO2SqtdhagqTzVTm5UT0busKjkjXeCtQta/PnhRg8LEWgRwQAR24k4vfTUUh6XLYL467LKXnwcmNXONkq8+3X50mFRj2SZrSWxJEbifj3avxztVGrFbHplK5n+KeDEVi8zzSlSS7mDG82dM+fH6Wnz5M6UII8qdSMbPyw7waux6c++2CicsRAytzoe6RiQgFt4fkGMpmAj7vreqVWHrmNVUduF/s/1oUh15CSoUZdV3u88dR/44IgYFBOr9TGYiadJ6RmYEeoLrH8nZwK7E2rO6K+mz0y1Vr8cbbsSjXQf8edxDQMW30Kk7dcRIuv9mHYqlPYcvbec6+Dl/g4E39djAUADMuTZJ6Xm6MVfKrZQisCx2+VbHgv6sETDFt9Cu+sPYO7SaUvcHs+6hFikjMglwkAdPXjfj4UUerzlZa+9EHj6oUHUm18qkAuExCRmIY7xSjqq9WKGL/xPBaEXMeETaFcQ5RMioGUualSG7C0AbLTgPvXijy0Y+2qeLNFdWhFYPZf4Zi6JeyZhTqvxaXit5z8p5l96sEi5490Xn2b5iadn7377KTz9ScikaXRomkNRzSroavBIwgChub0Sq0/EQktl7WhElp64CY0WhE2CgtotCIOXr+Pj/64AP85ezFy3VnsCostVYmNzaejkKXRorGHAxrnTN0vSLuc4b1DJVwuZvXRO9BoRYgisP186Weu/p0T7L3cyBUfda0NAPhy1xWsN+JweXxKBhJSMyETgHquhQdSdipLNK9ZGUDxZu/98O9N7M8plxAem1LuuVWiKGLDybuYtvUiUrggNT2FgZS5kVnkLBeDIvOkAF2w8lW/hpjRuy5kArD5TBQG/3ICiYUMg+jLHWhF3dBgq1pOBR5nr8pNOv/tVNFJ5xnZGmzI+cP+bttaBvteaewOO5UckQ+e4CCX3KASiHrwBFvP6ZZCWvduS+z7qAMmvOSLWlVtkKXWIvhyHEZvOIeAOXsxcXMoDl2/X6xeDbVGK71fhxbSG6XXPmd4r7jnBnRlP/QlFQBge2h0qUuJ7ArTB1JuGNvZR1rZ4NMdl7D1XOlrvZWEvjeqtrMdrBQWRR7byU+/XEzRv+v7ryXg+3261ALfnNpdq4/eec6WFu5xphpjfzuP6dsuYeOpKCzZz9UbyBADKXOUt57UMwiCgHfb1cLKt5vDTinH6TsPEfTjUYTHpOQ7du+VBBy5mQiFXIZPetUt8ryDipl0vv18NJLSsuDuaIXu9Z0N9lkpLPCGv27ocP1xJp2X1KHr9+H/RQhWHy3dzMyKbNnBW1BrRbTzrYJmNSrBu6otJrxUG/smdsDf49vi/Q614O5ohceZamw9H42hq07hrZUnceMZ+TZ7ryQgJjkDlW0U6N3ItchjW9aqDEsLAfcepkvrzD3LxlORSMvSwLuqDawsLXA7MQ0XcoKRkjgb+RBxKRmwU8rRrnYVCIKAKT3qYGigJ0QR+PiPC9idE2iVp7Bi5Efp6fOkTkQk4UmWusBjIpOe4ION5yGKwOCWNbB0sO5v3T/hcYh6UPZL8txMSEXQj0fwd1gshJzO91+P3cX91LLNufuvEEURh67fx2vLjqHH94dw7KZ5FK1lIGWOpArnRfdI5dWpTjVsG9MGXlVsEP0oHa8tO4bgS7l/aDPVGnz5dzgA4N22Xqhe2brI8zUpRtK5KIpYlfMh/3brmgZTyPXeaqULyP69llAufyjNVXxKBiZsDkVSWha+2nUF1+L+Owm5MY/S8UfOEkjjOvsa7BMEAfXdHDCtZ10cntwJW0YFYmigJxRyGY7eTELPRYfx5d/hheZR/Xr8DgBgYPPqUFkW3cNirZAjwFM3XHWoGD2qWWqtFPS+38Fb+sdi+/mSLzL+V04x2671naGU69opCAI+71Mfr/t7QCsC4zedL/cimBej9TP2nh1IeVe1hUclK2SptTh2Mynf/vQsDd5ffxYpGWo0qe6Iz/rUg6+zHdr5VoFWhFSOoqz8dTEGQT8exa37aXCxV+F/I1ujcXVHpGdrsPzgrVKfN1OtwboTd8tkLcbbiWm4FF3yQNsUTt1+gAE/ncDQVadw9u5DXI1LxaBfTmLSHxcq/IQiBlLmSJ9wHn8ZyC5+DRmfarbYProN2vlWQXq2BiPXn8OivTcgiiJ+PXYXd5KeoKqdEqM7+TzzXIIgYHBOr9RvJwtOOj98IxHX4x/DRmGBAS3yTyEHgFpVbdHOtwpE8dnDhKSj0YqYsCkUD9KyIAhAtkbE5P9d+M9U2f7p4C1ka0S09KqMFl6VCz1OJhPg71kZs4MaYO+HHdC1njPUWhE/H76NzgsOYuu5ewbv25sJqTh2S1eiY3Arz2K1pV3t4teT+utiDOJTMlHVTomgJm7o29QdgG7manYJfnYarYhdl+IAQBpi15PJBHz9WiP0buSKbI2IkevO4kRE/qClICUdYhRFUSp98PQaewURBAGd6hRc5VwURUzfFoYrsSlwslFg2VvNpABxRBvdBJXNZ6LwOLPgnqySyNZoMfvPcIz97TzSsjQIrOWEv8a3hb9nJUzMyTVbf+JuqetzfbvnGj7dfgnDVp96rmWwoh48Qe/Fh/HyD0de6F7ni/ceYeiqU+j/03GcuvMACrkMI9p44a1WNSAIwB9n7+GlhQexo5TD2C8CBlLmyLEGYO0EaNVAXFiJHupgbYnVbzfH8DY1Aehm+oxcfxaL990AAEzqXge2SnmxzhXU1B1Wlha4kVBw0vkvOSUP+jevDvsCCvXpvZXzobX5dFSFX38vNOoR9l9NKNcKzssP3sLxiCRYKyyw4d2WsFPJceFestT7Z84SUjKw8bSuN+qDLr7PODpXDSdr/Dw0AKuHN4dXFRvcT83ExN8v4I3lx3E5RhcM/JozvPxSXWe4O1oV67z6PKnjtxKLDIZEUcSKnBl1b7euCaXcAm19qqCKrQJJaVk4UoKE9VO3H+B+aiYcrCylwqB5WcgEfNe/Cbr4VUOmWot31pzG+TwrEWRka3AlNgU7L8Rg4T/XMGr9WXRdeBC1Z+zGqPXF7+WOSc5AUloW5DIBfi52xXpMbhkEw7yy9SfuYuv5aMgE4IdBTeHqkHv9O9SuilpVbJCaocaWs8+X+5WQkoFBP5+QfldGdvDGundaoEpOiYv2vlXg71kJmWotlpYiVyo8JgWrcvK5biemSX9XS0oURUzbGoYnOeulzvoz/IXL3boWl4r3153BKz8exaHr9yGX6WZ0H5zUEZ/1qYc5fRvifyMDUdvZFklpWfhgUyiGrT5dIUceGEiZI0HIUwbh2XlST5NbyDCzT33Me7UhLC0E7Lkcj9RMNRq6O+D1Zh7FPo+9yhJ9GuvySH57qtL5jfhUHLp+HzIBGN7aq8jzdPGrBjcHFR6kZWH3pfLP6ygvt+4/xhvLj2H4mtNo+dU+TN8WhrN3H5Tpf2Fn7z7AwhBdIu6sV+qjtXcVfNq7HgBgwT/XEXH/cZk914toxaEIZKm18PeshEDvgidDFKVTnWoIntAOk3vUgZWlBc7cfYg+PxzB9G1h0of0sNY1i32+eq72cLJRIC1Lg/ORjwo97ujNJFyNS4WVpYXUkyu3kKFPY12P0tYSDO/9HaYb1ute3xkKecF/4hVyGZYMbobW3k5Iy9Jg2KpTeO/XM+j07QHU+ywYPRcdxviN57H435vYfSkONxIeI1sjYvelOBy/VbwerIs5qxLUcbF75jCoXqtaTlDKZYh+lI4bCbr36tm7DzH7L11awdSefmjtbRgcymSC9I/fmmN3Sj3D99TtB+j9wxGcvvMQdko5fhrij6k9/QxSDgRBkGZAbjwVhZhH6cU+v0Yr4pNtYdBoRSlJ/qdDEVKgXhJ/nL2HIzcToZTLMCTnH81v9lzD18FXTd6rE3H/MT7YdB49Fh3CnsvxEATg1abu2PdRB3zVr6FBEOzvWRl/jWuHj7vVhkIuw6Hr99H1u4NYcehWhepBZyBlrvR5UjdCgIzSjaEPbFEDG95tBScbBeQyATP71IOsgHIHRXmzhe5D4a+wWDx6kjsOrv+Pr1s9F9RwKjrfSm4hk5LX11XgpPO5u64iWyPCQibg0ZNsbDgZideWHUfHbw/gu5DrxaqfU5Tk9GyM3xgKjVbEK43d8Lq/Luh9I8AD7XyrIFOtxdQtYWZbSiLxcSbWn9S9P8Z19oEglOy9qqeUW2B0Rx/8+3EH9GnsBq2oW/JInwTeugQBmkwmoG0xqpyvOKzrjRrQvDocrRXS9n45w3v/XI4rVv0rtUaL3WG6Yb3eTw3rPU1laYGfhwbA37MSUjLUCAmPx+3ENGhFwF4lh79nJQxsXh0zetfF2hEtpPdTcXtRcvOjHIt1PKCbYKIPgP+9moD7qZkYveEssjUiejV0wXvtahX4uFebecBeJcftxDQcuF7yvK9fj9/Bmz+fwP3UTNRxtsOOsW3Qvb5LgccGejuhpVdlZGm0+LEEvUC/nYpEaNQj2CrlWP9uS/Rs4AKNVsTULWElChoSUjIwJyewnNi1Nr7o2wDTcyb/LDtwC5/vvGz03/H7qZlYe+wOXlt2DJ0XHMSO0BiIItCroQv+mdAeCwc0gaeTTYGPVchlGNvZF8EftEOrWpWRka3FV7uu4pUfj0rFXF90DKTMVfXmuq83Q4BvfICNg4Cw/wFZJfuwbuFVGfsndcS/H3VEQM3C800K06S6I+q62iNLrZWmoyc9zsSWnO/faVd0b5Re/+bVYWkh4FzkowqTXJnXsZuJ2HslHhYyAcEftMP6d1ri1WbusFZY4G7SEyzadwMdvz2AV5cexboTd0ucfCmKIj7ZGoboR+moUdkaX/ZrIAUS+jIX1goLnLrzQAo2zM0vh28jI1uLRh4O6FC76nOfz9XBCj+82RQb32uF2s66HoTRHUseoOmrnBdWT+pqXIrUO6vP99Fr6O6AWlVtkKnWYs/lZ1c6P3n7AZLSslDJ2rJYAZ+NUo7Vw5vjo6618Xmfetjwbkuc+qQLLszshi2jWmPea43wbrta6FC7Kj7sWhuWFgKORyQVq25T7tIwz040z0ufJ7U3PB5jfzuH+JRM+FSzxfzXGxd67W2UcgzM+adt1ZE7JXq+kPB4fLbjMjRaEX2buGHbmNaoVdW20OMFQZBypX4/HVWsoaiE1AzMD74KAPi4W20426swK6g+7FVyhEWXbNj9sx2XkZKhRiMPB6mA8Xvta+X8zgNrj9/F5C0Xy71HJyUjG7+ficKQlSfR8qu9mLnzMs7efQhB0I0i/DWuLZYO9oevc/GGdWtVtcXG91ph/muN4GBlifDYFPRdchTzg6++8ItZFy/ZhSqeWp2Alz4HQn8DEq8D1/7W3SytgdrdgQavAT5ddQsdP4O9yrLIHKaiCIKAQS2q49Mdl/HbqUgMb1MTG05GIkutK2gY4FmpWOepZqdCjwau+PNCDNafuIt5rzUqVXvy0mhFrD56G1vPRUMEoLAQoJDLYGmhuynkMigsZLC0EGBvZYnhbbzgVaXg/6qe9Txz/tYt8PxWyxrwdbaDr7Md2vpWwZy+avxzOR5bz0fjyI37OBf5COciH+HLv8Mxvosv3mtXC5YFzGZ82qbTUfg7LBZymYDFbzbNtzhs9crWmNLDDzN3Xsa83VfRqU61Z868rEgepmVhXc6MuvGdfUvdG1WQQG8n7BrfDvcfZxoMSxSXvjDnxXuP8OhJlkGPEwBp4fAeDfL3zgqCgH5N3LEg5Dq2n4+WeoUK89fFGOlcxXnfALrf73HFyCdzd7TC6/7VsfFUJBbvu4H177Ys9FhRFIu1NExBOtWphpm4jDM5eZU2Cgssf8v/mbmZQwM98cvhCBy5mYhrcamoU4y8rKgHT/DR76EAdLlpM/vUK9Z7p2UtJ7T1qYIjNxPx47838fXrRf89+uKvK0jNCX6G5NQfq2anwoze9TB5y0UsDLmO7vVdCu210dsdFovgy3GQ50wayDvsOLilJ6wVFvj4j4v439l7SM/S4LsBTQod3i2NjGwN9l1JwM4L0dh/9T6y8gRrjas74pXGbni5kSuc7Z/9uVIQQRDQv3l1dK5bDbP/DMfOCzFYeuAW/r2agIX9m6Cem31ZvZQyxR4pcyWzANp+CIw5BYw8CrT7CKhUE8h+AlzeBmx+S9dTtfV9IPJkuTZFn3R+M+Exjt1KkpJ2R7T1KtEHnj4XYHtodKnWL8srIidfac7fVxAem4IrsSm4cC8Zp+88xLFbSTh4/T5CwuPxd1gstofG4Nfjd/HWLydL9bxbzt1DeGwK7FRyfPBSbYN91go5+jZ1x68jWuDEtC6Y0bsu/FzskJGtxfzga+i9+DDO3Cn6v/8b8amY9edlALrJAE0KqbY9pJUnmteshCdZGnyyLczkuRRlafXR20jL0qCeqz261K1W5ueXW8hKFUQBgLO9CnWc7SCKulyovOJTMrAjVNc7+24hw1b62XtHbyUWOVMsW6NFcM5svZefMaxXWqM7ekMuE3DkZiLO3i38fRn54AlSMtRQyGWoXcweCb0aTtbwrpobUHz7RmP4VCu8h0jPo5I1ejTQDcetOfbsHp5MtQZjfjsnlVP4pFfdEv09+jCnV+p/5+4VOSx/6Pp9/HkhBjIB+KpfQ4PVIN4I8EBrbydkZGsxbWvRv5OPnmTh0x263/PRHb1R1zV/UNGvqQeWDGoGSwsBf4fF4v11Z8pkgo4oivjtZCSaf7kXY347hz2X45Gl0cK3mi0+6lobBz7uiB1j2uCdtl6lDqLyqmKrxOI3m2L5W81Q2UaBq3GpCFpyBEv233whc6cYSJk7QQBcGgBdPgPGhwLv7QcCxwL2HkBWKnBxE7C6B3B2Tbk1IW/S+YebQ5H4OBOuDir0alh0QcOnNa9ZSQoy/lfK2TlarYiVR26j56LDOBepy1eYHVQfv45ogZXDArD8rWZYNLAJvn2jMea+2hCzXqmPGb3rokZla0Q/SsfULRdLFIA8yVLj2z26pXrGd/ZFZRtFocdWs1fh3Xa1sPuDdljwRmNUtlHgevxjvL78OKZuuWiQY6aXka3BuI3nkZGtRTvfKoXmkAC5U9+VchkO30jEH885w6m8ZGRrsP9qQrGHN5PTs6XK1s+TG1We2hWSJ7X22B1ka0QEeFaSlkd6WvXK1mhesxJEEdgZWviSMcduJeHhk2w42SjQsoiyD8+jemVrvNZMnytVeH6QvqJ5XVf7UvWIBDXRBY+jOnqjZwn+TgzPGRrdei4aD57x/vny7yu4eC8ZjtaWWDK4WYnb6e9ZCR3rVIVGK2LxvwXnjWVkazBj+yUAukkKDZ7qnRMEAXNfbQiVpQzHbiXhjzOF/07O+fsKEh/rhjnHdC68BE2PBi74ZVhzqCxl2H/tPoavPv1cZSHiUzIwfM1pfLItDKkZarg7WmFUR2/s/qAd/vmwPcZ18UXNUvTUF0ePBq7YM6E9utZzRrZGxDd7ruH15cdx6wWbNMNA6r9EEHRVz7t/CUwIA0bsAer3A0Qt8OcHwMH5QDn1UgxqqetNSsipCDysdc1iDz3oCYIglUJYf+JuiRMq7ySmYeCKE/jir3BkqrVo61MFez5sj6GBNdG+dlV0qeuMHg1cEdTEHa/7e+DNFjUwrHVNvNuuFn54synkMgG7L8WVqJ7VTwcjkJCaiRqVrTG0dfFqDwmCgNf8PbBvYgcMbK6rr7XpdBQ6LziILWcNaxt9+fcVXI1LRRVbBRb0b/zMyQC1qtpK+R1f/BVe6lo45SEjW4O1x+6gwzf7MXzNaXRecACbTz97ncW1x+4gNVON2s62hSYIm1q72vmXi0nLVEvr3hXWG6Wn75XaVsTsvb9zhvV6NnQpsLhtWRnTyQcWMgEHr99HaM7MvKeF6RPNSzispze6ozf2f9wRU3r4lehxAZ6V0NDdAZlqLTYW8Xv654UYqWf8uwFNil3O4mkf5vQwbz8fjZsJ+T/cf/z3JiIfPIGLvQofdatT4Dk8nWyk38k5f4cjoYDfyUPX7+N/Z+9BEJDzz1DRsyA71K6KtcNbwFYpx/GIJAz+5aRBiYvi2nkhBt2+O4QD1+5DIZdhRm9dIdspPfxQ19XeKP+0VLVTYsUQfyx4ozHsVHKERj1C78WHsfro7Rdm4gwDqf8qmQyo0Qp4fTXQfpJu2/4vgV0fA9qyT+xr7OEgdUVbWVrgzeY1SnWevk3dYavUzc45Vsxp2FqtiLXH7qDnosM4decBrBUW+LJfA6x7p0Wx/4A2ru4o/VGf/Wd4sSqFxyVn4KdDugrI03r6PfOP39Mq2Sgw77VG+COn1sqDtCx89McFDPr5JG7df4w9l+OwLueDeEH/JqhmV7wu9XfaeqGRhwNSM9SYvu2SyYf48gZQM3deRnxKJhQWMjx8ko0pW8LQ/6fjuBqXf8kiQLcO2sqcemRjO/uWeFapsbSoWRkKuQwxyRm4dV83DPTHmSikZKhR08kaXes5F/n43g1dYWkhIDw2pcD3XpY6d1ivd8PyGdbTq+FkLc0mLGwG34WcAKthCRPN9eQWslLlIwqCgBFtawLQzcQrqHbXrfuPMXXLRQDAmE7eUnJ7aTSu7oiX6jpDK+a/FjcTUqXf/89fqVdkjteINl5o6O6AlAw1Zu68bLAvLVONaVt19QDfbl0T/sXMK21Zywkb3m0JBytLXIh6hH5Lj6HvkqPYERr9zMXpH6ZlYcxv5zB+43kkp2ejobsD/h7XFu+2q2WS3zH9P5d7JrRHO98qyMjWYtaf4Rj8y0nce2j6ulMMpP7rBAHoPAPo9S0AATj9C/C/4SWqiF68pxHwfnvdf91vt6kJB+vSJa/bKuV4tZnuj/j3e69j67l7OHT9PsJjUnA/NROap/5DiXrwBIN+OYGZOy8jPVtXpXjPhPYY3NKzxP9NvdPWCx1qV0WmWouxv51DelbRAec3e64hI1uL5jUrSbkbpdG8pq7WypQeflBZynA8Igk9vz+Mj36/AAD4v/a1SjRLTW4hw/zXG8HSQsDeK/H486JpanMVFEC5Oagwp28DhM7sium96sJaoavl1HvxEXy16wrSnhqi+PX4HSSnZ6NWVRv0LuFQsTFZKSzQImfW6+Eb96HRiliZM1PrnXa1DPJmCuJorZA+8LeH5u+VOnozESkZalS1UxZZzb2sjOnkA5mgK1EQ9tRagFqtKM2sLemMvbLQu6EbqtopEZ+SKS3crJeepcHo9eeQlqVBS6/KUo/S8/iwqy5R/8+LMVKQK4oiPtl2CdkaEV38qj2zp1RuIcPXrzWCRU6vtz4oBnR/R6IfpcOjkhU+LqRXqzCNqzti+5g2eK2ZBxQWMoRGPcIHm0LR9ut/sXjfjQIXqP/3ajy6fX8If1+MhYVMwISXfLF1dOtiz74rT26OVvh1RAt8EVQfVpYWOB6RhB7fH5aWhDIVBlKk0+I94I3VgIUCCN8BbHi91PWnCtO3qTsOT+6ESSX8Y/A0fdL5mbsPMfH3Cxi66hR6LT6M5l/uhe/0XQiYsxc9vj+EIStPovv3h3Ai4gGsLC0wO6g+NrzbstSz1WQyAQv6N0ZVOyVuJDyWigQWJOxeMrac0+U7TO9dvJlARVHIZRjV0RshH3ZApzpVkaXR4nGmGo09HEr8xxUA/FzsMSZnqZ/Pd15GUgF/UMtLUQHU/kkd8VYrT1gr5HivfS3sndgB3es7Q6PVVf7uuvAg9lyOgyiKeJKllma8jc0ZbnqRta+tz5NKxJ7LcYh6kI5K1pbFLnKr7wXacT4635DGnznDer0auBjlOnhVsZHymBY91RMTkZiGtCwNVJYy+BRRRqC8KPIUqdRXEdf7bMclXItPRRVbpW64vgyGQOu7OaBnAxeIIrBon64Y7h9n7+HUbd3fnVlB9Yv1+1/PzV76Z/OzHZeQnJ6NM3ceYG3ObNS5rzaETTFXlcjLq4oNFvRvjGPTOmNi19qoaqdEQmomFoZcR+u5/+Kj3y/gUnQyUjOyMeV/FzFizRncT9XlYm0b3RoTXqpd4jSM8iQIAoYE1sTuD9rB37MSHmeqEfGcNfieu02iqfv1zVhKSgocHByQnJwMe/sXc9pmPrcP6WpOZaUCzg2Bt/4H2L14eSe/n4nCsZuJSHychfupmUh8nIkHT7IKTPFqUbMyvnmj0TOnFhfX0ZuJeGvlSYgi8OOgpvlmSImiiIErTuDk7Qfo28QN3w9sWibPm/f8wZficOjGfYzt7Fvq/I4stRav/HgEV+NS0b2+M5YN9i/3bvt9V+LxybYwxKfoAjc3BxVGd/LBGwEeRQ59/ntVV+vn3kNdJekuftXgXc0WKw5FwNPJGvsmdijXvKCycCU2BT0XHYaVpQV8qtkiLDoZ4zv7YGIxA+GMbA2af7kXqRlqbPq/VmhVS1cnKlOtQcAXe5Gaqcbv7wcapUcKAG4mPEbX7w5CFIG/xrWVEqm3nb+HDzdfQIBnJfxvVGujtOVpiY8z0Xrev7r6daNbo1mNSvj9TBQm/+8iZAKw/t2W+SqkP49rcanosegQRBHY8G5LjP3tHB4+yca0nn54v4N3sc+Tka1Br0WHEZGYhteaeSA06iFu3U/DG/4e+OaNxmXS1iy1FrsvxWL10TsGOW52SjlSM9UQBOCdNl74uHudYlekNxWNVsTvZ6LwajP3EqdOPEtJPr8ZSJWjChlIAUDsBWD960Bagm7dviHbAafi/zEwFbVGiwdpWbj/ODMnuMqCnUqOrnWdyzxA+GbPVSzZfwt2Sjl2fdDOoJfrn8tx+L91Z6GUy/Dvxx1LHegYQ9i9ZPRbehRqrYj/a18Ln+RUSC4P/zt7D1O2XIRGKxY7gMorPUuDH/ffwIpDEcjW5P7Z+vq1hhhQypw7YxJFEc2/3CcNpyjkMhyd0hlV7ZTFPsfULRex6XQUBgRUl2oXhYTH471fz8DZXonjU7sYNYdl/Mbz2HkhBj3qu2D5EN2yVLP+vIzVR+9geJuamNmnvtHa8rRJf1zAH2fv4eVGrhjTyQd9lxxFplqLj7vVxtjOxV+HsbjGbTyPPy/EQGEhQ5ZGCz8XO/w5rm2Je3NORiRhwIoT0v2qdkrs/bBDqdMhinI+8iHWHLuDvy/GQq0V4VHJCt++0VgK0v/LSvL5/WL/C0em4doYeOcfoHIt4FEksLIrEF38xUpNRW4hQzV7Feq7OaBjnWp43d8D3eu7lMsHy4SXaqNZDUekZqoxbuN5Kak1S63F3N26Csbvtav1QgdRgC4ZeH7OB/KKQxHltor8L4cj8PEfF6DRinjd30MawivJf5FWCgtM6u6H3TlLSQBAjcrW6Ne0+Os/mpIgCGjvm9sL8mpT9xIFUUDu7L1dYbFSfSD9bL1eDV2Nngg8trMPBAEIvhyHK7G6CQGlrWhe1vSlEHZfisP/rTuDTLUWHWpXxeiOhZcOeB4fdPGFTACyNFoIAvBlv4alGhJrWctJWm8RAL4Iql8uQRQANK1RCYsGNsXRqZ3xw5tNETyhPYOoUmAgRQWr7AWM+AdwbQI8SQJ+7gIsbQ3sHA+cXw/cvwZoX7zCaMZiaSHDooFNYZ8zHXfBP7rciPUn7uJ2Yhqq2CoxsuOL34sH6NYpm9RdN7w0+69wBJfhwtCiKOKbPVelyu7vtfPCN68/e/p2UXyq2WHje63w+/uB+P39wDKt3Fze2tXODaTeLebySHm1qFkZbg4qpGaq8e/VBGRkaxASrls6pryKcBaltrMdejXQJfn/+K+uWOLlGF1A1dDd0ejtyauemz0CazlBoxUR9SAdrg4qfDegSbkFmz7VbKWgflCLGsWeXVeQKT390LFOVfxf+1ro0aD8J1E426vQp7HbM6vHU8F41ahwtlWBt/8CtrwHXN8NJFzW3c6t1e1XOQDuAYBH85ybP2BV+j8eFU31ytb4+rVGGLXhHJYfvIX6bvZS4u1H3WpXqD9Kozt6I+ZROjacjMQHm0Kx4V1lqdZWzEujFTFj+yWpns/kHnUwqoN3mdSeEQTBaLlAZemlus4I8KyEpjUc4VOt5LOgZDIBQU3dsezALWw7Hw2ZICAtSwM3BxWaFlLRvryN6+KDv8NisetSLHpcckF6tgY2CgvUKqcijSUxoq0XjkckQS4T8OOgZkUWxC0Lc/o2QPf6zujk93zV9e1VllgzvEUZtYrKG3OkylGFzZEqSGoccO90zu0MEH0OUKcbHiNYAC1HAp2mAUrTT5U1lunbwrDhZG7xPz8XO/w9vt0LP4vsaWqNFiPXn8XeKwlwtLbE/0a2LtbSHAXJVGswcfMF/B0WC1nOMMebLV78PKaK4Hp8Krp9dwiWFgJa1XLC4RuJeK+dF6b3rmeyNo1cdxbBl+PgaG2JR0+y0dKrMja/H2iy9uiJoq6GnGcVm+eqF0X/PUw2f0GYVSD1NE02EH85N7C6dwp4EKHbZ+cG9PwaqNtHV6fKzGVkaxD041Fci9fVkFn3Tgu08y1+XacXSXqWBm/+fAKhUY/gUckKW0e3LnahT720TDVGrj+LwzcSobCQ4fuBTUq8HBAVrdeiwwiPzS1Sun1Mm0LXWDSGyzHJ6L34iHTf1IEd0fNisjmVPwtLwK2Jrv7Uqz8B488Dg7foFkZOjQF+HwJsHAg8vGvqlpY7laUFfhzUFNXslHi1qXuFDaIAXUL3ymEBqOlkjXsP0zFiTcnW6XqYloVBv5zE4RuJsFZYYNXbzRlElQN9TSkA8KhkhcYmTuyu7+ZgUJ29oYej6RpDZGQMpKjs+L4EjD4BtPsYkFkC14OBJS2BI9/perDMmK+zHU5NfwkLBzQxdVOem5OtEmtHtICTjQKXolMwZsO5ApfayEurFXE1LgX9fzqOC1GP4Ghtid/ea4W2vmVXq4dyvdLEDfqR496NXF+IhZrH5ykpYOrAjsiYOLRXjsx6aO9Z7l8D/poI3M3p7q9aF3j5O8DT9HkTVDyhUY/w5ooTSM/W4A1/D8x/vREEQUBGtgbX41NxOSYF4TEpCI9NwZXYFDzJWTLHxV6F9e+2KFUyNRXfh5tDsedyHP4c1xbeJqggXpB1J+4iM1vzzEWYiV50zJF6QfynAykAEEXgwkbgnxm6EgoA0HQIEDACcG4AyEsxg0arARLCdXlZmixdoVAnX8DBA5C92FV4K6J9V3TFHrUiEFjLCUlpmbh1Py3fmoYAoJTL0LxmZcx7rSE8KpVuGR4qPo1WRLZG+8JXnyaqiBhIvSD+84GU3pMHwN6ZwLlfc7dZKACXRoC7P+ARoPtauVb+5PT0h7qgKeokEHVKN1swKzX/c1godY+v4gM45bnZueiGFTVZgDqzgK+ZgDoLyH4CqDN0X7MzgOx03azE7HTdNk02YFMFcKgBOFbXBW4O1QF7N12+mBn77WQkPtkWZrCtso0C9d3sUc/VHvXc7FHfzR41nWxerGVa1Jm62aYAYOdausCdiP6TGEi9IBhIPSXyBHB4oW6GX/rD/PtVjoB7M10R0LQEIOo0kHgt/3EKO13NKksb4MEt3WxBTVZ5t75ggkw3S9HBQxdgWTsBCltAYaMrAaGwBZQ59xV2uu8FWU7Alp57MwjingBatS7YtFDoAjW5Ms/9PNsBADm/wqL41Pd52iizAGRyXYkKmUXONnnO9zn3BQGAkBvM5rn/T3gCYpPT4VvNBr5VbVHFVgHh6efN91VbyL68108fOAuG3+tfl/4cBufS5t6ePABSY3UBU2pcnu9jgfQHeZ8IsK0G2LsDDu66r9L3HrogWU9q49PXsoD7Be0TNbqeU1Gb81Wj+3lK32uBrMdAxiPdwuDpOV+fvg/o2mXtpLvZVAGsq+T53gmwtAYyU546x6M895N17yeFra7um5Wj7qsqz1crR0BpD1g8o+7Z0z+7wq6LOkP3vJkpOW1IyXP/ke6+Jkv3+yHd7HU3lX3uNkubPL8XT4CsJ0B2mu73Rfo+Q/d7YGkNWFrl/6qw1n0vk+veX4JM914QZLnveWmbUMR7Vsx9H2qycv4Ry8r9XtqWqfs5yxWAXJXze6vUfZWrcrfJcyra698nYt73izb3++Iy+N3N+xW6rxB1bVNnFP5Vk61rl/76yVV5rmfOTW4FyJ7xz5J0nbJz3vfZgEb/NVv3urTZumOAQtqd56v0s5Ll/i17+udn6wLYl+2klgoVSC1duhTffPMNYmNjUb9+fXz//fdo165docdnZmZi9uzZWL9+PeLi4uDh4YHp06djxIgRAIDs7GzMnTsXa9euRXR0NOrUqYOvv/4aPXr0kM6xbNkyLFu2DHfu3AEA1K9fH5999hl69uwpHSOKImbNmoUVK1bg4cOHaNmyJZYsWYL69Yu/dhQDqUKIIvDwNnDvrG7pmeizuvX9NJkFH1/ZG6jeEqjeHPBoAVSraziMp9XolrJJugUk3cy53dDdT7uf84dMUcjXnADl6T8WT9+3kAOPE4BHUUBypO5rSrTpAjgqHgslAJE/JyJz1nYi8NLMMj1lST6/TVp6efPmzZgwYQKWLl2KNm3a4KeffkLPnj0RHh6OGjUKLt7Xv39/xMfHY+XKlfDx8UFCQgLU6tzp2TNmzMD69evx888/w8/PD3v27EG/fv1w7NgxNG3aFADg4eGBefPmwcdHt+bS2rVrERQUhPPnz0uB0vz587Fw4UKsWbMGtWvXxpw5c9C1a1dcu3YNdnZMon0ugqAbhqtcC2j0hm6bOkuX+xR9Boi9qPtvu3oLXcV0m2fM/JJZ6Ja0qeylmzloLFqtrudMH1wl39P1BGQ9BrLSgMzUPN8/1g1JZj4GID71H5/qqQDOWvdfn0ad/79dTbYu4NRk5cyEzNt7BOTv2RGf6hnJ89VgmzbnWKDQ/8YhGv73XtB/vtL9p3u4nv6KZ/fs5P2PM+/z5j23laNu2M7OVTeM+/RXfaX9J0m6n09KNJAcrfua9/snD566hsj9/unrW+T1Rm4vn0yWpwcwp0dQv62wHqK890Wtrt1PEoG0nK9PkoC0xJzvH+h6alQOhZwjZ5ulte49aNDj9Si3Bys9GchMLt6ST/muQwHXyUKhe26lfU4b7PPct9e1ycJS97uQmZpzS879PiNF9zX7Sc7viJWuR1fqZbLJ+T2x0f3uaLLz9O4+eepruq7nSqvWvbX0vT0GPZti7rZC39953tfyp3uGFYbbZBa6NhXY+5OVex946j0iy/O+yfm+OLMxC+1BQ+59QcjTK1bIV5lFTvv01y0j9zrm3SYW430ik+lmbltY5rzv5TnfW+r+KdX3jhfZc63vBXyqF9rglrNPZeKOCtGEWrRoIY4cOdJgm5+fnzh16tQCj9+9e7fo4OAgJiUlFXpOV1dX8ccffzTYFhQUJA4ePLjItlSqVEn85ZdfRFEURa1WK7q4uIjz5s2T9mdkZIgODg7i8uXLizxPXsnJySIAMTk5udiPISIiItMqyee3yTJDs7KycPbsWXTr1s1ge7du3XDs2LECH7Nz504EBARg/vz5cHd3R+3atfHxxx8jPT13qZLMzEyoVIaVmK2srHDkyJGnTwcA0Gg02LRpE9LS0hAYqJuaf/v2bcTFxRm0TalUokOHDoW2Tf/cKSkpBjciIiIyXyYb2ktMTIRGo4Gzs7PBdmdnZ8TFxRX4mIiICBw5cgQqlQrbtm1DYmIiRo8ejQcPHmDVqlUAgO7du2PhwoVo3749vL29sW/fPuzYsQMajWHiXlhYGAIDA5GRkQFbW1ts27YN9erpljTQP39Bbbt7t/BK3XPnzsWsWbNKdiGIiIiowjL5XOWnK/KKolholV6tVgtBELBhwwa0aNECvXr1kvKY9L1SixYtgq+vL/z8/KBQKDB27FgMHz4cFhaGtVbq1KmD0NBQnDhxAqNGjcKwYcMQHh5e6rYBwLRp05CcnCzdoqKiin0diIiIqOIxWSBVpUoVWFhY5Ot9SkhIyNcTpOfq6gp3d3c4OOQuP1C3bl2Iooh79+4BAKpWrYrt27cjLS0Nd+/exdWrV2FrawsvLy+DcykUCvj4+CAgIABz585F48aNsWjRIgCAi4sLAJSobYBu+M/e3t7gRkRERObLZIGUQqGAv78/QkJCDLaHhISgdevWBT6mTZs2iImJwePHj6Vt169fh0wmg4eHh8GxKpUK7u7uUKvV2LJlC4KCgopsjyiKyMzUzaTw8vKCi4uLQduysrJw8ODBQttGRERE/z0mHdqbOHEifvnlF6xatQpXrlzBhx9+iMjISIwcORKAbqhs6NCh0vGDBg2Ck5MThg8fjvDwcBw6dAiTJk3CiBEjYGVlBQA4efIktm7dioiICBw+fBg9evSAVqvF5MmTpfN88sknOHz4MO7cuYOwsDBMnz4dBw4cwODBgwHohvQmTJiAr776Ctu2bcOlS5fw9ttvw9raGoMGDTLiFSIiIqIXmUnrSA0YMABJSUmYPXs2YmNj0aBBA+zatQuenp4AgNjYWERGRkrH29raIiQkBOPGjUNAQACcnJzQv39/zJkzRzomIyMDM2bMQEREBGxtbdGrVy+sW7cOjo6O0jHx8fEYMmQIYmNj4eDggEaNGiE4OBhdu3aVjpk8eTLS09MxevRoqSDnP//8wxpSREREJDF5ZXNzxsrmREREFU9JPr9NPmuPiIiIqKJiIEVERERUSgykiIiIiEqJgRQRERFRKTGQIiIiIiolBlJEREREpWTSOlLmTl9ZIiUlxcQtISIiouLSf24Xp0IUA6lylJqaCgCoXr26iVtCREREJZWammqwvm9BWJCzHGm1WsTExMDOzg6CIJTpuVNSUlC9enVERUWx2KcR8HobF6+3cfF6Gxevt3GV5nqLoojU1FS4ublBJis6C4o9UuWooMWUy5q9vT1/EY2I19u4eL2Ni9fbuHi9jauk1/tZPVF6TDYnIiIiKiUGUkRERESlxECqglIqlZg5cyaUSqWpm/KfwOttXLzexsXrbVy83sZV3tebyeZEREREpcQeKSIiIqJSYiBFREREVEoMpIiIiIhKiYEUERERUSkxkKqAli5dCi8vL6hUKvj7++Pw4cOmbpJZOHToEPr06QM3NzcIgoDt27cb7BdFEZ9//jnc3NxgZWWFjh074vLly6ZprBmYO3cumjdvDjs7O1SrVg19+/bFtWvXDI7hNS87y5YtQ6NGjaSihIGBgdi9e7e0n9e6/MydOxeCIGDChAnSNl7vsvX5559DEASDm4uLi7S/PK83A6kKZvPmzZgwYQKmT5+O8+fPo127dujZsyciIyNN3bQKLy0tDY0bN8aPP/5Y4P758+dj4cKF+PHHH3H69Gm4uLiga9eu0pqKVDIHDx7EmDFjcOLECYSEhECtVqNbt25IS0uTjuE1LzseHh6YN28ezpw5gzNnzqBz584ICgqSPkx4rcvH6dOnsWLFCjRq1MhgO6932atfvz5iY2OlW1hYmLSvXK+3SBVKixYtxJEjRxps8/PzE6dOnWqiFpknAOK2bduk+1qtVnRxcRHnzZsnbcvIyBAdHBzE5cuXm6CF5ichIUEEIB48eFAURV5zY6hUqZL4yy+/8FqXk9TUVNHX11cMCQkRO3ToIH7wwQeiKPK9XR5mzpwpNm7cuMB95X292SNVgWRlZeHs2bPo1q2bwfZu3brh2LFjJmrVf8Pt27cRFxdncO2VSiU6dOjAa19GkpOTAQCVK1cGwGtenjQaDTZt2oS0tDQEBgbyWpeTMWPGoHfv3njppZcMtvN6l48bN27Azc0NXl5eGDhwICIiIgCU//XmosUVSGJiIjQaDZydnQ22Ozs7Iy4uzkSt+m/QX9+Crv3du3dN0SSzIooiJk6ciLZt26JBgwYAeM3LQ1hYGAIDA5GRkQFbW1ts27YN9erVkz5MeK3LzqZNm3Du3DmcPn063z6+t8tey5Yt8euvv6J27dqIj4/HnDlz0Lp1a1y+fLncrzcDqQpIEASD+6Io5ttG5YPXvnyMHTsWFy9exJEjR/Lt4zUvO3Xq1EFoaCgePXqELVu2YNiwYTh48KC0n9e6bERFReGDDz7AP//8A5VKVehxvN5lp2fPntL3DRs2RGBgILy9vbF27Vq0atUKQPldbw7tVSBVqlSBhYVFvt6nhISEfJE2lS397A9e+7I3btw47Ny5E/v374eHh4e0nde87CkUCvj4+CAgIABz585F48aNsWjRIl7rMnb27FkkJCTA398fcrkccrkcBw8exOLFiyGXy6VryutdfmxsbNCwYUPcuHGj3N/fDKQqEIVCAX9/f4SEhBhsDwkJQevWrU3Uqv8GLy8vuLi4GFz7rKwsHDx4kNe+lERRxNixY7F161b8+++/8PLyMtjPa17+RFFEZmYmr3UZ69KlC8LCwhAaGirdAgICMHjwYISGhqJWrVq83uUsMzMTV65cgaura/m/v587XZ2MatOmTaKlpaW4cuVKMTw8XJwwYYJoY2Mj3rlzx9RNq/BSU1PF8+fPi+fPnxcBiAsXLhTPnz8v3r17VxRFUZw3b57o4OAgbt26VQwLCxPffPNN0dXVVUxJSTFxyyumUaNGiQ4ODuKBAwfE2NhY6fbkyRPpGF7zsjNt2jTx0KFD4u3bt8WLFy+Kn3zyiSiTycR//vlHFEVe6/KWd9aeKPJ6l7WPPvpIPHDggBgRESGeOHFCfPnll0U7Ozvps7E8rzcDqQpoyZIloqenp6hQKMRmzZpJ08Xp+ezfv18EkO82bNgwURR1U2hnzpwpuri4iEqlUmzfvr0YFhZm2kZXYAVdawDi6tWrpWN4zcvOiBEjpL8bVatWFbt06SIFUaLIa13eng6keL3L1oABA0RXV1fR0tJSdHNzE1999VXx8uXL0v7yvN6CKIri8/drEREREf33MEeKiIiIqJQYSBERERGVEgMpIiIiolJiIEVERERUSgykiIiIiEqJgRQRERFRKTGQIiIiIiolBlJEREREpcRAioj+M95++20IgpDv1qNHD1M3jYgqKLmpG0BEZEw9evTA6tWrDbYplUoTtYaIKjr2SBHRf4pSqYSLi4vBrVKlSgAAQRCwbNky9OzZE1ZWVvDy8sIff/xh8PiwsDB07twZVlZWcHJywv/93//h8ePHBsesWrUK9evXh1KphKurK8aOHSvtW7hwIRo2bAgbGxtUr14do0ePzvd4Iqo4GEgREeXx6aef4rXXXsOFCxfw1ltv4c0338SVK1cAAE+ePEGPHj1QqVIlnD59Gn/88Qf27t1rECgtW7YMY8aMwf/93/8hLCwMO3fuhI+Pj7RfJpNh8eLFuHTpEtauXYt///0XkydPNvrrJKKywUWLieg/4+2338b69euhUqkMtk+ZMgWffvopBEHAyJEjsWzZMmlfq1at0KxZMyxduhQ///wzpkyZgqioKNjY2AAAdu3ahT59+iAmJgbOzs5wd3fH8OHDMWfOnGK16Y8//sCoUaOQmJhYdi+UiIyGOVJE9J/SqVMng0AJACpXrix9HxgYaLAvMDAQoaGhAIArV66gcePGUhAFAG3atIFWq8W1a9cgCAJiYmLQpUuXQp9///79+OqrrxAeHo6UlBSo1WpkZGQgLS3N4LxEVDEwkCKi/xQbGxuDobbiEAQBACCKovR9QcdYWVkVeZ67d++iV69eGDlyJL744gtUrlwZR44cwTvvvIPs7OwStYmIXgzMkSIiyuPEiRP57vv5+QEA6tWrh9DQUKSlpUn7jx49CplMhtq1a8POzg41a9bEvn37Cjz3mTNnoFarsWDBArRq1Qq1a9dGTExM+b0YIip37JEiov+UzMxMxMXFGWyTy+WoUqUKAF3OUkBAANq2bYsNGzbg1KlTWLlyJQBg8ODBmDlzJoYNG4bPP/8c9+/fx7hx4zBkyBA4OzsDAD7//HOMHDkS1apVQ8+ePZGamoqjR49i3Lhx8Pb2hlqtxg8//IA+ffrg6NGjWL58uXEvABGVLZGI6D9i2LBhIoB8tzp16oiiKIoAxCVLlohdu3YVlUql6OnpKW7cuNHgHBcvXhQ7deokqlQqsXLlyuJ7770npqamGhyzfPlysU6dOqKlpaXo6uoqjhs3Ttq3cOFC0dXVVbSyshK7d+8u/vrrryIA8eHDh+X++omo7HHWHhFRDkEQsG3bNvTt29fUTSGiCoI5UkRERESlxECKiIiIqJSYbE5ElIOZDkRUUuyRIiIiIiolBlJEREREpcRAioiIiKiUGEgRERERlRIDKSIiIqJSYiBFREREVEoMpIiIiIhKiYEUERERUSkxkCIiIiIqpf8HJjcKw4ylo08AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5088\n",
      "--> Acuracia (validacao): 0.5083\n",
      "--> E_val - E_in = -0.0000\n",
      "--> acc_in - acc_val = 0.0005\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> Acuracia (treino): 0.5087\n",
      "--> Acuracia (validacao): 0.5087\n",
      "--> E_val - E_in = -0.0001\n",
      "--> acc_in - acc_val = 0.0005\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_resampled, y_resampled, test_size=0.2, random_state=RANDOM_STATE)\n",
    "\n",
    "skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=RANDOM_STATE)\n",
    "\n",
    "accuracies_train = []\n",
    "accuracies_val = []\n",
    "history_list = []\n",
    "\n",
    "i = 0\n",
    "for train_index, val_index in skf.split(X_train, y_train):\n",
    "    i = i + 1\n",
    "    X_train_fold, X_val_fold = X_train.iloc[train_index], X_train.iloc[val_index]\n",
    "    y_train_fold, y_val_fold = y_train.iloc[train_index], y_train.iloc[val_index]\n",
    "\n",
    "    model = Sequential()\n",
    "    model.add(Dense(10, input_dim=9, kernel_initializer='normal', kernel_regularizer=regularizers.l2(0.01), activation='relu'))\n",
    "    model.add(Dense(5, kernel_initializer='normal', kernel_regularizer=regularizers.l2(0.01), activation='relu'))\n",
    "    model.add(Dense(1, kernel_initializer='normal', activation='sigmoid'))\n",
    "\n",
    "\n",
    "    model.compile(loss='binary_crossentropy', optimizer=\"adam\", metrics=['accuracy'])\n",
    "\n",
    "    history = model.fit(X_train_fold, y_train_fold, validation_data=(X_val_fold, y_val_fold), epochs=50, batch_size=BATCH_SIZE, verbose=0)\n",
    "    history_list.append(history)\n",
    "\n",
    "\n",
    "    _, acc_train = model.evaluate(X_train_fold, y_train_fold, verbose=0)\n",
    "    _, acc_val = model.evaluate(X_val_fold, y_val_fold, verbose=0)\n",
    "    accuracies_train.append(acc_train)\n",
    "    accuracies_val.append(acc_val)\n",
    "\n",
    "    plt.plot(history.history['loss'])\n",
    "    plt.plot(history.history['val_loss'])\n",
    "    plt.title(f'Metrica de erro - Fold {i}')\n",
    "    plt.ylabel('Erro')\n",
    "    plt.xlabel('Epoca')\n",
    "    plt.legend(['Treinamento'])\n",
    "    plt.show()\n",
    "\n",
    "    print(f'--> Acuracia (treino): {acc_train:.4f}')\n",
    "    print(f'--> Acuracia (validacao): {acc_val:.4f}')\n",
    "    print(f\"--> E_val - E_in = {(history.history['val_loss'][-1] - history.history['loss'][-1]):.4f}\")\n",
    "    print(f\"--> acc_in - acc_val = {acc_train - acc_val:.4f}\\n\")    \n",
    "avg_accuracy_train = np.mean(accuracies_train)\n",
    "avg_accuracy_val = np.mean(accuracies_val)\n",
    "\n",
    "# Historico com valores medios dos folds\n",
    "history_loss_avg = []\n",
    "history_val_loss_avg = []\n",
    "aux_list1 = []\n",
    "aux_list2 = []\n",
    "\n",
    "for i in range(len(history.history['loss'])):\n",
    "    for j in range(len(history_list)):\n",
    "        aux_list1.append(history_list[j].history['loss'][i])\n",
    "        aux_list2.append(history_list[j].history['val_loss'][i])\n",
    "    history_loss_avg.append(np.mean(aux_list1))\n",
    "    history_val_loss_avg.append(np.mean(aux_list2))                            \n",
    "                   \n",
    "plt.plot(history_loss_avg)\n",
    "plt.plot(history_val_loss_avg)\n",
    "plt.title('Metrica de erro - Media dos Folds')\n",
    "plt.ylabel('Erro')\n",
    "plt.xlabel('Epoca')\n",
    "plt.legend(['Treinamento'])\n",
    "plt.show()\n",
    "\n",
    "print(f'--> Acuracia (treino): {avg_accuracy_train:.4f}')\n",
    "print(f'--> Acuracia (validacao): {avg_accuracy_val:.4f}')\n",
    "print(f\"--> E_val - E_in = {(history_val_loss_avg[-1] - history_loss_avg[-1]):.4f}\")\n",
    "print(f\"--> acc_in - acc_val = {acc_train - acc_val:.4f}\\n\")    \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acuracia (teste): 0.4653\n"
     ]
    }
   ],
   "source": [
    "_, accuracy_test = model.evaluate(X_test, y_test, batch_size=BATCH_SIZE, verbose=0)\n",
    "print(f'Acuracia (teste): {accuracy_test:.4f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conclusão: \n",
    "\n",
    "O modelo está generalizando bem, porém apresenta uma acurácia baixa, o que indica que ele não está capturando completamente o padrão dos dados. Talvez por ter poucos dados, complexidade inadequada do modelo ou as características dos dados que precisam ser ajustadas.\n",
    "\n",
    "Ao aplicar o método de validação cruzada k-fold para testar a rede neural, esperava-se que essa técnica ajudasse a melhorar a robustez do modelo e os resultados gerais. No entanto, os resultados obtidos foram inferiores aos testes anteriores, sugerindo que o k-fold pode não ter sido a melhor abordagem neste contexto"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}