ann_clasif_binaria_detect_corazon.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importación de módulos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Creación del dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lectura de datos históricos\n",
    "df = pd.read_excel('./datasets/Heart_diagnosis.xlsx', sheet_name='heart')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**INFORMACIÓN DEL DATASET**\n",
    "\n",
    "https://www.kaggle.com/ronitf/heart-disease-uci\n",
    "1. age\n",
    "2. sex (0=female, 1=male)\n",
    "3. chest pain type (4 values)\n",
    "4. resting blood pressure\n",
    "5. serum cholestoral in mg/dl\n",
    "6. fasting blood sugar > 120 mg/dl\n",
    "7. resting electrocardiographic results (values 0,1,2)\n",
    "8. maximum heart rate achieved\n",
    "9. exercise induced angina\n",
    "10. oldpeak = ST depression induced by exercise relative to rest\n",
    "11. the slope of the peak exercise ST segment\n",
    "12. number of major vessels (0-3) colored by flourosopy\n",
    "13. thal: 3 = normal; 6 = fixed defect; 7 = reversable defect\n",
    "\n",
    "Variable salida: \"target\" => \"0\"=\"No hay problema de corazón\" / \"1\" = \"Sí hay problema de corazón\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>fbs</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>145</td>\n",
       "      <td>233</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>250</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>204</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>236</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>178</td>\n",
       "      <td>0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>354</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
       "0   63    1   3       145   233    1        0      150      0      2.3      0   \n",
       "1   37    1   2       130   250    0        1      187      0      3.5      0   \n",
       "2   41    0   1       130   204    0        0      172      0      1.4      2   \n",
       "3   56    1   1       120   236    0        1      178      0      0.8      2   \n",
       "4   57    0   0       120   354    0        1      163      1      0.6      2   \n",
       "\n",
       "   ca  thal  target  \n",
       "0   0     1       1  \n",
       "1   0     2       1  \n",
       "2   0     2       1  \n",
       "3   0     2       1  \n",
       "4   0     2       1  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 303 entries, 0 to 302\n",
      "Data columns (total 14 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   age       303 non-null    int64  \n",
      " 1   sex       303 non-null    int64  \n",
      " 2   cp        303 non-null    int64  \n",
      " 3   trestbps  303 non-null    int64  \n",
      " 4   chol      303 non-null    int64  \n",
      " 5   fbs       303 non-null    int64  \n",
      " 6   restecg   303 non-null    int64  \n",
      " 7   thalach   303 non-null    int64  \n",
      " 8   exang     303 non-null    int64  \n",
      " 9   oldpeak   303 non-null    float64\n",
      " 10  slope     303 non-null    int64  \n",
      " 11  ca        303 non-null    int64  \n",
      " 12  thal      303 non-null    int64  \n",
      " 13  target    303 non-null    int64  \n",
      "dtypes: float64(1), int64(13)\n",
      "memory usage: 33.3 KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age         0\n",
       "sex         0\n",
       "cp          0\n",
       "trestbps    0\n",
       "chol        0\n",
       "fbs         0\n",
       "restecg     0\n",
       "thalach     0\n",
       "exang       0\n",
       "oldpeak     0\n",
       "slope       0\n",
       "ca          0\n",
       "thal        0\n",
       "target      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verificamos si el dataframe contiene valores nulos\n",
    "df.isnull().sum()"
   ]
  },
  {
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>303.0</td>\n",
       "      <td>54.366337</td>\n",
       "      <td>9.082101</td>\n",
       "      <td>29.0</td>\n",
       "      <td>47.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>77.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sex</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.683168</td>\n",
       "      <td>0.466011</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cp</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.966997</td>\n",
       "      <td>1.032052</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trestbps</th>\n",
       "      <td>303.0</td>\n",
       "      <td>131.623762</td>\n",
       "      <td>17.538143</td>\n",
       "      <td>94.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chol</th>\n",
       "      <td>303.0</td>\n",
       "      <td>246.264026</td>\n",
       "      <td>51.830751</td>\n",
       "      <td>126.0</td>\n",
       "      <td>211.0</td>\n",
       "      <td>240.0</td>\n",
       "      <td>274.5</td>\n",
       "      <td>564.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fbs</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.148515</td>\n",
       "      <td>0.356198</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>restecg</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.528053</td>\n",
       "      <td>0.525860</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thalach</th>\n",
       "      <td>303.0</td>\n",
       "      <td>149.646865</td>\n",
       "      <td>22.905161</td>\n",
       "      <td>71.0</td>\n",
       "      <td>133.5</td>\n",
       "      <td>153.0</td>\n",
       "      <td>166.0</td>\n",
       "      <td>202.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>exang</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.326733</td>\n",
       "      <td>0.469794</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>oldpeak</th>\n",
       "      <td>303.0</td>\n",
       "      <td>1.039604</td>\n",
       "      <td>1.161075</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>1.6</td>\n",
       "      <td>6.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>slope</th>\n",
       "      <td>303.0</td>\n",
       "      <td>1.399340</td>\n",
       "      <td>0.616226</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ca</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.729373</td>\n",
       "      <td>1.022606</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thal</th>\n",
       "      <td>303.0</td>\n",
       "      <td>2.313531</td>\n",
       "      <td>0.612277</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>303.0</td>\n",
       "      <td>0.544554</td>\n",
       "      <td>0.498835</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          count        mean        std    min    25%    50%    75%    max\n",
       "age       303.0   54.366337   9.082101   29.0   47.5   55.0   61.0   77.0\n",
       "sex       303.0    0.683168   0.466011    0.0    0.0    1.0    1.0    1.0\n",
       "cp        303.0    0.966997   1.032052    0.0    0.0    1.0    2.0    3.0\n",
       "trestbps  303.0  131.623762  17.538143   94.0  120.0  130.0  140.0  200.0\n",
       "chol      303.0  246.264026  51.830751  126.0  211.0  240.0  274.5  564.0\n",
       "fbs       303.0    0.148515   0.356198    0.0    0.0    0.0    0.0    1.0\n",
       "restecg   303.0    0.528053   0.525860    0.0    0.0    1.0    1.0    2.0\n",
       "thalach   303.0  149.646865  22.905161   71.0  133.5  153.0  166.0  202.0\n",
       "exang     303.0    0.326733   0.469794    0.0    0.0    0.0    1.0    1.0\n",
       "oldpeak   303.0    1.039604   1.161075    0.0    0.0    0.8    1.6    6.2\n",
       "slope     303.0    1.399340   0.616226    0.0    1.0    1.0    2.0    2.0\n",
       "ca        303.0    0.729373   1.022606    0.0    0.0    0.0    1.0    4.0\n",
       "thal      303.0    2.313531   0.612277    0.0    2.0    2.0    3.0    3.0\n",
       "target    303.0    0.544554   0.498835    0.0    0.0    1.0    1.0    1.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Vemos los datos estadísticos del dataframe\n",
    "df.describe().transpose()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualización"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='target', ylabel='count'>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Verificamos que haya una proporción equilibrada entre 0 y 1 en la variable objetivo para obtener un buen modelado posterior\n",
    "sns.countplot(x='target', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "exang      -0.436757\n",
       "oldpeak    -0.430696\n",
       "ca         -0.391724\n",
       "thal       -0.344029\n",
       "sex        -0.280937\n",
       "age        -0.225439\n",
       "trestbps   -0.144931\n",
       "chol       -0.085239\n",
       "fbs        -0.028046\n",
       "restecg     0.137230\n",
       "slope       0.345877\n",
       "thalach     0.421741\n",
       "cp          0.433798\n",
       "target      1.000000\n",
       "Name: target, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualizamos la correlación entre las variables de entrada y la variable objetivo\n",
    "df.corr()['target'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Podemos generar un plot de barras con la correlación entre las variables de entrada y la variable objetivo\n",
    "df.corr()['target'][:-1].sort_values().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Podemos generar un mapa de calor con la correlación entre las variables\n",
    "plt.figure(figsize=(12,8))\n",
    "sns.set()\n",
    "sns.heatmap(df.corr(), annot=True, annot_kws={'size': 8}, cmap='PiYG')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x1c82eb07ee0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizamos las variables altamente correlacionadas con la variable objetivo\n",
    "\n",
    "# Chest Pain (Si los valores están normalizados, lo mejor es utilizar un histograma 2D)\n",
    "plt.hist2d(df['cp'], df['target'], bins=(4, 2), cmap=plt.cm.viridis)\n",
    "plt.xlabel('Chest Pain Type')\n",
    "plt.xticks(range(0, 4, 1))\n",
    "plt.ylabel('Target')\n",
    "plt.yticks(range(0, 2, 1))\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x1c82f1ad7e0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh0AAAG6CAYAAACskfQsAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAuSklEQVR4nO3dfVRVdb7H8c8ROCCiVCggrTEVQ9JUNFGc0mXU8npnqolIZyrwZlqYFj2ZlTijaUxNQVqWT6GRqfkwuJwmm7kl9965txkl9Y42VyQfUtMQGB/KSB4Ezv2DgTqpjbj37xzYvl9r7bXkt/c5+4v94afv77d/2+XxeDwCAAAwrJ2/CwAAAJcGQgcAAPAJQgcAAPAJQgcAAPAJQgcAAPAJQgcAAPAJQgcAAPAJQgcAAPAJQgcAAPCJVhU6Fi9erPT0dH+XAQAADGg1oWPlypWaN2+ev8sAAACGBPq7gPLycs2cOVNFRUXq3r27v8sBAACG+L3TsWvXLgUFBendd9/VgAED/F0OAAAwxO+djuTkZCUnJ/u7DAAAYJjfQ4fd6hrqVVr5tb/LAFqd9sG1/i4BaFWucEcowBVg/D4eT51Uf9TalwR0lcvV9v/Jbvu/wfeUVn6tEe/k+bsMoNUZPXinv0sAWpU51+aqS3Ck+RvVl6r+mLWOfkDn/5ACu9lUkP/4fU0HAAC4NDiu0wEAQGtT72mw9Hnzk0C+QegAAMAgj6QGeSx/h8uWavyL0AEAgGENstbpcIpWFTpeeOEFf5cAAAAMaVWhAwAAJ6r3WJtecQpCBwAABnnksWFNhzNCC4/MAgAAn6DTAQCAYfUO6VRYRacDAADDGv4xxXKxh50WL16s9PT0s8YPHTqkhIQEHTlyxGu8pqZGzz77rIYNG6aBAwfqiSee0IkTJy7q3oQOAAAuEStXrtS8efPOGt+/f7/uu+8+VVVVnXVu1qxZ+uijjzR//ny99dZb+uyzz5SZmXlR9yd0AABgkEeNT69YOaz2OsrLyzVp0iTl5OSoe/fuXucWL16sO++8U+Hh4ef83IYNGzRjxgwNHjxY/fv318svv6ytW7fqr3/9a4vrIHQAAGBYg8XDql27dikoKEjvvvuuBgwY4HVu06ZNev755/XUU0+d9bnt27dLkpKSkprHevTooaioKG3durXFdbCQFACANqC0tPScazGaFBYWnvdccnKykpPP/abbdevWSZKKiorOOldeXq7LL79cwcHBXuORkZEqKyu7kLK9EDoAADCsrT69UlVVJbfbfdZ4cHCwampqWvx9hA4AAAxqXNNh/TtiYmJ+sJthQkhIiGpra88ar6mpUfv27Vv8fYQOAAAMa6uve4uOjtaXX36p2tpar45HRUWFoqKiWvx9LCQFAADndN1116mhoaF5QakkHThwQOXl5UpMTGzx9xE6AAAwrF4uS4e/REVF6ac//almzJihoqIiffLJJ3r88cc1ZMgQJSQktPj7mF4BAMAgj6QGG9Z0+MucOXP061//Wg899JAkacSIEZoxY8ZFfZfL43HW+3Y/P/WlRryT5+8ygFZn9OCd/i4BaFXmXJurLsGRxu9TW3dIB44Os/QdPbpuljvwKpsq8h86HQAAGObPKZLWhNABAIBhhI5GLCQFAAA+QacDAACDGheSWut0OGXxJaEDAACj7Hjs1RnTM0yvAAAAn6DTAQCAQR5J9Rb/H5/pFQAA8M95rK/pcErqIHQAAGAYj8w2Yk0HAADwCTodAAAY5JFU72FNh0ToAADAMJcaLE8sOGN6hukVAADgE3Q6AAAwjIWkjQgdAAAYxJqObzG9AgAAfIJOBwAAhjUwvSKJ0AEAgHFWt0F3Cv4WAACAT9DpAADAII9cNiwkdcb0DKEDAADDrG8O5gyEDgAADKu3+pZZhyB6AQAAn6DTAQCAQR5Zf3rFKZuDEToAADDKpQaLC0l54RsAAEAL0OkAAMAwNgdrROgAAMCgxhe+WZseccqaDqIXAADwCUIHAACGNaidpcNOixcvVnp6utfY7t27lZaWpoSEBCUnJ2v58uXe9Tc06NVXX9Xw4cOVkJCg+++/X4cPH27xvQkdAACY5GncBt3KIZs2F1u5cqXmzZvnNXby5EmNHz9e3bp1U0FBgaZMmaKcnBwVFBQ0X7NgwQKtWrVKc+bM0erVq9XQ0KCJEyeqtra2RfdnTQcAAA5XXl6umTNnqqioSN27d/c6t3btWgUFBWn27NkKDAxUbGysDh06pCVLlig1NVW1tbVatmyZpk6dqpEjR0qS5s6dq+HDh+uDDz7QLbfccsF10OkAAMAgj6QGuSwdVheS7tq1S0FBQXr33Xc1YMAAr3Pbtm3TkCFDFBj4bR8iKSlJBw8e1LFjx1RSUqJvvvlGw4YNaz7fqVMn9enTR1u3bm1RHXQ6AAAwzOpbZiWptLT0rLUY31VYWHjec8nJyUpOTj7nubKyMsXFxXmNRUZGSpKOHj2qsrIySVLXrl3Puqbp3IUidAAAYFBr3wa9urpabrfbayw4OFiSVFNTo6qqKkk65zVfffVVi+5F6AAAoA2IiYn5wW7GxQoJCTlrQWhNTY0kKTQ0VCEhIZKk2tra5j83XdO+ffsW3Ys1HQAAGNbgcVk6TIqOjlZFRYXXWNPPUVFRzdMq57omKiqqRfcidAAAYJRL9Wpn6TD5wrfExERt375d9fX1zWNbtmxRjx49FBERofj4eIWFhamoqKj5/KlTp1RcXKzExMQW3YvQAQDAJSw1NVWVlZXKysrSvn37tH79euXn5ysjI0NS41qOtLQ05eTkqLCwUCUlJXrssccUHR2tUaNGteherOkAAMAw66+2NyciIkJ5eXnKzs5WSkqKunTpomnTpiklJaX5mszMTNXV1WnGjBmqrq5WYmKili5dqqCgoBbdy+XxeJzyHhlJ0uenvtSId/L8XQbQ6owevNPfJQCtypxrc9UlONL4fb6sPaq8/fda+o6Jsfm6zN31n1/YyrXe6AUAAByF6RUAAAxrzdMrvkToAADAoMbNwaw9feKUdRBELwAA4BN0OgAAMMplw/SK2Q3CfIXQAQCAYXa88M0JCB0AABjW4JBOhVVELwAA4BN0OgAAMMgj69MrTnl6hdABAIBJHll/U6xDUgfTKwAAwCfodAAAYJDnH6+2t/odTkDoAADAMMvTKw7B9AoAAPAJOh0AABjWwP/jSyJ0AABgXD3TK5KYXgEAAD5CpwMAAIM8sr6Q1CHbdBA6AAAwi7fMNiF0AABgWL1DQoNVrOkAAAA+QacDAACDWNPxLUIHAACGWV/T4Qz8LQAAAJ+g0wEAgGENLCSVROgAAMAoj8f6jqQehyzqYHoFAAD4BJ0OAACMYnOwJoQOAAAMs/rIrFMQOgAAMIyFpI1Y0wEAgMNVVlZq5syZuuGGGzRkyBBNnTpVx48fbz6/efNm3XHHHRowYIBGjx6tjRs3GqmD0AEAgEFNO5JaOaw+vPLII4/oT3/6k7Kzs7Vy5UpVVVVp3Lhxqq2t1f79+5WRkaHhw4dr/fr1GjNmjKZNm6bNmzfb8et7YXoFAADD/Lkj6e7du/XRRx/pjTfe0IgRIyRJL774okaOHKmNGzfqr3/9q3r37q3HHntMkhQbG6vi4mLl5eVp2LBhttZC6AAAoA0oLS1Venr6ec8XFhaec/zgwYOSpMGDBzePdejQQVdddZU+/vhj7dy5UzfffLPXZ5KSkpSdnS2PxyOXy771KEyvAABglLWplcYnXy7+H/7IyEhJ0tGjR5vH6uvrVVZWphMnTqisrEzR0dFnfaaqqkonT5686PueC50OAAAMs+PplZiYmPN2M35Iv3791LNnT82cOVO5ubkKDw/Xq6++qpMnT+rMmTOqrq6W2+32+kzTz7W1tZbr/i46HQAAOJjb7dZrr72mqqoqjRgxQoMHD9bJkyd14403KiwsTMHBwWeFi6af27dvb2stdDoAADCo6ekVq99hRWxsrAoKCvTll18qMDBQYWFhuvPOO5WUlKSuXbuqoqLC6/qKigqFhoaqY8eOFu/sjU4HAACGWV/TcfEqKyuVlpamkpISXXbZZQoLC9ORI0dUXFys66+/XoMHD9bHH3/s9ZktW7Zo0KBBatfO3phA6AAAwMHCwsLk8XiUnZ2tvXv36m9/+5sefPBBJSUladiwYUpPT9cnn3yinJwc7d+/X8uWLdMf//hHTZw40fZaCB0AAJjksaHTYXF+5eWXX1Z4eLjuuusuZWRk6LrrrtP8+fMlSVdffbUWLFigP/3pT7r99tu1bt06vfTSS7bv0SGxpgMAAOP8/cK3qKgovfbaa+c9P2LEiOaNw0widAAAYJBH1h+ZtbqQtLVgegUAAPgEnQ4AAAzz9/RKa0HoAADAKOuPvVrZBr01YXoFAAD4BJ0OAAAMag07krYWhA4AAAxjTUcjplcAAIBP0OkAAMAwD50OSYQOAACMs7o5mFMwvQIAAHyCTgcAACZ5bFhI6pDHVwgdAAAY5JH1NR0OyRyEDgAATOOR2Uas6QAAAD5BpwMAAKNcNjwy64xOCaEDAADDmF5pxPQKAADwCTodAAAY5nHK4ycWEToAADDII+s7kjolszC9AgAAfIJOBwAAhvHCt0aEDgAADOPplUZMrwAAAJ+g0wEAgEkeG55ecchKUkIHAACGsaajEaEDAADDCB2NWNMBAAB8gk4HAAAGeeSy/PSKxyEvfKPTAQCAYR6PtcOquro6vfLKK7rxxhs1cOBA3XPPPdqxY0fz+d27dystLU0JCQlKTk7W8uXLrd/0HAgdAAA43MKFC7Vu3TrNmTNHGzZsUI8ePTRx4kRVVFTo5MmTGj9+vLp166aCggJNmTJFOTk5KigosL0OplcAADDM3wtJN23apFtuuUU33HCDJOnpp5/WunXrtGPHDh04cEBBQUGaPXu2AgMDFRsbq0OHDmnJkiVKTU21tQ5CBwAAhtkROkpLS5Wenn7e84WFhec9FxERof/8z/9UWlqaunbtqjVr1sjtdis+Pl7r1q3TkCFDFBj4bSRISkrS4sWLdezYMXXu3Nly7U0IHQAAOFxWVpYeeeQR3XTTTQoICFC7du00f/58devWTWVlZYqLi/O6PjIyUpJ09OhRQgcAAG2JHRuKxsTE/GA344fs27dPHTt21Ouvv66oqCitW7dOU6dO1YoVK1RdXS232+11fXBwsCSppqbGct3fRegAAMAwf67pOHr0qJ544gnl5+dr8ODBkqR+/fpp3759mj9/vkJCQlRbW+v1maawERoaamstPL0CAICD7dy5U2fOnFG/fv28xgcMGKBDhw4pOjpaFRUVXueafo6KirK1FkIHAAAmeWw6LlJ0dLQk6dNPP/Ua37Nnj7p3767ExERt375d9fX1zee2bNmiHj16KCIi4uJvfA6EDgAADPN4XJYOK/r376/rrrtOTz31lLZs2aKDBw9q3rx52rx5sx544AGlpqaqsrJSWVlZ2rdvn9avX6/8/HxlZGTY9Nt/izUdAAAY5JH1XUWtfLxdu3ZauHCh5s2bp2eeeUZfffWV4uLilJ+frwEDBkiS8vLylJ2drZSUFHXp0kXTpk1TSkqKtaLPgdABAIDDhYeHa+bMmZo5c+Y5z/fv319r1qwxXgehAwAAw/y9I2lrQegAAMA0QockFpICAAAfodMBAIBhdrye3gkIHQAAmEbokMT0CgAA8BE6HQAAmOSx4ekVh3RKCB0AAJjmkNBgFdMrAADAJ+h0AABglPX3p0jO2OeD0AEAgGlMr0gidAAA4APO6FRYxZoOAADgE3Q6AAAwjekVSYQOAADMI3RIYnoFAAD4iO2djvr6egUEBNj9tQAAtF282l7SRXQ6brrpJpWUlJzz3CeffKIf//jHlosCAMBJPB5rh1NcUKfjvffeU11dnSTpiy++0IcffnjO4LF582adOXPG3goBAIAjXFDo+Nvf/qa33npLkuRyufT666+f99rx48fbUxkAAE7gkfWFpA7pdlxQ6HjiiSc0btw4eTwe3XzzzXrttdd0zTXXeF0TEBCgsLAwhYWFGSkUAIA2izUdki4wdLjdbl155ZWSpMLCQnXp0kVut9toYQAAwFla/PTKlVdeqRMnTmjp0qX6y1/+or///e/Ky8vTpk2bFB8fr5tvvtlEnQAAtFkuh0yPWNXip1cOHz6s2267TWvXrlVUVJSOHz+u+vp6HThwQJmZmfqv//ovA2UCANCGeSweDtHiTsdvfvMbRURE6O2331ZoaKiuvfZaSVJubq5qamq0aNEijRw50u46AQBou1jTIekiOh2bN2/W5MmT1alTJ7lc3n+JP//5z7V3717bigMAAM5xUTuSBgae+2O1tbVnBREAAC55DpoisaLFnY7Bgwdr8eLFOn36dPOYy+VSQ0OD3nnnHQ0aNMjWAgEAaPNY0yHpIjodTzzxhO666y6NGjVKQ4cOlcvl0tKlS7V//34dOnRIq1atMlEnAABo41rc6YiLi9Nvf/tbDR06VEVFRQoICNBf/vIXdevWTatXrz5r0zAAAC5pVrscFrsdRUVF6t279zmPm266SZJ05MgRZWRkaNCgQbrhhhs0b9481dfXX/xNz+Oi1nT06NFDubm5dtcCAIAz+fHplYEDB+qjjz7yGtuxY4cefvhhTZ48WWfOnNGECRPUvXt3rV69Wp9//rmysrLUrl07ZWZm2lpLi0PH1q1bz3vO5XKpQ4cO+tGPfsR26AAAtAJut1tdunRp/vn06dN6/vnnlZKSotTUVL333nsqLS3V2rVrFR4erri4OB0/flwvvviiJk2aZOsO5C0OHenp6c1PqHi+877d7z610q5dO91+++2aPXu2AgICbCgTAIC2y44dSUtLS5Wenn7e84WFhRf0PYsWLVJVVZWeeuopSdK2bdvUt29fhYeHN1+TlJSkyspK7d69WwMGDLBW+He0eE3HwoUL5Xa7NXbsWL399tt6//33tWLFCqWlpSkwMFDPPPOMpk+frg8//FBLliyxrVAAANqsVvL0yokTJ5Sfn69JkybpsssukySVlZUpOjra67rIyEhJ0tGjR+27uS6i0/HGG2/o7rvvbk5IktSzZ08NHjxYoaGh+vDDD/X2229LkpYvX64HH3zQvmoBALhExcTEXHA343xWrVqljh076uc//3nzWHV1tTp16uR1XXBwsCSppqbG0v2+r8Wdjl27dmn48OHnPDd06FDt3LlTktS7d2/bExIAALh4GzZs0O23366QkJDmsZCQENXW1npd1xQ2QkNDbb1/i0NHly5dVFRUdM5zRUVF6ty5syTp5MmTZyUnAAAuRS6PtcMOJSUlOnz4sG699Vav8ejoaFVUVHiNNf0cFRVlz83/ocXTK3fddZdyc3NVVVWlf/mXf1FERISOHTumTZs2acWKFXr44YdVVlamhQsXaujQobYWeyGCjtXo6gfPHYqAS9mC0p3+LgFoVVwB9k4d/KBW8MK3bdu2KSIiQvHx8V7jiYmJ2rBhgyorK5ufPN2yZYs6dOhw1rVWtTh0TJgwQVVVVcrLy2teu+HxeNSxY0c9/PDDysjI0IYNG1RbW6vHH3/c1mIBAMDFKS4uVu/evc8av/nmmzVv3jw9+uijmjp1qo4cOaKXX35Z9913n62Py0oXETr279+vhx56SBMmTNCOHTt04sQJRUVF6ZprrlGHDh0kSbfeeqtuv/12WwsFAKDNagXvT/n73//e/MTKdwUHBysvL0/PPvusxo4dq/DwcN19992aPHmy7TW0OHTcfffdeuaZZ3T77bdr2LBh57yGvTkAAPiOVhA63njjjfOeu+qqq7Rs2TLjNbR4IWlQUJAuv/xyE7UAAAAHa3Gn45FHHtGLL76or7/+WvHx8ed8nCYmJsaW4gAAaPPseAKlFXRK7NDi0DFr1izV19frySefPO81u3fvtlQUAACO4pDQYFWLQ8dzzz1nog4AAOBwLQ4dKSkpJuoAAMC56HRIuojQIUnl5eXavn2717apDQ0Nqqqq0rZt2zR37lzbCgQAoK2za1fRtq7FoeOPf/yjpk6dqrq6Oq9X3Df9uWfPnvZWCAAAHKHFj8wuWrRIffv21fr163XHHXfoZz/7mTZu3Kgnn3xSAQEBmj59uok6AQBoo1yN26BbOeT/bdTt0OJOx4EDB5Sbm6s+ffpo6NChWrZsmWJjYxUbG6tjx45p0aJFuv76603UCgBA28T0iqSL6HS0a9dO4eHhkhp3MPvss8/U0NAgSRoxYoT27dtnb4UAALRhLll/y6wz+hwXETp69uyp//3f/23+c21trUpKSiRJp06d8lpcCgAA0OSCQse4ceO0f/9+SdIvfvELvfLKK5o7d646duyopKQkPfPMM3r77beVm5urvn37Gi0YAIA2xWPT4QAXFDo+/vhjffPNN5KkMWPGKCsrq7mjMWfOHNXU1Cg7O1t1dXXKysoyVy0AAG2Q1ekVp7iofTruueee5j//6Ec/0h/+8AedPHlSV1xxhW2FAQAAZ7mo0PF9LpeLwAEAwPk4qFthxQWHjilTpsjtdv/T61wulzZt2mSpKAAAHIXQIakFoaNPnz50MwAAwEVrUaejf//+JmsBAMCRnLQY1IoW79MBAABwMQgdAADAJy5oeiUlJUWXX3656VoAAHAmplckXWDoeP75503XAQCAY7Gmo5Et+3QAAIAfQOiQxJoOAADgI3Q6AAAwyY4XtjmkU0LoAADAMNZ0NGJ6BQAA+ASdDgAATKPTIYnQAQCAcUyvNGJ6BQCAS8CGDRv0k5/8RP369dNPf/pT/eEPf2g+d+TIEWVkZGjQoEG64YYbNG/ePNXX19teA50OAABM83On43e/+52ysrI0ffp0DR8+XBs3btTjjz+u6OhoXXvttZowYYK6d++u1atX6/PPP1dWVpbatWunzMxMW+sgdAAAYJofQ4fH49Err7yicePG6Z577pEkPfjgg9q2bZs+/vhjffHFFyotLdXatWsVHh6uuLg4HT9+XC+++KImTZokt9ttWy2EDgAA2oDS0lKlp6ef93xhYeE5xw8cOKAvvvhCt956q9f40qVLJUmzZs1S3759FR4e3nwuKSlJlZWV2r17twYMGGBD9Y1Y0wEAgGEuj7XDigMHDkiSTp8+rQkTJmjYsGEaM2aM/uM//kOSVFZWpujoaK/PREZGSpKOHj1q7ebfQ6cDAACTbNqRNCYm5rzdjB9SWVkpSXrqqaf00EMPaerUqfr3f/93TZ48WW+++aaqq6vVqVMnr88EBwdLkmpqaiwW7o3QAQCAaX5c0xEUFCRJmjBhglJSUiRJ11xzjYqLi/Xmm28qJCREtbW1Xp9pChuhoaG21sL0CgAADhYVFSVJiouL8xrv1auXjhw5oujoaFVUVHida/q56bN2IXQAAGCYP9d09O3bVx06dNDOnTu9xvfs2aNu3bopMTFRxcXFzdMwkrRlyxZ16NBB8fHx1m7+PYQOAABM81g8LAgJCdHEiRP1+uuv67333tPnn3+uhQsX6s9//rPGjx+vm2++WV26dNGjjz6qkpISbdq0SS+//LLuu+8+Wx+XlVjTAQCA402ePFnt27fX3LlzVV5ertjYWM2fP19Dhw6VJOXl5enZZ5/V2LFjFR4errvvvluTJ0+2vQ5CBwAABrlkfYrEZUMd48eP1/jx48957qqrrtKyZctsuMsPI3QAAGAaL3yTxJoOAADgI3Q6AAAwjU6HJEIHAADG2bEmwwmYXgEAAD5BpwMAANOYXpFE6AAAwCwbdhV1SmghdAAAYJpDQoNVrOkAAAA+QacDAADT6HRIInQAAGCc5TUdDsH0CgAA8Ak6HQAAmEanQxKhAwAA45heacT0CgAA8Ak6HQAAmEanQxKhAwAA45heacT0CgAA8Ak6HQAAmOSR9ekVh3RKCB0AAJjmkNBgFaEDAADDWNPRiDUdAADAJ+h0AABgGp0OSYQOAAAM88jlYSWpxPQKAADwETodAACY5oxGhWWEDgAADHLJ+tMrLlsq8T+mVwAAgE/Q6QAAwCR2JG1GpwMAAMNcHmuHVeXl5erdu/dZx/r16yVJu3fvVlpamhISEpScnKzly5dbv+k50OkAAMDhSkpKFBwcrE2bNsnl+naFSMeOHXXy5EmNHz9eycnJevbZZ7Vjxw49++yz6tChg1JTU22tg9ABAIBpfp4e2bNnj7p3767IyMizzr311lsKCgrS7NmzFRgYqNjYWB06dEhLliwhdAAA0NbYMUVSWlqq9PT0854vLCw877lPP/1UsbGx5zy3bds2DRkyRIGB30aCpKQkLV68WMeOHVPnzp0vvujvYU0HAACmeSweFu3Zs0cnTpzQPffcox//+Me666679N///d+SpLKyMkVHR3td39QROXr0qPWbfwedDgAA2oCYmJgf7GacT11dnT777DP16tVLTz/9tMLCwrRx40Y98MADevPNN1VdXS232+31meDgYElSTU2NLbU3IXQAAGCYP19tHxgYqKKiIgUEBCgkJESSdO2112rv3r1aunSpQkJCVFtb6/WZprARGhpqay1MrwAAYJrHY+2wqEOHDs2Bo8nVV1+t8vJyRUdHq6Kiwutc089RUVGW7/1dhA4AABxs7969GjRokIqKirzG/+///k+9evVSYmKitm/frvr6+uZzW7ZsUY8ePRQREWFrLYQOAAAM8+fmYLGxserZs6dmz56tbdu2af/+/Xr++ee1Y8cOPfjgg0pNTVVlZaWysrK0b98+rV+/Xvn5+crIyLDnl/8O1nQAAGCSn7dBb9eunRYtWqTc3Fw9+uijOnXqlPr06aM333xTcXFxkqS8vDxlZ2crJSVFXbp00bRp05SSkmKx6LMROgAAcLjOnTvr+eefP+/5/v37a82aNcbrIHQAAGCYq8HfFbQOhA4AAExzyFtirWIhKQAA8Ak6HQAAGObPzcFaE0IHAACm2bDBlxMQOgAAMMgl650Oly2V+B9rOgAAgE/Q6QAAwCQ/bw7WmhA6AAAwjIWkjZheAQAAPkGnAwAA03h6RRKhAwAA45heacT0CgAA8Ak6HQAAmEanQxKhAwAA45heacT0CgAA8Ak6HQAAmNZAq0MidAAAYBY7kjYjdAAAYBhrOhqxpgMAAPgEnQ4AAExjR1JJhA4AAIxjeqUR0ysAAMAn6HQAAGAanQ5JhA4AAIxzsaZDEtMrAADAR+h0AABgkkdSgw3f4QCEDgAAjPLYML3ijNTB9AoAAPAJQgcAAKZ5LB42OnDggAYOHKj169c3j+3evVtpaWlKSEhQcnKyli9fbu9N/4HQAQCAaR6PtcMmZ86c0dSpU3X69OnmsZMnT2r8+PHq1q2bCgoKNGXKFOXk5KigoMC2+zZhTQcAAAa5ZH1HUpctlUjz589XWFiY19jatWsVFBSk2bNnKzAwULGxsTp06JCWLFmi1NRUm+7ciE4HAACXgK1bt2rNmjV64YUXvMa3bdumIUOGKDDw2z5EUlKSDh48qGPHjtlaA50OAABMs2GKpLS0VOnp6ec9X1hYeN5zp06d0rRp0zRjxgx17drV61xZWZni4uK8xiIjIyVJR48eVefOnS1U7Y3QAQCAYS6r+3RYNGvWLA0cOFC33nrrWeeqq6vldru9xoKDgyVJNTU1ttZB6AAAoA2IiYn5wW7G+WzYsEHbtm3T73//+3OeDwkJUW1trddYU9gIDQ1teaE/gNABAIBJHlmfXrHw8YKCAh0/flwjR470Gp85c6bef/99RUdHq6Kiwutc089RUVEXf+NzIHQAAGCaHzcUzcnJUXV1tdfYqFGjlJmZqdtuu02/+93vtHr1atXX1ysgIECStGXLFvXo0UMRERG21sLTKwAAOFhUVJSuuuoqr0OSIiIiFBUVpdTUVFVWViorK0v79u3T+vXrlZ+fr4yMDNtrodMBAIBhrfnV9hEREcrLy1N2drZSUlLUpUsXTZs2TSkpKbbfi9ABAIBprSx0fPrpp14/9+/fX2vWrDF+X6ZXAACAT9DpAADAND/v09FaEDoAADCsNa/p8CVCBwAAJvl5n47WhDUdAADAJ+h0AABglMeGp1ec0eogdAAAYBoLSSUxvQIAAHyETgcAAIbx9EojQgcAAKYROiQxvQIAAHyETgcAAKbR6ZBE6AAAwDxChyRCBwAAZnlk/ZFZh2QW1nQAAACfoNMBAIBhPDLbiNABAIBphA5JTK8AAAAfodMBAIBRHqmBF75JhA4AAMxjekUS0ysAAMBH6HQAAGAanQ5JhA4AAMzyyHrocEhmYXoFAAD4BJ0OAABMs/z0ijMQOgAAMM1j9eUrzkDoAADANBaSSmJNBwAA8BE6HQAAGMWOpE3odAAAYFLTI7OWDmslHD9+XE8++aSSkpI0cOBAPfDAA9q/f3/z+d27dystLU0JCQlKTk7W8uXLrd3wPAgdAAA43JQpU3To0CEtWbJEv/3tbxUSEqJ7771XVVVVOnnypMaPH69u3bqpoKBAU6ZMUU5OjgoKCmyvg+kVAABM8+NC0q+++kpXXnmlMjIyFBcXJ0maPHmyfvazn2nv3r3avHmzgoKCNHv2bAUGBio2NrY5oKSmptpaC50OAABMszq9YkF4eLhyc3ObA8eJEyeUn5+v6Oho9erVS9u2bdOQIUMUGPhtHyIpKUkHDx7UsWPHLN37++h0AADQBpSWlio9Pf285wsLC//pd/zyl7/U2rVr5Xa7tXDhQoWGhqqsrKw5kDSJjIyUJB09elSdO3e2Vvh30OkAAMC0hgZrh03+7d/+TQUFBbrllls0ZcoU7dq1S9XV1XK73V7XBQcHS5Jqampsu7dEpwMAAPNsWNMRExNzQd2MH9KrVy9JUnZ2tnbu3KkVK1YoJCREtbW1Xtc1hY3Q0FBL9/s+Oh0AADjYiRMntHHjRtXV1TWPtWvXTr169VJFRYWio6NVUVHh9Zmmn6OiomythdABAIBpflxIeuzYMT3++OPavHlz89iZM2dUXFys2NhYJSYmavv27aqvr28+v2XLFvXo0UMRERGW7v19hA4AAEzy/GNHUiuHheARFxenESNG6LnnntPWrVu1Z88ePf300zp16pTuvfdepaamqrKyUllZWdq3b5/Wr1+v/Px8ZWRk2PiX0Ig1HQAAGObx81tmX375ZeXm5uqxxx7T119/rcGDB2vlypWKiYmRJOXl5Sk7O1spKSnq0qWLpk2bppSUFNvrIHQAAOBwHTt21KxZszRr1qxznu/fv7/WrFljvA5CBwAApll+4ZszEDoAADDNj9ugtyYsJAUAAD5BpwMAANNs3FW0LSN0AABgkg17bThleobpFQAA4BN0OgAAMMzD9IokQgcAAOY5ZHrEKqZXAACAT9DpAADANDYHk0ToAADAPD+/e6W1IHQAAGCSR/JY7XQ4pFHCmg4AAOATfg8dDQ0NevXVVzV8+HAlJCTo/vvv1+HDh/1dFgAANvE0Tq9YORzS6vB76FiwYIFWrVqlOXPmaPXq1WpoaNDEiRNVW1vr79IAALCFp8Fj6XAKv4aO2tpaLVu2TJmZmRo5cqTi4+M1d+5clZWV6YMPPvBnaQAAwGZ+XUhaUlKib775RsOGDWse69Spk/r06aOtW7fqlltuafF3dvlRhJbve83OMgFHcHWmewh4Cejqk9tEduus5XtftfwdTuDX0FFWViZJ6trV+z98ZGRk87mWCgwKVNeeUZZrAwDADgGBAfy79A9+nV6pqqqSJLndbq/x4OBg1dTU+KMkAABgiF9DR0hIiCSdtWi0pqZG7du390dJAADAEL+GjqZplYqKCq/xiooKRUXRigIAwEn8Gjri4+MVFhamoqKi5rFTp06puLhYiYmJfqwMAADYza8LSd1ut9LS0pSTk6MrrrhCV155pV566SVFR0dr1KhR/iwNAADYzO/vXsnMzFRdXZ1mzJih6upqJSYmaunSpQoKCvJ3aQAAwEYuj8fjnK3OAABAq+X3bdABAMClgdABAAB8gtABAAB8gtABAAB8gtABAAB8gtABAAB8wjGho6GhQa+++qqGDx+uhIQE3X///Tp8+LC/ywIAtHKLFy9Wenq6v8u4JDgmdCxYsECrVq3SnDlztHr1ajU0NGjixIlnvUwOAIAmK1eu1Lx58/xdxiXDEaGjtrZWy5YtU2ZmpkaOHKn4+HjNnTtXZWVl+uCDD/xdHgCglSkvL9ekSZOUk5Oj7t27+7ucS4YjQkdJSYm++eYbDRs2rHmsU6dO6tOnj7Zu3erHygAArdGuXbsUFBSkd999VwMGDPB3OZcMv797xQ5lZWWSpK5du3qNR0ZGNp8DAKBJcnKykpOT/V3GJccRnY6qqipJjW+t/a7g4GDV1NT4oyQAAPA9jggdISEhknTWotGamhq1b9/eHyUBAIDvcUToaJpWqaio8BqvqKhQVFSUP0oCAADf44jQER8fr7CwMBUVFTWPnTp1SsXFxUpMTPRjZQAAoIkjFpK63W6lpaUpJydHV1xxha688kq99NJLio6O1qhRo/xdHgAAkENChyRlZmaqrq5OM2bMUHV1tRITE7V06VIFBQX5uzQAACDJ5fF4PP4uAgAAOJ8j1nQAAIDWj9ABAAB8gtABAAB8gtABAAB8gtABAAB8gtABAAB8gtABAAB8gtABAAB8wjE7kgKXmj179mjhwoX6+OOP9dVXX+myyy7T4MGDNWnSJMXHx/u7PAA4CzuSAm3Q3r17NXbsWCUkJGjs2LGKiIhQWVmZVqxYoZKSEi1fvlwJCQn+LhMAvBA6gDZo+vTp2rJliz744AMFBn7bsDx9+rRGjx6t+Ph4LVmyxI8VAsDZWNMBtEHHjh2Tx+NRQ0OD13hoaKimT5+uf/3Xf20e27Rpk+644w7169dP119/vZ577jmdPn1aklRZWakbb7xRo0ePVm1trSTJ4/Fo3Lhxuv7663XixAnf/VIAHI/QAbRBI0eOVGlpqX7xi19o5cqV2r9/v5qalqNHj1ZKSook6fe//72mTJminj176vXXX9dDDz2kd999V5MnT5bH41FYWJiys7N18OBBLVq0SJK0fPlyFRUV6de//rWuuOIKv/2OAJyH6RWgjXrllVe0dOlS1dTUSJIuv/xy3XDDDRo3bpz69+8vj8ejkSNH6uqrr1ZeXl7z5zZv3qx7771Xixcv1siRIyVJM2fOVEFBgV5//XVlZmYqNTVVv/rVr/zxawFwMEIH0IZ99dVX+p//+R9t3rxZRUVFOnz4sFwul6ZPn67rr79eP/nJTzRz5kyNHTvW63NDhw7VHXfcoaysLEnSN998o9tuu02lpaXq0aOH1q9fr5CQEH/8SgAcjNABOEhxcbGefPJJff7558rPz9fdd9993mtHjx6tV155pfnn3/zmN1q2bJnS0tL0y1/+0hflArjEsE8H0MaUl5crNTVVjzzyiMaMGeN1rk+fPnrsscc0ZcoU1dfXS5KmTZumIUOGnPU94eHhzX/es2eP3n77bV1zzTV65513dNttt2nAgAFmfxEAlxwWkgJtTOfOnRUYGKhVq1Y1r+f4rs8++0zBwcG6+uqrFRERoSNHjqhfv37NR1RUlHJzc1VcXCxJqqur09NPP61u3bpp9erVio+P11NPPXXO7wYAK+h0AG1MQECAZs2apSlTpig1NVX33HOPYmNjVVVVpT//+c9auXKlHnnkEV1++eV67LHH9Ktf/UoBAQG68cYbderUKS1YsEDl5eXq27evJGnRokUqLi7WqlWrFBISojlz5mjMmDGaO3eunn76aT//tgCchDUdQBu1a9cuLV26VNu3b9eJEyfkdrvVp08fpaena9SoUc3Xvf/++8rLy9PevXsVGhqqQYMG6dFHH1Xv3r1VUlKiO++8U2PGjNHMmTObP/PCCy/orbfe0ooVK3Tdddf549cD4ECEDgAA4BOs6QAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD5B6AAAAD7x//j9idfs574WAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sex\n",
    "plt.hist2d(df['sex'], df['target'], bins=(2, 2), cmap=plt.cm.viridis)\n",
    "plt.xlabel('Sex')\n",
    "plt.xticks(range(0, 2, 1))\n",
    "plt.ylabel('Target')\n",
    "plt.yticks(range(0, 2, 1))\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Limpieza Final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>145</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>0</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>178</td>\n",
       "      <td>0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  restecg  thalach  exang  oldpeak  slope  ca  thal  \\\n",
       "0   63    1   3       145        0      150      0      2.3      0   0     1   \n",
       "1   37    1   2       130        1      187      0      3.5      0   0     2   \n",
       "2   41    0   1       130        0      172      0      1.4      2   0     2   \n",
       "3   56    1   1       120        1      178      0      0.8      2   0     2   \n",
       "4   57    0   0       120        1      163      1      0.6      2   0     2   \n",
       "\n",
       "   target  \n",
       "0       1  \n",
       "1       1  \n",
       "2       1  \n",
       "3       1  \n",
       "4       1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Eliminamos las variables poco correlacionadas con la variable objetivo\n",
    "df = df.drop(['chol', 'fbs'], axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train/Test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dividimos los conjuntos X e Y\n",
    "X = df.drop('target', axis=1)\n",
    "y = df['target'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importamos train_test_split\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Realizamos el split en 75/25\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.25, random_state=101)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Escalado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Impotamos MinMaxScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "# Creamos una instancia de MinMaxScaler\n",
    "scaler = MinMaxScaler()\n",
    "\n",
    "# Adaptamos scaler a nuestro conjunto de entrenamiento\n",
    "scaler.fit(X_train)\n",
    "\n",
    "# Escalamos los conjuntos de prueba y entrenamiento\n",
    "X_train = scaler.transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Creación del modelo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para un problema de clasificación binario utilizamos como optimizador 'rmsprop' o 'adam', en el argumento loss indicamos 'binary_crossentropy', y \n",
    "como métrica utilizamos 'accuracy'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Realizamos las importaciones necesarias\n",
    "import tensorflow as tf\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Activation, Dropout\n",
    "\n",
    "# Creamos un modelo secuencial\n",
    "model = Sequential()\n",
    "\n",
    "# Iteramos una variable con un número de neuronas igual a la cantidad de variables de entrada\n",
    "num_neuronas = X_train.shape[1]\n",
    "\n",
    "# Elegimos un número de capas (https://stats.stackexchange.com/questions/181/how-to-choose-the-number-of-hidden-layers-and-nodes-in-a-feedforward-neural-netw)\n",
    "# Creamos una capa oculta con una neurona por cada variable de entrada con función de activación ReLu\n",
    "model.add(Dense(units=num_neuronas, activation='relu'))\n",
    "\n",
    "# Creamos otra capa oculta con la mitad de neuronas (suele hacerse para acelerar el entrenamiento) con función de activación ReLu\n",
    "model.add(Dense(units=np.round(num_neuronas/2), activation='relu'))\n",
    "\n",
    "# Creamos una capa de salida con una neurona por cada variable de salida con función de activación Sigmoid (recomendada para clasificación binaria)\n",
    "model.add(Dense(units=1, activation='sigmoid'))\n",
    "\n",
    "# Compilamos el modelo con optimizador adam y binary_crossentropy en las pérdidas\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics='accuracy')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Entrenamiento del modelo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Información para poder elegir el número de iteraciones**\n",
    "\n",
    "https://stats.stackexchange.com/questions/164876/tradeoff-batch-size-vs-number-of-iterations-to-train-a-neural-network\n",
    "\n",
    "https://datascience.stackexchange.com/questions/18414/are-there-any-rules-for-choosing-the-size-of-a-mini-batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "8/8 [==============================] - 1s 45ms/step - loss: 0.6940 - val_loss: 0.6824\n",
      "Epoch 2/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6910 - val_loss: 0.6802\n",
      "Epoch 3/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6882 - val_loss: 0.6773\n",
      "Epoch 4/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.6847 - val_loss: 0.6727\n",
      "Epoch 5/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6808 - val_loss: 0.6673\n",
      "Epoch 6/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6763 - val_loss: 0.6600\n",
      "Epoch 7/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6708 - val_loss: 0.6525\n",
      "Epoch 8/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6641 - val_loss: 0.6438\n",
      "Epoch 9/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6565 - val_loss: 0.6317\n",
      "Epoch 10/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6469 - val_loss: 0.6192\n",
      "Epoch 11/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6357 - val_loss: 0.6057\n",
      "Epoch 12/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6221 - val_loss: 0.5917\n",
      "Epoch 13/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6099 - val_loss: 0.5784\n",
      "Epoch 14/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.5974 - val_loss: 0.5624\n",
      "Epoch 15/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.5847 - val_loss: 0.5472\n",
      "Epoch 16/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5720 - val_loss: 0.5342\n",
      "Epoch 17/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5606 - val_loss: 0.5227\n",
      "Epoch 18/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5483 - val_loss: 0.5095\n",
      "Epoch 19/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5364 - val_loss: 0.4961\n",
      "Epoch 20/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5237 - val_loss: 0.4829\n",
      "Epoch 21/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5126 - val_loss: 0.4683\n",
      "Epoch 22/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.5012 - val_loss: 0.4550\n",
      "Epoch 23/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.4895 - val_loss: 0.4461\n",
      "Epoch 24/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4792 - val_loss: 0.4368\n",
      "Epoch 25/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4706 - val_loss: 0.4266\n",
      "Epoch 26/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4619 - val_loss: 0.4162\n",
      "Epoch 27/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4533 - val_loss: 0.4074\n",
      "Epoch 28/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4484 - val_loss: 0.3996\n",
      "Epoch 29/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4404 - val_loss: 0.3933\n",
      "Epoch 30/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4338 - val_loss: 0.3888\n",
      "Epoch 31/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4286 - val_loss: 0.3859\n",
      "Epoch 32/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4213 - val_loss: 0.3780\n",
      "Epoch 33/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4171 - val_loss: 0.3721\n",
      "Epoch 34/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.4128 - val_loss: 0.3676\n",
      "Epoch 35/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4082 - val_loss: 0.3653\n",
      "Epoch 36/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4049 - val_loss: 0.3622\n",
      "Epoch 37/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4040 - val_loss: 0.3578\n",
      "Epoch 38/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3996 - val_loss: 0.3558\n",
      "Epoch 39/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3968 - val_loss: 0.3531\n",
      "Epoch 40/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3958 - val_loss: 0.3507\n",
      "Epoch 41/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3929 - val_loss: 0.3495\n",
      "Epoch 42/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3888 - val_loss: 0.3464\n",
      "Epoch 43/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3883 - val_loss: 0.3450\n",
      "Epoch 44/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3862 - val_loss: 0.3459\n",
      "Epoch 45/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3842 - val_loss: 0.3453\n",
      "Epoch 46/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.3827 - val_loss: 0.3427\n",
      "Epoch 47/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3799 - val_loss: 0.3394\n",
      "Epoch 48/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3772 - val_loss: 0.3365\n",
      "Epoch 49/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3784 - val_loss: 0.3351\n",
      "Epoch 50/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3786 - val_loss: 0.3337\n",
      "Epoch 51/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3771 - val_loss: 0.3323\n",
      "Epoch 52/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3745 - val_loss: 0.3328\n",
      "Epoch 53/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3712 - val_loss: 0.3309\n",
      "Epoch 54/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3714 - val_loss: 0.3303\n",
      "Epoch 55/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3743 - val_loss: 0.3298\n",
      "Epoch 56/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3711 - val_loss: 0.3298\n",
      "Epoch 57/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3670 - val_loss: 0.3329\n",
      "Epoch 58/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3674 - val_loss: 0.3352\n",
      "Epoch 59/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3679 - val_loss: 0.3354\n",
      "Epoch 60/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3668 - val_loss: 0.3318\n",
      "Epoch 61/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3636 - val_loss: 0.3311\n",
      "Epoch 62/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3623 - val_loss: 0.3304\n",
      "Epoch 63/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3614 - val_loss: 0.3292\n",
      "Epoch 64/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3611 - val_loss: 0.3298\n",
      "Epoch 65/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3605 - val_loss: 0.3276\n",
      "Epoch 66/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3601 - val_loss: 0.3278\n",
      "Epoch 67/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3591 - val_loss: 0.3284\n",
      "Epoch 68/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3577 - val_loss: 0.3304\n",
      "Epoch 69/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3567 - val_loss: 0.3296\n",
      "Epoch 70/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3566 - val_loss: 0.3290\n",
      "Epoch 71/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3556 - val_loss: 0.3311\n",
      "Epoch 72/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3553 - val_loss: 0.3311\n",
      "Epoch 73/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3554 - val_loss: 0.3341\n",
      "Epoch 74/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3547 - val_loss: 0.3325\n",
      "Epoch 75/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3532 - val_loss: 0.3305\n",
      "Epoch 76/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3518 - val_loss: 0.3304\n",
      "Epoch 77/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3512 - val_loss: 0.3299\n",
      "Epoch 78/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3508 - val_loss: 0.3307\n",
      "Epoch 79/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3503 - val_loss: 0.3302\n",
      "Epoch 80/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3490 - val_loss: 0.3271\n",
      "Epoch 81/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3511 - val_loss: 0.3251\n",
      "Epoch 82/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3522 - val_loss: 0.3246\n",
      "Epoch 83/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3511 - val_loss: 0.3246\n",
      "Epoch 84/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3480 - val_loss: 0.3254\n",
      "Epoch 85/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3460 - val_loss: 0.3279\n",
      "Epoch 86/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3456 - val_loss: 0.3279\n",
      "Epoch 87/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3447 - val_loss: 0.3287\n",
      "Epoch 88/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3443 - val_loss: 0.3266\n",
      "Epoch 89/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3447 - val_loss: 0.3251\n",
      "Epoch 90/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3426 - val_loss: 0.3268\n",
      "Epoch 91/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3422 - val_loss: 0.3307\n",
      "Epoch 92/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3433 - val_loss: 0.3289\n",
      "Epoch 93/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3427 - val_loss: 0.3244\n",
      "Epoch 94/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3415 - val_loss: 0.3251\n",
      "Epoch 95/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3406 - val_loss: 0.3258\n",
      "Epoch 96/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3393 - val_loss: 0.3278\n",
      "Epoch 97/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3387 - val_loss: 0.3312\n",
      "Epoch 98/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3408 - val_loss: 0.3353\n",
      "Epoch 99/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3407 - val_loss: 0.3328\n",
      "Epoch 100/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3382 - val_loss: 0.3293\n",
      "Epoch 101/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3371 - val_loss: 0.3274\n",
      "Epoch 102/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3402 - val_loss: 0.3251\n",
      "Epoch 103/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3369 - val_loss: 0.3266\n",
      "Epoch 104/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3352 - val_loss: 0.3289\n",
      "Epoch 105/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3354 - val_loss: 0.3323\n",
      "Epoch 106/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3351 - val_loss: 0.3321\n",
      "Epoch 107/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3343 - val_loss: 0.3306\n",
      "Epoch 108/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3338 - val_loss: 0.3310\n",
      "Epoch 109/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3326 - val_loss: 0.3288\n",
      "Epoch 110/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3313 - val_loss: 0.3263\n",
      "Epoch 111/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3323 - val_loss: 0.3256\n",
      "Epoch 112/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3312 - val_loss: 0.3277\n",
      "Epoch 113/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3300 - val_loss: 0.3310\n",
      "Epoch 114/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3309 - val_loss: 0.3318\n",
      "Epoch 115/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3305 - val_loss: 0.3293\n",
      "Epoch 116/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3291 - val_loss: 0.3294\n",
      "Epoch 117/600\n",
      "8/8 [==============================] - 0s 21ms/step - loss: 0.3285 - val_loss: 0.3274\n",
      "Epoch 118/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3284 - val_loss: 0.3267\n",
      "Epoch 119/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3278 - val_loss: 0.3264\n",
      "Epoch 120/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3285 - val_loss: 0.3264\n",
      "Epoch 121/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3284 - val_loss: 0.3310\n",
      "Epoch 122/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3266 - val_loss: 0.3327\n",
      "Epoch 123/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3262 - val_loss: 0.3320\n",
      "Epoch 124/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3262 - val_loss: 0.3325\n",
      "Epoch 125/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3249 - val_loss: 0.3284\n",
      "Epoch 126/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3249 - val_loss: 0.3290\n",
      "Epoch 127/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3240 - val_loss: 0.3301\n",
      "Epoch 128/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.3240 - val_loss: 0.3274\n",
      "Epoch 129/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3262 - val_loss: 0.3240\n",
      "Epoch 130/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3262 - val_loss: 0.3241\n",
      "Epoch 131/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3246 - val_loss: 0.3245\n",
      "Epoch 132/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3230 - val_loss: 0.3279\n",
      "Epoch 133/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3206 - val_loss: 0.3313\n",
      "Epoch 134/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3204 - val_loss: 0.3291\n",
      "Epoch 135/600\n",
      "8/8 [==============================] - 0s 24ms/step - loss: 0.3195 - val_loss: 0.3286\n",
      "Epoch 136/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3199 - val_loss: 0.3280\n",
      "Epoch 137/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3197 - val_loss: 0.3269\n",
      "Epoch 138/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3205 - val_loss: 0.3317\n",
      "Epoch 139/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3192 - val_loss: 0.3419\n",
      "Epoch 140/600\n",
      "8/8 [==============================] - 0s 23ms/step - loss: 0.3232 - val_loss: 0.3449\n",
      "Epoch 141/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3217 - val_loss: 0.3366\n",
      "Epoch 142/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3166 - val_loss: 0.3299\n",
      "Epoch 143/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3181 - val_loss: 0.3289\n",
      "Epoch 144/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.3178 - val_loss: 0.3296\n",
      "Epoch 145/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3170 - val_loss: 0.3309\n",
      "Epoch 146/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3175 - val_loss: 0.3360\n",
      "Epoch 147/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3148 - val_loss: 0.3348\n",
      "Epoch 148/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3144 - val_loss: 0.3348\n",
      "Epoch 149/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3137 - val_loss: 0.3352\n",
      "Epoch 150/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3134 - val_loss: 0.3338\n",
      "Epoch 151/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3145 - val_loss: 0.3328\n",
      "Epoch 152/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3139 - val_loss: 0.3356\n",
      "Epoch 153/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3117 - val_loss: 0.3364\n",
      "Epoch 154/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3114 - val_loss: 0.3334\n",
      "Epoch 155/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3136 - val_loss: 0.3301\n",
      "Epoch 156/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3156 - val_loss: 0.3306\n",
      "Epoch 157/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3119 - val_loss: 0.3335\n",
      "Epoch 158/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3109 - val_loss: 0.3383\n",
      "Epoch 159/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3096 - val_loss: 0.3361\n",
      "Epoch 160/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3093 - val_loss: 0.3348\n",
      "Epoch 161/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3090 - val_loss: 0.3363\n",
      "Epoch 162/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3084 - val_loss: 0.3378\n",
      "Epoch 163/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3078 - val_loss: 0.3363\n",
      "Epoch 164/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3087 - val_loss: 0.3370\n",
      "Epoch 165/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3112 - val_loss: 0.3329\n",
      "Epoch 166/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3093 - val_loss: 0.3335\n",
      "Epoch 167/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3064 - val_loss: 0.3379\n",
      "Epoch 168/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3059 - val_loss: 0.3394\n",
      "Epoch 169/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3057 - val_loss: 0.3426\n",
      "Epoch 170/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3056 - val_loss: 0.3421\n",
      "Epoch 171/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3048 - val_loss: 0.3398\n",
      "Epoch 172/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3051 - val_loss: 0.3462\n",
      "Epoch 173/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3074 - val_loss: 0.3552\n",
      "Epoch 174/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3082 - val_loss: 0.3515\n",
      "Epoch 175/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3034 - val_loss: 0.3392\n",
      "Epoch 176/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3049 - val_loss: 0.3357\n",
      "Epoch 177/600\n",
      "8/8 [==============================] - 0s 23ms/step - loss: 0.3046 - val_loss: 0.3359\n",
      "Epoch 178/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3049 - val_loss: 0.3356\n",
      "Epoch 179/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3023 - val_loss: 0.3404\n",
      "Epoch 180/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3027 - val_loss: 0.3475\n",
      "Epoch 181/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3028 - val_loss: 0.3448\n",
      "Epoch 182/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3015 - val_loss: 0.3429\n",
      "Epoch 183/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3009 - val_loss: 0.3449\n",
      "Epoch 184/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3006 - val_loss: 0.3466\n",
      "Epoch 185/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3009 - val_loss: 0.3475\n",
      "Epoch 186/600\n",
      "8/8 [==============================] - 0s 28ms/step - loss: 0.2997 - val_loss: 0.3463\n",
      "Epoch 187/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2992 - val_loss: 0.3441\n",
      "Epoch 188/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2989 - val_loss: 0.3429\n",
      "Epoch 189/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2985 - val_loss: 0.3436\n",
      "Epoch 190/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2982 - val_loss: 0.3453\n",
      "Epoch 191/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2977 - val_loss: 0.3444\n",
      "Epoch 192/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2984 - val_loss: 0.3386\n",
      "Epoch 193/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3000 - val_loss: 0.3408\n",
      "Epoch 194/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2988 - val_loss: 0.3441\n",
      "Epoch 195/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.2977 - val_loss: 0.3466\n",
      "Epoch 196/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2965 - val_loss: 0.3454\n",
      "Epoch 197/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2966 - val_loss: 0.3441\n",
      "Epoch 198/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2958 - val_loss: 0.3462\n",
      "Epoch 199/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2957 - val_loss: 0.3480\n",
      "Epoch 200/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2955 - val_loss: 0.3415\n",
      "Epoch 201/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2980 - val_loss: 0.3389\n",
      "Epoch 202/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2969 - val_loss: 0.3447\n",
      "Epoch 203/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2944 - val_loss: 0.3486\n",
      "Epoch 204/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2944 - val_loss: 0.3514\n",
      "Epoch 205/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2946 - val_loss: 0.3498\n",
      "Epoch 206/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2939 - val_loss: 0.3494\n",
      "Epoch 207/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2931 - val_loss: 0.3502\n",
      "Epoch 208/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2947 - val_loss: 0.3552\n",
      "Epoch 209/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2947 - val_loss: 0.3505\n",
      "Epoch 210/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2934 - val_loss: 0.3461\n",
      "Epoch 211/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2924 - val_loss: 0.3479\n",
      "Epoch 212/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2920 - val_loss: 0.3493\n",
      "Epoch 213/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2922 - val_loss: 0.3523\n",
      "Epoch 214/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2916 - val_loss: 0.3513\n",
      "Epoch 215/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2906 - val_loss: 0.3493\n",
      "Epoch 216/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2899 - val_loss: 0.3484\n",
      "Epoch 217/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2902 - val_loss: 0.3490\n",
      "Epoch 218/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2905 - val_loss: 0.3507\n",
      "Epoch 219/600\n",
      "8/8 [==============================] - 0s 23ms/step - loss: 0.2903 - val_loss: 0.3490\n",
      "Epoch 220/600\n",
      "8/8 [==============================] - 0s 35ms/step - loss: 0.2889 - val_loss: 0.3620\n",
      "Epoch 221/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2929 - val_loss: 0.3669\n",
      "Epoch 222/600\n",
      "8/8 [==============================] - 0s 21ms/step - loss: 0.2932 - val_loss: 0.3620\n",
      "Epoch 223/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2882 - val_loss: 0.3531\n",
      "Epoch 224/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2874 - val_loss: 0.3505\n",
      "Epoch 225/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2877 - val_loss: 0.3506\n",
      "Epoch 226/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2873 - val_loss: 0.3534\n",
      "Epoch 227/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2873 - val_loss: 0.3567\n",
      "Epoch 228/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2870 - val_loss: 0.3541\n",
      "Epoch 229/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2867 - val_loss: 0.3525\n",
      "Epoch 230/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2868 - val_loss: 0.3558\n",
      "Epoch 231/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2861 - val_loss: 0.3623\n",
      "Epoch 232/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.2870 - val_loss: 0.3758\n",
      "Epoch 233/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2937 - val_loss: 0.3751\n",
      "Epoch 234/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2910 - val_loss: 0.3634\n",
      "Epoch 235/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2860 - val_loss: 0.3598\n",
      "Epoch 236/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2853 - val_loss: 0.3586\n",
      "Epoch 237/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2859 - val_loss: 0.3531\n",
      "Epoch 238/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2857 - val_loss: 0.3522\n",
      "Epoch 239/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2845 - val_loss: 0.3545\n",
      "Epoch 240/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2831 - val_loss: 0.3585\n",
      "Epoch 241/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2825 - val_loss: 0.3607\n",
      "Epoch 242/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2828 - val_loss: 0.3599\n",
      "Epoch 243/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2822 - val_loss: 0.3586\n",
      "Epoch 244/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2818 - val_loss: 0.3589\n",
      "Epoch 245/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2812 - val_loss: 0.3600\n",
      "Epoch 246/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2812 - val_loss: 0.3564\n",
      "Epoch 247/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2829 - val_loss: 0.3549\n",
      "Epoch 248/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2868 - val_loss: 0.3547\n",
      "Epoch 249/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2862 - val_loss: 0.3559\n",
      "Epoch 250/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2836 - val_loss: 0.3571\n",
      "Epoch 251/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2808 - val_loss: 0.3595\n",
      "Epoch 252/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2806 - val_loss: 0.3612\n",
      "Epoch 253/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2801 - val_loss: 0.3631\n",
      "Epoch 254/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2792 - val_loss: 0.3643\n",
      "Epoch 255/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2786 - val_loss: 0.3655\n",
      "Epoch 256/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2782 - val_loss: 0.3652\n",
      "Epoch 257/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2778 - val_loss: 0.3660\n",
      "Epoch 258/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2775 - val_loss: 0.3635\n",
      "Epoch 259/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2770 - val_loss: 0.3596\n",
      "Epoch 260/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2790 - val_loss: 0.3566\n",
      "Epoch 261/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2788 - val_loss: 0.3596\n",
      "Epoch 262/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2777 - val_loss: 0.3589\n",
      "Epoch 263/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2810 - val_loss: 0.3562\n",
      "Epoch 264/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2812 - val_loss: 0.3575\n",
      "Epoch 265/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2771 - val_loss: 0.3599\n",
      "Epoch 266/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2755 - val_loss: 0.3620\n",
      "Epoch 267/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2751 - val_loss: 0.3645\n",
      "Epoch 268/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2755 - val_loss: 0.3601\n",
      "Epoch 269/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2755 - val_loss: 0.3590\n",
      "Epoch 270/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2762 - val_loss: 0.3580\n",
      "Epoch 271/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2748 - val_loss: 0.3615\n",
      "Epoch 272/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2740 - val_loss: 0.3642\n",
      "Epoch 273/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2734 - val_loss: 0.3645\n",
      "Epoch 274/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2730 - val_loss: 0.3673\n",
      "Epoch 275/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2739 - val_loss: 0.3675\n",
      "Epoch 276/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2728 - val_loss: 0.3636\n",
      "Epoch 277/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2718 - val_loss: 0.3624\n",
      "Epoch 278/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2715 - val_loss: 0.3631\n",
      "Epoch 279/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2713 - val_loss: 0.3652\n",
      "Epoch 280/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2711 - val_loss: 0.3648\n",
      "Epoch 281/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2708 - val_loss: 0.3636\n",
      "Epoch 282/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2701 - val_loss: 0.3617\n",
      "Epoch 283/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2702 - val_loss: 0.3637\n",
      "Epoch 284/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2707 - val_loss: 0.3711\n",
      "Epoch 285/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2718 - val_loss: 0.3748\n",
      "Epoch 286/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2719 - val_loss: 0.3738\n",
      "Epoch 287/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2708 - val_loss: 0.3685\n",
      "Epoch 288/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2697 - val_loss: 0.3693\n",
      "Epoch 289/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2676 - val_loss: 0.3635\n",
      "Epoch 290/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2679 - val_loss: 0.3619\n",
      "Epoch 291/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2691 - val_loss: 0.3615\n",
      "Epoch 292/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2679 - val_loss: 0.3636\n",
      "Epoch 293/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2668 - val_loss: 0.3659\n",
      "Epoch 294/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2675 - val_loss: 0.3726\n",
      "Epoch 295/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2706 - val_loss: 0.3721\n",
      "Epoch 296/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2692 - val_loss: 0.3702\n",
      "Epoch 297/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2673 - val_loss: 0.3656\n",
      "Epoch 298/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2689 - val_loss: 0.3718\n",
      "Epoch 299/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2701 - val_loss: 0.3738\n",
      "Epoch 300/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2694 - val_loss: 0.3672\n",
      "Epoch 301/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2665 - val_loss: 0.3648\n",
      "Epoch 302/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2666 - val_loss: 0.3608\n",
      "Epoch 303/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2654 - val_loss: 0.3624\n",
      "Epoch 304/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2652 - val_loss: 0.3630\n",
      "Epoch 305/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2649 - val_loss: 0.3638\n",
      "Epoch 306/600\n",
      "8/8 [==============================] - 0s 29ms/step - loss: 0.2639 - val_loss: 0.3658\n",
      "Epoch 307/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2639 - val_loss: 0.3683\n",
      "Epoch 308/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2632 - val_loss: 0.3670\n",
      "Epoch 309/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2627 - val_loss: 0.3636\n",
      "Epoch 310/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2631 - val_loss: 0.3593\n",
      "Epoch 311/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2657 - val_loss: 0.3602\n",
      "Epoch 312/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2638 - val_loss: 0.3626\n",
      "Epoch 313/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2621 - val_loss: 0.3656\n",
      "Epoch 314/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2612 - val_loss: 0.3673\n",
      "Epoch 315/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2612 - val_loss: 0.3690\n",
      "Epoch 316/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2611 - val_loss: 0.3677\n",
      "Epoch 317/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2607 - val_loss: 0.3670\n",
      "Epoch 318/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2606 - val_loss: 0.3599\n",
      "Epoch 319/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2628 - val_loss: 0.3605\n",
      "Epoch 320/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2616 - val_loss: 0.3647\n",
      "Epoch 321/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2601 - val_loss: 0.3650\n",
      "Epoch 322/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2595 - val_loss: 0.3665\n",
      "Epoch 323/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2592 - val_loss: 0.3688\n",
      "Epoch 324/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2615 - val_loss: 0.3783\n",
      "Epoch 325/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2613 - val_loss: 0.3768\n",
      "Epoch 326/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2602 - val_loss: 0.3736\n",
      "Epoch 327/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2578 - val_loss: 0.3680\n",
      "Epoch 328/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2575 - val_loss: 0.3655\n",
      "Epoch 329/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2583 - val_loss: 0.3654\n",
      "Epoch 330/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2570 - val_loss: 0.3661\n",
      "Epoch 331/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2568 - val_loss: 0.3688\n",
      "Epoch 332/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2585 - val_loss: 0.3737\n",
      "Epoch 333/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2579 - val_loss: 0.3710\n",
      "Epoch 334/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2557 - val_loss: 0.3645\n",
      "Epoch 335/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2579 - val_loss: 0.3629\n",
      "Epoch 336/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2569 - val_loss: 0.3655\n",
      "Epoch 337/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2562 - val_loss: 0.3726\n",
      "Epoch 338/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2553 - val_loss: 0.3717\n",
      "Epoch 339/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2547 - val_loss: 0.3679\n",
      "Epoch 340/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2566 - val_loss: 0.3627\n",
      "Epoch 341/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2577 - val_loss: 0.3632\n",
      "Epoch 342/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2561 - val_loss: 0.3664\n",
      "Epoch 343/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2540 - val_loss: 0.3691\n",
      "Epoch 344/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2543 - val_loss: 0.3729\n",
      "Epoch 345/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2563 - val_loss: 0.3750\n",
      "Epoch 346/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2555 - val_loss: 0.3654\n",
      "Epoch 347/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2548 - val_loss: 0.3667\n",
      "Epoch 348/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2568 - val_loss: 0.3841\n",
      "Epoch 349/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2591 - val_loss: 0.3839\n",
      "Epoch 350/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2574 - val_loss: 0.3745\n",
      "Epoch 351/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2534 - val_loss: 0.3729\n",
      "Epoch 352/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2529 - val_loss: 0.3731\n",
      "Epoch 353/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2534 - val_loss: 0.3783\n",
      "Epoch 354/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2555 - val_loss: 0.3779\n",
      "Epoch 355/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2523 - val_loss: 0.3685\n",
      "Epoch 356/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2534 - val_loss: 0.3672\n",
      "Epoch 357/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2524 - val_loss: 0.3695\n",
      "Epoch 358/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2515 - val_loss: 0.3706\n",
      "Epoch 359/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2509 - val_loss: 0.3694\n",
      "Epoch 360/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2513 - val_loss: 0.3683\n",
      "Epoch 361/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2519 - val_loss: 0.3711\n",
      "Epoch 362/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.2508 - val_loss: 0.3707\n",
      "Epoch 363/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2503 - val_loss: 0.3744\n",
      "Epoch 364/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2511 - val_loss: 0.3750\n",
      "Epoch 365/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2497 - val_loss: 0.3646\n",
      "Epoch 366/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2513 - val_loss: 0.3653\n",
      "Epoch 367/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2487 - val_loss: 0.3714\n",
      "Epoch 368/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2499 - val_loss: 0.3689\n",
      "Epoch 369/600\n",
      "8/8 [==============================] - 0s 33ms/step - loss: 0.2487 - val_loss: 0.3655\n",
      "Epoch 370/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2518 - val_loss: 0.3618\n",
      "Epoch 371/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2541 - val_loss: 0.3632\n",
      "Epoch 372/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2506 - val_loss: 0.3690\n",
      "Epoch 373/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2481 - val_loss: 0.3720\n",
      "Epoch 374/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2481 - val_loss: 0.3787\n",
      "Epoch 375/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2486 - val_loss: 0.3804\n",
      "Epoch 376/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2474 - val_loss: 0.3766\n",
      "Epoch 377/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2473 - val_loss: 0.3718\n",
      "Epoch 378/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2483 - val_loss: 0.3771\n",
      "Epoch 379/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2469 - val_loss: 0.3763\n",
      "Epoch 380/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2463 - val_loss: 0.3760\n",
      "Epoch 381/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2461 - val_loss: 0.3750\n",
      "Epoch 382/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2454 - val_loss: 0.3715\n",
      "Epoch 383/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2460 - val_loss: 0.3729\n",
      "Epoch 384/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2454 - val_loss: 0.3735\n",
      "Epoch 385/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2454 - val_loss: 0.3757\n",
      "Epoch 386/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2444 - val_loss: 0.3713\n",
      "Epoch 387/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2455 - val_loss: 0.3717\n",
      "Epoch 388/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2455 - val_loss: 0.3732\n",
      "Epoch 389/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2444 - val_loss: 0.3784\n",
      "Epoch 390/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2441 - val_loss: 0.3774\n",
      "Epoch 391/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2439 - val_loss: 0.3776\n",
      "Epoch 392/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2440 - val_loss: 0.3743\n",
      "Epoch 393/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2433 - val_loss: 0.3753\n",
      "Epoch 394/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2430 - val_loss: 0.3755\n",
      "Epoch 395/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2427 - val_loss: 0.3769\n",
      "Epoch 396/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2426 - val_loss: 0.3775\n",
      "Epoch 397/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2424 - val_loss: 0.3783\n",
      "Epoch 398/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2429 - val_loss: 0.3797\n",
      "Epoch 399/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2418 - val_loss: 0.3730\n",
      "Epoch 400/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2433 - val_loss: 0.3724\n",
      "Epoch 401/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2436 - val_loss: 0.3717\n",
      "Epoch 402/600\n",
      "8/8 [==============================] - 0s 23ms/step - loss: 0.2432 - val_loss: 0.3729\n",
      "Epoch 403/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.2430 - val_loss: 0.3786\n",
      "Epoch 404/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2428 - val_loss: 0.3797\n",
      "Epoch 405/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2429 - val_loss: 0.3719\n",
      "Epoch 406/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2451 - val_loss: 0.3737\n",
      "Epoch 407/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2431 - val_loss: 0.3743\n",
      "Epoch 408/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2418 - val_loss: 0.3749\n",
      "Epoch 409/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2419 - val_loss: 0.3731\n",
      "Epoch 410/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2425 - val_loss: 0.3761\n",
      "Epoch 411/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2416 - val_loss: 0.3774\n",
      "Epoch 412/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2415 - val_loss: 0.3763\n",
      "Epoch 413/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2406 - val_loss: 0.3777\n",
      "Epoch 414/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2409 - val_loss: 0.3756\n",
      "Epoch 415/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2419 - val_loss: 0.3750\n",
      "Epoch 416/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2422 - val_loss: 0.3740\n",
      "Epoch 417/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2415 - val_loss: 0.3731\n",
      "Epoch 418/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2426 - val_loss: 0.3725\n",
      "Epoch 419/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2408 - val_loss: 0.3762\n",
      "Epoch 420/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2388 - val_loss: 0.3787\n",
      "Epoch 421/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2389 - val_loss: 0.3801\n",
      "Epoch 422/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2390 - val_loss: 0.3742\n",
      "Epoch 423/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2402 - val_loss: 0.3729\n",
      "Epoch 424/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2400 - val_loss: 0.3763\n",
      "Epoch 425/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2387 - val_loss: 0.3855\n",
      "Epoch 426/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2394 - val_loss: 0.3854\n",
      "Epoch 427/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2389 - val_loss: 0.3827\n",
      "Epoch 428/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2379 - val_loss: 0.3805\n",
      "Epoch 429/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2361 - val_loss: 0.3848\n",
      "Epoch 430/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2378 - val_loss: 0.3881\n",
      "Epoch 431/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2381 - val_loss: 0.3856\n",
      "Epoch 432/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2369 - val_loss: 0.3856\n",
      "Epoch 433/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2372 - val_loss: 0.3819\n",
      "Epoch 434/600\n",
      "8/8 [==============================] - 0s 30ms/step - loss: 0.2363 - val_loss: 0.3789\n",
      "Epoch 435/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2366 - val_loss: 0.3861\n",
      "Epoch 436/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2368 - val_loss: 0.3862\n",
      "Epoch 437/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2360 - val_loss: 0.3894\n",
      "Epoch 438/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2363 - val_loss: 0.3874\n",
      "Epoch 439/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2350 - val_loss: 0.3838\n",
      "Epoch 440/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2345 - val_loss: 0.3810\n",
      "Epoch 441/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2340 - val_loss: 0.3823\n",
      "Epoch 442/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2338 - val_loss: 0.3843\n",
      "Epoch 443/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2334 - val_loss: 0.3839\n",
      "Epoch 444/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2337 - val_loss: 0.3780\n",
      "Epoch 445/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2335 - val_loss: 0.3788\n",
      "Epoch 446/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2331 - val_loss: 0.3767\n",
      "Epoch 447/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2335 - val_loss: 0.3816\n",
      "Epoch 448/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2337 - val_loss: 0.3811\n",
      "Epoch 449/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2325 - val_loss: 0.3788\n",
      "Epoch 450/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2328 - val_loss: 0.3777\n",
      "Epoch 451/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2327 - val_loss: 0.3803\n",
      "Epoch 452/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2312 - val_loss: 0.3849\n",
      "Epoch 453/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2308 - val_loss: 0.3969\n",
      "Epoch 454/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2345 - val_loss: 0.3962\n",
      "Epoch 455/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2319 - val_loss: 0.3855\n",
      "Epoch 456/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2314 - val_loss: 0.3813\n",
      "Epoch 457/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2309 - val_loss: 0.3813\n",
      "Epoch 458/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2315 - val_loss: 0.3804\n",
      "Epoch 459/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2307 - val_loss: 0.3816\n",
      "Epoch 460/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2308 - val_loss: 0.3799\n",
      "Epoch 461/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2304 - val_loss: 0.3830\n",
      "Epoch 462/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2295 - val_loss: 0.3852\n",
      "Epoch 463/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.2287 - val_loss: 0.3869\n",
      "Epoch 464/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2293 - val_loss: 0.3852\n",
      "Epoch 465/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2285 - val_loss: 0.3870\n",
      "Epoch 466/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2291 - val_loss: 0.3854\n",
      "Epoch 467/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2283 - val_loss: 0.3866\n",
      "Epoch 468/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2285 - val_loss: 0.3906\n",
      "Epoch 469/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2271 - val_loss: 0.3891\n",
      "Epoch 470/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2275 - val_loss: 0.3885\n",
      "Epoch 471/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2270 - val_loss: 0.3886\n",
      "Epoch 472/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2269 - val_loss: 0.3883\n",
      "Epoch 473/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2269 - val_loss: 0.3920\n",
      "Epoch 474/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2267 - val_loss: 0.3880\n",
      "Epoch 475/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2273 - val_loss: 0.3865\n",
      "Epoch 476/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2271 - val_loss: 0.3931\n",
      "Epoch 477/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2261 - val_loss: 0.3933\n",
      "Epoch 478/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2257 - val_loss: 0.3919\n",
      "Epoch 479/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2248 - val_loss: 0.3887\n",
      "Epoch 480/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2263 - val_loss: 0.3898\n",
      "Epoch 481/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2255 - val_loss: 0.3913\n",
      "Epoch 482/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2247 - val_loss: 0.3921\n",
      "Epoch 483/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2244 - val_loss: 0.3922\n",
      "Epoch 484/600\n",
      "8/8 [==============================] - 0s 25ms/step - loss: 0.2242 - val_loss: 0.3897\n",
      "Epoch 485/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2240 - val_loss: 0.3915\n",
      "Epoch 486/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2227 - val_loss: 0.3948\n",
      "Epoch 487/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2237 - val_loss: 0.3963\n",
      "Epoch 488/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2232 - val_loss: 0.3972\n",
      "Epoch 489/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2236 - val_loss: 0.3963\n",
      "Epoch 490/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2235 - val_loss: 0.3987\n",
      "Epoch 491/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2253 - val_loss: 0.3961\n",
      "Epoch 492/600\n",
      "8/8 [==============================] - 0s 36ms/step - loss: 0.2213 - val_loss: 0.3850\n",
      "Epoch 493/600\n",
      "8/8 [==============================] - 0s 24ms/step - loss: 0.2254 - val_loss: 0.3872\n",
      "Epoch 494/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2247 - val_loss: 0.3941\n",
      "Epoch 495/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2227 - val_loss: 0.4004\n",
      "Epoch 496/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2223 - val_loss: 0.3989\n",
      "Epoch 497/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2224 - val_loss: 0.4022\n",
      "Epoch 498/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2214 - val_loss: 0.4000\n",
      "Epoch 499/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2214 - val_loss: 0.3934\n",
      "Epoch 500/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2224 - val_loss: 0.3917\n",
      "Epoch 501/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2201 - val_loss: 0.3980\n",
      "Epoch 502/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2207 - val_loss: 0.4012\n",
      "Epoch 503/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2214 - val_loss: 0.4023\n",
      "Epoch 504/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2192 - val_loss: 0.3956\n",
      "Epoch 505/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2208 - val_loss: 0.3926\n",
      "Epoch 506/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2204 - val_loss: 0.3951\n",
      "Epoch 507/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2190 - val_loss: 0.3984\n",
      "Epoch 508/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2190 - val_loss: 0.4009\n",
      "Epoch 509/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2186 - val_loss: 0.3979\n",
      "Epoch 510/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2208 - val_loss: 0.3938\n",
      "Epoch 511/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2217 - val_loss: 0.3939\n",
      "Epoch 512/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2215 - val_loss: 0.3920\n",
      "Epoch 513/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2211 - val_loss: 0.3896\n",
      "Epoch 514/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2206 - val_loss: 0.3918\n",
      "Epoch 515/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2200 - val_loss: 0.3946\n",
      "Epoch 516/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2192 - val_loss: 0.3947\n",
      "Epoch 517/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2169 - val_loss: 0.3980\n",
      "Epoch 518/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2172 - val_loss: 0.3977\n",
      "Epoch 519/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2176 - val_loss: 0.3990\n",
      "Epoch 520/600\n",
      "8/8 [==============================] - 0s 30ms/step - loss: 0.2166 - val_loss: 0.3970\n",
      "Epoch 521/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2168 - val_loss: 0.3978\n",
      "Epoch 522/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2163 - val_loss: 0.4150\n",
      "Epoch 523/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2251 - val_loss: 0.4265\n",
      "Epoch 524/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2233 - val_loss: 0.4143\n",
      "Epoch 525/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2185 - val_loss: 0.3991\n",
      "Epoch 526/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2159 - val_loss: 0.3959\n",
      "Epoch 527/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2149 - val_loss: 0.3920\n",
      "Epoch 528/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2160 - val_loss: 0.3916\n",
      "Epoch 529/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2140 - val_loss: 0.3992\n",
      "Epoch 530/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2157 - val_loss: 0.4090\n",
      "Epoch 531/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2190 - val_loss: 0.4151\n",
      "Epoch 532/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2173 - val_loss: 0.4093\n",
      "Epoch 533/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2147 - val_loss: 0.3994\n",
      "Epoch 534/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2138 - val_loss: 0.3929\n",
      "Epoch 535/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2143 - val_loss: 0.3944\n",
      "Epoch 536/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2134 - val_loss: 0.4004\n",
      "Epoch 537/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2129 - val_loss: 0.4013\n",
      "Epoch 538/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2134 - val_loss: 0.4041\n",
      "Epoch 539/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2131 - val_loss: 0.4030\n",
      "Epoch 540/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2126 - val_loss: 0.4033\n",
      "Epoch 541/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2123 - val_loss: 0.4017\n",
      "Epoch 542/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2117 - val_loss: 0.4033\n",
      "Epoch 543/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2118 - val_loss: 0.4028\n",
      "Epoch 544/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2115 - val_loss: 0.4018\n",
      "Epoch 545/600\n",
      "8/8 [==============================] - 0s 29ms/step - loss: 0.2112 - val_loss: 0.4007\n",
      "Epoch 546/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2111 - val_loss: 0.4020\n",
      "Epoch 547/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2113 - val_loss: 0.3956\n",
      "Epoch 548/600\n",
      "8/8 [==============================] - 0s 21ms/step - loss: 0.2130 - val_loss: 0.3890\n",
      "Epoch 549/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2137 - val_loss: 0.3928\n",
      "Epoch 550/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2118 - val_loss: 0.3970\n",
      "Epoch 551/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2109 - val_loss: 0.3986\n",
      "Epoch 552/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2104 - val_loss: 0.4001\n",
      "Epoch 553/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2099 - val_loss: 0.3998\n",
      "Epoch 554/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2099 - val_loss: 0.4007\n",
      "Epoch 555/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2101 - val_loss: 0.3969\n",
      "Epoch 556/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2095 - val_loss: 0.3970\n",
      "Epoch 557/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2102 - val_loss: 0.4059\n",
      "Epoch 558/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2100 - val_loss: 0.4060\n",
      "Epoch 559/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2096 - val_loss: 0.4036\n",
      "Epoch 560/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2085 - val_loss: 0.4005\n",
      "Epoch 561/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2094 - val_loss: 0.3925\n",
      "Epoch 562/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2218 - val_loss: 0.3859\n",
      "Epoch 563/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2257 - val_loss: 0.3900\n",
      "Epoch 564/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2164 - val_loss: 0.3989\n",
      "Epoch 565/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2080 - val_loss: 0.4078\n",
      "Epoch 566/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2074 - val_loss: 0.4138\n",
      "Epoch 567/600\n",
      "8/8 [==============================] - 0s 29ms/step - loss: 0.2107 - val_loss: 0.4206\n",
      "Epoch 568/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.2098 - val_loss: 0.4082\n",
      "Epoch 569/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.2082 - val_loss: 0.4026\n",
      "Epoch 570/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2076 - val_loss: 0.4072\n",
      "Epoch 571/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2071 - val_loss: 0.4089\n",
      "Epoch 572/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2067 - val_loss: 0.4077\n",
      "Epoch 573/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2072 - val_loss: 0.4043\n",
      "Epoch 574/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2082 - val_loss: 0.4046\n",
      "Epoch 575/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2073 - val_loss: 0.4088\n",
      "Epoch 576/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2057 - val_loss: 0.4033\n",
      "Epoch 577/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2078 - val_loss: 0.4020\n",
      "Epoch 578/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2069 - val_loss: 0.4085\n",
      "Epoch 579/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2057 - val_loss: 0.4125\n",
      "Epoch 580/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2057 - val_loss: 0.4116\n",
      "Epoch 581/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.2046 - val_loss: 0.4050\n",
      "Epoch 582/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2070 - val_loss: 0.4042\n",
      "Epoch 583/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2044 - val_loss: 0.4152\n",
      "Epoch 584/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2100 - val_loss: 0.4220\n",
      "Epoch 585/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2084 - val_loss: 0.4059\n",
      "Epoch 586/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.2067 - val_loss: 0.4040\n",
      "Epoch 587/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2063 - val_loss: 0.4041\n",
      "Epoch 588/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2057 - val_loss: 0.4037\n",
      "Epoch 589/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2049 - val_loss: 0.4069\n",
      "Epoch 590/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2039 - val_loss: 0.4093\n",
      "Epoch 591/600\n",
      "8/8 [==============================] - 0s 27ms/step - loss: 0.2039 - val_loss: 0.4197\n",
      "Epoch 592/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.2044 - val_loss: 0.4211\n",
      "Epoch 593/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.2038 - val_loss: 0.4154\n",
      "Epoch 594/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2028 - val_loss: 0.4021\n",
      "Epoch 595/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.2079 - val_loss: 0.4033\n",
      "Epoch 596/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2063 - val_loss: 0.4085\n",
      "Epoch 597/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.2032 - val_loss: 0.4132\n",
      "Epoch 598/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2039 - val_loss: 0.4083\n",
      "Epoch 599/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.2032 - val_loss: 0.4107\n",
      "Epoch 600/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.2040 - val_loss: 0.4192\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c838506290>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Entrenamos el modelo utilizando 600 iteraciones\n",
    "model.fit(\n",
    "    x=X_train,\n",
    "    y=y_train,\n",
    "    validation_data=(X_test, y_test),\n",
    "    epochs=600,\n",
    "    verbose=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizamos las pérdidas\n",
    "model_loss = pd.DataFrame(model.history.history)\n",
    "model_loss.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Al realizar 600 iteraciones se generó un Overfitting. Para evitarlo, aplicamos un Early Stopping para seguir a 'val_loss' y parar en base a la condición indicada\n",
    "\n",
    "\n",
    "\n",
    "    EarlyStopping() arguments:\n",
    "        monitor: Quantity to be monitored.\n",
    "        min_delta: Minimum change in the monitored quantity\n",
    "            to qualify as an improvement, i.e. an absolute\n",
    "            change of less than min_delta, will count as no\n",
    "            improvement.\n",
    "        patience: Number of epochs with no improvement\n",
    "            after which training will be stopped.\n",
    "        verbose: verbosity mode.\n",
    "        mode: One of `{\"auto\", \"min\", \"max\"}`. In `min` mode,\n",
    "            training will stop when the quantity\n",
    "            monitored has stopped decreasing; in `max`\n",
    "            mode it will stop when the quantity\n",
    "            monitored has stopped increasing; in `auto`\n",
    "            mode, the direction is automatically inferred\n",
    "            from the name of the monitored quantity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creamos el modelo nuevamente\n",
    "model = Sequential()\n",
    "num_neuronas = X_train.shape[1]\n",
    "model.add(Dense(units=num_neuronas, activation='relu'))\n",
    "model.add(Dense(units=np.round(num_neuronas/2), activation='relu'))\n",
    "model.add(Dense(units=1, activation='sigmoid'))\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "8/8 [==============================] - 1s 33ms/step - loss: 0.6926 - val_loss: 0.6981\n",
      "Epoch 2/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6911 - val_loss: 0.6972\n",
      "Epoch 3/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6895 - val_loss: 0.6956\n",
      "Epoch 4/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6882 - val_loss: 0.6940\n",
      "Epoch 5/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6869 - val_loss: 0.6927\n",
      "Epoch 6/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6856 - val_loss: 0.6913\n",
      "Epoch 7/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6838 - val_loss: 0.6900\n",
      "Epoch 8/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6819 - val_loss: 0.6886\n",
      "Epoch 9/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6799 - val_loss: 0.6869\n",
      "Epoch 10/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6779 - val_loss: 0.6850\n",
      "Epoch 11/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6754 - val_loss: 0.6831\n",
      "Epoch 12/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6731 - val_loss: 0.6810\n",
      "Epoch 13/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6701 - val_loss: 0.6786\n",
      "Epoch 14/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6673 - val_loss: 0.6759\n",
      "Epoch 15/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6647 - val_loss: 0.6729\n",
      "Epoch 16/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6620 - val_loss: 0.6698\n",
      "Epoch 17/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6586 - val_loss: 0.6667\n",
      "Epoch 18/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6548 - val_loss: 0.6638\n",
      "Epoch 19/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6511 - val_loss: 0.6609\n",
      "Epoch 20/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6475 - val_loss: 0.6577\n",
      "Epoch 21/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6434 - val_loss: 0.6543\n",
      "Epoch 22/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6393 - val_loss: 0.6504\n",
      "Epoch 23/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6355 - val_loss: 0.6463\n",
      "Epoch 24/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6305 - val_loss: 0.6418\n",
      "Epoch 25/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6256 - val_loss: 0.6368\n",
      "Epoch 26/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6210 - val_loss: 0.6313\n",
      "Epoch 27/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6162 - val_loss: 0.6265\n",
      "Epoch 28/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6122 - val_loss: 0.6227\n",
      "Epoch 29/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6082 - val_loss: 0.6184\n",
      "Epoch 30/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6034 - val_loss: 0.6148\n",
      "Epoch 31/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.5983 - val_loss: 0.6112\n",
      "Epoch 32/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.5939 - val_loss: 0.6077\n",
      "Epoch 33/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5888 - val_loss: 0.6046\n",
      "Epoch 34/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5845 - val_loss: 0.6004\n",
      "Epoch 35/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5801 - val_loss: 0.5962\n",
      "Epoch 36/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5757 - val_loss: 0.5933\n",
      "Epoch 37/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.5713 - val_loss: 0.5904\n",
      "Epoch 38/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5676 - val_loss: 0.5871\n",
      "Epoch 39/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.5634 - val_loss: 0.5821\n",
      "Epoch 40/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.5585 - val_loss: 0.5782\n",
      "Epoch 41/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5540 - val_loss: 0.5744\n",
      "Epoch 42/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5498 - val_loss: 0.5687\n",
      "Epoch 43/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5431 - val_loss: 0.5590\n",
      "Epoch 44/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5340 - val_loss: 0.5446\n",
      "Epoch 45/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5236 - val_loss: 0.5278\n",
      "Epoch 46/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5101 - val_loss: 0.5123\n",
      "Epoch 47/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4967 - val_loss: 0.4960\n",
      "Epoch 48/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.4849 - val_loss: 0.4810\n",
      "Epoch 49/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4754 - val_loss: 0.4675\n",
      "Epoch 50/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.4651 - val_loss: 0.4572\n",
      "Epoch 51/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4564 - val_loss: 0.4473\n",
      "Epoch 52/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4495 - val_loss: 0.4405\n",
      "Epoch 53/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4434 - val_loss: 0.4331\n",
      "Epoch 54/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4369 - val_loss: 0.4281\n",
      "Epoch 55/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4318 - val_loss: 0.4232\n",
      "Epoch 56/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4276 - val_loss: 0.4184\n",
      "Epoch 57/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4224 - val_loss: 0.4132\n",
      "Epoch 58/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4182 - val_loss: 0.4105\n",
      "Epoch 59/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4146 - val_loss: 0.4059\n",
      "Epoch 60/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4119 - val_loss: 0.4002\n",
      "Epoch 61/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4098 - val_loss: 0.3961\n",
      "Epoch 62/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.4056 - val_loss: 0.3941\n",
      "Epoch 63/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.4027 - val_loss: 0.3929\n",
      "Epoch 64/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4009 - val_loss: 0.3899\n",
      "Epoch 65/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3988 - val_loss: 0.3849\n",
      "Epoch 66/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3987 - val_loss: 0.3821\n",
      "Epoch 67/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3961 - val_loss: 0.3792\n",
      "Epoch 68/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3934 - val_loss: 0.3768\n",
      "Epoch 69/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3910 - val_loss: 0.3751\n",
      "Epoch 70/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3891 - val_loss: 0.3739\n",
      "Epoch 71/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3875 - val_loss: 0.3741\n",
      "Epoch 72/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3872 - val_loss: 0.3733\n",
      "Epoch 73/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3854 - val_loss: 0.3732\n",
      "Epoch 74/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3847 - val_loss: 0.3716\n",
      "Epoch 75/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3859 - val_loss: 0.3674\n",
      "Epoch 76/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3805 - val_loss: 0.3681\n",
      "Epoch 77/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3817 - val_loss: 0.3709\n",
      "Epoch 78/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3812 - val_loss: 0.3666\n",
      "Epoch 79/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3787 - val_loss: 0.3633\n",
      "Epoch 80/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3787 - val_loss: 0.3610\n",
      "Epoch 81/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.3774 - val_loss: 0.3606\n",
      "Epoch 82/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3765 - val_loss: 0.3615\n",
      "Epoch 83/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3760 - val_loss: 0.3630\n",
      "Epoch 84/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3776 - val_loss: 0.3680\n",
      "Epoch 85/600\n",
      "8/8 [==============================] - 0s 24ms/step - loss: 0.3793 - val_loss: 0.3640\n",
      "Epoch 86/600\n",
      "8/8 [==============================] - 0s 21ms/step - loss: 0.3757 - val_loss: 0.3573\n",
      "Epoch 87/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3730 - val_loss: 0.3567\n",
      "Epoch 88/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.3725 - val_loss: 0.3592\n",
      "Epoch 89/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3733 - val_loss: 0.3596\n",
      "Epoch 90/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3722 - val_loss: 0.3575\n",
      "Epoch 91/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3695 - val_loss: 0.3541\n",
      "Epoch 92/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3709 - val_loss: 0.3534\n",
      "Epoch 93/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3706 - val_loss: 0.3524\n",
      "Epoch 94/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3690 - val_loss: 0.3531\n",
      "Epoch 95/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3684 - val_loss: 0.3527\n",
      "Epoch 96/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3680 - val_loss: 0.3516\n",
      "Epoch 97/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3681 - val_loss: 0.3529\n",
      "Epoch 98/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3676 - val_loss: 0.3532\n",
      "Epoch 99/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3670 - val_loss: 0.3518\n",
      "Epoch 100/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3660 - val_loss: 0.3500\n",
      "Epoch 101/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3659 - val_loss: 0.3507\n",
      "Epoch 102/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3656 - val_loss: 0.3502\n",
      "Epoch 103/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3637 - val_loss: 0.3467\n",
      "Epoch 104/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3692 - val_loss: 0.3467\n",
      "Epoch 105/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3729 - val_loss: 0.3452\n",
      "Epoch 106/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3678 - val_loss: 0.3440\n",
      "Epoch 107/600\n",
      "8/8 [==============================] - 0s 24ms/step - loss: 0.3633 - val_loss: 0.3476\n",
      "Epoch 108/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3644 - val_loss: 0.3466\n",
      "Epoch 109/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3636 - val_loss: 0.3469\n",
      "Epoch 110/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3625 - val_loss: 0.3450\n",
      "Epoch 111/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3620 - val_loss: 0.3452\n",
      "Epoch 112/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3613 - val_loss: 0.3439\n",
      "Epoch 113/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3606 - val_loss: 0.3432\n",
      "Epoch 114/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3605 - val_loss: 0.3435\n",
      "Epoch 115/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3601 - val_loss: 0.3448\n",
      "Epoch 116/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3594 - val_loss: 0.3441\n",
      "Epoch 117/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3600 - val_loss: 0.3456\n",
      "Epoch 118/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3595 - val_loss: 0.3454\n",
      "Epoch 119/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3588 - val_loss: 0.3435\n",
      "Epoch 120/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3587 - val_loss: 0.3421\n",
      "Epoch 121/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3583 - val_loss: 0.3422\n",
      "Epoch 122/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3579 - val_loss: 0.3430\n",
      "Epoch 123/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3578 - val_loss: 0.3418\n",
      "Epoch 124/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3571 - val_loss: 0.3416\n",
      "Epoch 125/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3566 - val_loss: 0.3395\n",
      "Epoch 126/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3575 - val_loss: 0.3389\n",
      "Epoch 127/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3590 - val_loss: 0.3386\n",
      "Epoch 128/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3571 - val_loss: 0.3384\n",
      "Epoch 129/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3564 - val_loss: 0.3368\n",
      "Epoch 130/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3567 - val_loss: 0.3365\n",
      "Epoch 131/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3555 - val_loss: 0.3384\n",
      "Epoch 132/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3543 - val_loss: 0.3398\n",
      "Epoch 133/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3561 - val_loss: 0.3458\n",
      "Epoch 134/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3559 - val_loss: 0.3427\n",
      "Epoch 135/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3523 - val_loss: 0.3368\n",
      "Epoch 136/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3548 - val_loss: 0.3363\n",
      "Epoch 137/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3535 - val_loss: 0.3376\n",
      "Epoch 138/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3521 - val_loss: 0.3402\n",
      "Epoch 139/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3540 - val_loss: 0.3434\n",
      "Epoch 140/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3531 - val_loss: 0.3403\n",
      "Epoch 141/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3529 - val_loss: 0.3395\n",
      "Epoch 142/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3512 - val_loss: 0.3390\n",
      "Epoch 143/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3523 - val_loss: 0.3380\n",
      "Epoch 144/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3525 - val_loss: 0.3387\n",
      "Epoch 145/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3510 - val_loss: 0.3361\n",
      "Epoch 146/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3504 - val_loss: 0.3366\n",
      "Epoch 147/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3503 - val_loss: 0.3379\n",
      "Epoch 148/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3500 - val_loss: 0.3385\n",
      "Epoch 149/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3497 - val_loss: 0.3385\n",
      "Epoch 150/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3491 - val_loss: 0.3381\n",
      "Epoch 151/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.3495 - val_loss: 0.3397\n",
      "Epoch 152/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.3492 - val_loss: 0.3373\n",
      "Epoch 153/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3484 - val_loss: 0.3372\n",
      "Epoch 154/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3499 - val_loss: 0.3384\n",
      "Epoch 155/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3467 - val_loss: 0.3352\n",
      "Epoch 156/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3493 - val_loss: 0.3348\n",
      "Epoch 157/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3538 - val_loss: 0.3360\n",
      "Epoch 158/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3520 - val_loss: 0.3352\n",
      "Epoch 159/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3467 - val_loss: 0.3378\n",
      "Epoch 160/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.3471 - val_loss: 0.3391\n",
      "Epoch 161/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3473 - val_loss: 0.3388\n",
      "Epoch 162/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3469 - val_loss: 0.3367\n",
      "Epoch 163/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3462 - val_loss: 0.3330\n",
      "Epoch 164/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3484 - val_loss: 0.3321\n",
      "Epoch 165/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3456 - val_loss: 0.3334\n",
      "Epoch 166/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3456 - val_loss: 0.3337\n",
      "Epoch 167/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3450 - val_loss: 0.3319\n",
      "Epoch 168/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3445 - val_loss: 0.3320\n",
      "Epoch 169/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3439 - val_loss: 0.3322\n",
      "Epoch 170/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3433 - val_loss: 0.3323\n",
      "Epoch 171/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3451 - val_loss: 0.3316\n",
      "Epoch 172/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3433 - val_loss: 0.3316\n",
      "Epoch 173/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3439 - val_loss: 0.3313\n",
      "Epoch 174/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3438 - val_loss: 0.3312\n",
      "Epoch 175/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3424 - val_loss: 0.3324\n",
      "Epoch 176/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3433 - val_loss: 0.3358\n",
      "Epoch 177/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3418 - val_loss: 0.3320\n",
      "Epoch 178/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3419 - val_loss: 0.3311\n",
      "Epoch 179/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3420 - val_loss: 0.3314\n",
      "Epoch 180/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3407 - val_loss: 0.3335\n",
      "Epoch 181/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3416 - val_loss: 0.3397\n",
      "Epoch 182/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3422 - val_loss: 0.3396\n",
      "Epoch 183/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3426 - val_loss: 0.3348\n",
      "Epoch 184/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3395 - val_loss: 0.3328\n",
      "Epoch 185/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3409 - val_loss: 0.3309\n",
      "Epoch 186/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3407 - val_loss: 0.3310\n",
      "Epoch 187/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3395 - val_loss: 0.3318\n",
      "Epoch 188/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3384 - val_loss: 0.3330\n",
      "Epoch 189/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3387 - val_loss: 0.3358\n",
      "Epoch 190/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3384 - val_loss: 0.3359\n",
      "Epoch 191/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3381 - val_loss: 0.3373\n",
      "Epoch 192/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3407 - val_loss: 0.3386\n",
      "Epoch 193/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3378 - val_loss: 0.3323\n",
      "Epoch 194/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3362 - val_loss: 0.3308\n",
      "Epoch 195/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3363 - val_loss: 0.3306\n",
      "Epoch 196/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3361 - val_loss: 0.3297\n",
      "Epoch 197/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3368 - val_loss: 0.3288\n",
      "Epoch 198/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3360 - val_loss: 0.3302\n",
      "Epoch 199/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3346 - val_loss: 0.3311\n",
      "Epoch 200/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3350 - val_loss: 0.3348\n",
      "Epoch 201/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3359 - val_loss: 0.3335\n",
      "Epoch 202/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3346 - val_loss: 0.3334\n",
      "Epoch 203/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3338 - val_loss: 0.3316\n",
      "Epoch 204/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3352 - val_loss: 0.3282\n",
      "Epoch 205/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3333 - val_loss: 0.3285\n",
      "Epoch 206/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3338 - val_loss: 0.3302\n",
      "Epoch 207/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3321 - val_loss: 0.3298\n",
      "Epoch 208/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3320 - val_loss: 0.3265\n",
      "Epoch 209/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3318 - val_loss: 0.3275\n",
      "Epoch 210/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3309 - val_loss: 0.3271\n",
      "Epoch 211/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3306 - val_loss: 0.3281\n",
      "Epoch 212/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3298 - val_loss: 0.3290\n",
      "Epoch 213/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3300 - val_loss: 0.3296\n",
      "Epoch 214/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3287 - val_loss: 0.3272\n",
      "Epoch 215/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3297 - val_loss: 0.3263\n",
      "Epoch 216/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3288 - val_loss: 0.3271\n",
      "Epoch 217/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3279 - val_loss: 0.3290\n",
      "Epoch 218/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3286 - val_loss: 0.3317\n",
      "Epoch 219/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3286 - val_loss: 0.3315\n",
      "Epoch 220/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3282 - val_loss: 0.3294\n",
      "Epoch 221/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3273 - val_loss: 0.3285\n",
      "Epoch 222/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3265 - val_loss: 0.3292\n",
      "Epoch 223/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3264 - val_loss: 0.3270\n",
      "Epoch 224/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3253 - val_loss: 0.3256\n",
      "Epoch 225/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3272 - val_loss: 0.3254\n",
      "Epoch 226/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3261 - val_loss: 0.3257\n",
      "Epoch 227/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3255 - val_loss: 0.3306\n",
      "Epoch 228/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3267 - val_loss: 0.3331\n",
      "Epoch 229/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3265 - val_loss: 0.3300\n",
      "Epoch 230/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3249 - val_loss: 0.3242\n",
      "Epoch 231/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3281 - val_loss: 0.3249\n",
      "Epoch 232/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3262 - val_loss: 0.3246\n",
      "Epoch 233/600\n",
      "8/8 [==============================] - 0s 32ms/step - loss: 0.3242 - val_loss: 0.3252\n",
      "Epoch 234/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.3247 - val_loss: 0.3284\n",
      "Epoch 235/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3236 - val_loss: 0.3304\n",
      "Epoch 236/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3223 - val_loss: 0.3279\n",
      "Epoch 237/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3221 - val_loss: 0.3269\n",
      "Epoch 238/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3237 - val_loss: 0.3272\n",
      "Epoch 239/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3216 - val_loss: 0.3283\n",
      "Epoch 240/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3215 - val_loss: 0.3310\n",
      "Epoch 241/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3222 - val_loss: 0.3316\n",
      "Epoch 242/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3209 - val_loss: 0.3295\n",
      "Epoch 243/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3207 - val_loss: 0.3307\n",
      "Epoch 244/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3203 - val_loss: 0.3320\n",
      "Epoch 245/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3228 - val_loss: 0.3357\n",
      "Epoch 246/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3224 - val_loss: 0.3326\n",
      "Epoch 247/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3208 - val_loss: 0.3284\n",
      "Epoch 248/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3182 - val_loss: 0.3272\n",
      "Epoch 249/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3167 - val_loss: 0.3242\n",
      "Epoch 250/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.3187 - val_loss: 0.3240\n",
      "Epoch 251/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3185 - val_loss: 0.3244\n",
      "Epoch 252/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3182 - val_loss: 0.3231\n",
      "Epoch 253/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3204 - val_loss: 0.3236\n",
      "Epoch 254/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3220 - val_loss: 0.3226\n",
      "Epoch 255/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3193 - val_loss: 0.3219\n",
      "Epoch 256/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3171 - val_loss: 0.3265\n",
      "Epoch 257/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3157 - val_loss: 0.3278\n",
      "Epoch 258/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3154 - val_loss: 0.3249\n",
      "Epoch 259/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3143 - val_loss: 0.3247\n",
      "Epoch 260/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3139 - val_loss: 0.3260\n",
      "Epoch 261/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3135 - val_loss: 0.3295\n",
      "Epoch 262/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3154 - val_loss: 0.3311\n",
      "Epoch 263/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.3153 - val_loss: 0.3305\n",
      "Epoch 264/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3142 - val_loss: 0.3271\n",
      "Epoch 265/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3140 - val_loss: 0.3237\n",
      "Epoch 266/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3132 - val_loss: 0.3251\n",
      "Epoch 267/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3132 - val_loss: 0.3230\n",
      "Epoch 268/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3122 - val_loss: 0.3225\n",
      "Epoch 269/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3131 - val_loss: 0.3258\n",
      "Epoch 270/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3111 - val_loss: 0.3244\n",
      "Epoch 271/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3107 - val_loss: 0.3233\n",
      "Epoch 272/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3100 - val_loss: 0.3237\n",
      "Epoch 273/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.3094 - val_loss: 0.3245\n",
      "Epoch 274/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3102 - val_loss: 0.3241\n",
      "Epoch 275/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.3114 - val_loss: 0.3223\n",
      "Epoch 276/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.3101 - val_loss: 0.3233\n",
      "Epoch 277/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.3106 - val_loss: 0.3275\n",
      "Epoch 278/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.3104 - val_loss: 0.3272\n",
      "Epoch 279/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3130 - val_loss: 0.3309\n",
      "Epoch 280/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.3089 - val_loss: 0.3251\n",
      "Epoch 280: early stopping\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c839847100>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importamos la clase EarlyStopping desde keras.callbacks\n",
    "from keras.callbacks import EarlyStopping\n",
    "\n",
    "# Definimos un objeto de EarlyStopping con monitorización en 'val_loss', modo 'min' y patience de 25.\n",
    "early_stop = EarlyStopping(\n",
    "    monitor='val_loss',\n",
    "    mode='min',\n",
    "    patience=25,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "model.fit(\n",
    "    x=X_train,\n",
    "    y=y_train, \n",
    "    validation_data=(X_test, y_test),\n",
    "    epochs=600,\n",
    "    callbacks=[early_stop]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizamos las pérdidas\n",
    "model_loss = pd.DataFrame(model.history.history)\n",
    "model_loss.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Otro forma de evitar el Overfiting es añadiendo DropOut Layers, lo que consiste en desactivar capas en cada epoch, a las cuales no se le ajustaran los pesos y bias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importamos la clase Dropout desde keras.layers\n",
    "from keras.layers import Dropout\n",
    "\n",
    "# Creamos el modelo añadiendo los DropOuts\n",
    "model = Sequential()\n",
    "num_neuronas = X_train.shape[1]\n",
    "\n",
    "# Creamos la primer capa oculta\n",
    "model.add(Dense(units=num_neuronas, activation='relu'))\n",
    "\n",
    "# Desactivamos la mitad de las neuronas en cada epoch para esta capa\n",
    "model.add(Dropout(.5))\n",
    "\n",
    "# Creamos la segunda capa oculta\n",
    "model.add(Dense(units=np.round(num_neuronas/2), activation='relu'))\n",
    "\n",
    "# Desactivamos la mitad de las neuronas en cada epoch para esta capa\n",
    "model.add(Dropout(.5))\n",
    "\n",
    "model.add(Dense(units=1, activation='sigmoid'))\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "8/8 [==============================] - 1s 34ms/step - loss: 0.7223 - val_loss: 0.7065\n",
      "Epoch 2/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.7670 - val_loss: 0.6986\n",
      "Epoch 3/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.7203 - val_loss: 0.6934\n",
      "Epoch 4/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.7472 - val_loss: 0.6887\n",
      "Epoch 5/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6885 - val_loss: 0.6839\n",
      "Epoch 6/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.7222 - val_loss: 0.6817\n",
      "Epoch 7/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.7305 - val_loss: 0.6807\n",
      "Epoch 8/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6963 - val_loss: 0.6786\n",
      "Epoch 9/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6836 - val_loss: 0.6771\n",
      "Epoch 10/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6714 - val_loss: 0.6747\n",
      "Epoch 11/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.7052 - val_loss: 0.6738\n",
      "Epoch 12/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6852 - val_loss: 0.6743\n",
      "Epoch 13/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6999 - val_loss: 0.6729\n",
      "Epoch 14/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6855 - val_loss: 0.6714\n",
      "Epoch 15/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.7120 - val_loss: 0.6719\n",
      "Epoch 16/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6818 - val_loss: 0.6719\n",
      "Epoch 17/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6524 - val_loss: 0.6704\n",
      "Epoch 18/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.7031 - val_loss: 0.6683\n",
      "Epoch 19/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6540 - val_loss: 0.6651\n",
      "Epoch 20/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6779 - val_loss: 0.6615\n",
      "Epoch 21/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6677 - val_loss: 0.6578\n",
      "Epoch 22/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6720 - val_loss: 0.6549\n",
      "Epoch 23/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6711 - val_loss: 0.6509\n",
      "Epoch 24/600\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.6674 - val_loss: 0.6477\n",
      "Epoch 25/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6652 - val_loss: 0.6479\n",
      "Epoch 26/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.6557 - val_loss: 0.6466\n",
      "Epoch 27/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6793 - val_loss: 0.6449\n",
      "Epoch 28/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.6378 - val_loss: 0.6433\n",
      "Epoch 29/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.6519 - val_loss: 0.6401\n",
      "Epoch 30/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6388 - val_loss: 0.6358\n",
      "Epoch 31/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6586 - val_loss: 0.6317\n",
      "Epoch 32/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.6473 - val_loss: 0.6268\n",
      "Epoch 33/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.6519 - val_loss: 0.6216\n",
      "Epoch 34/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.6547 - val_loss: 0.6202\n",
      "Epoch 35/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.6396 - val_loss: 0.6171\n",
      "Epoch 36/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.6250 - val_loss: 0.6120\n",
      "Epoch 37/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.6346 - val_loss: 0.6071\n",
      "Epoch 38/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6427 - val_loss: 0.6013\n",
      "Epoch 39/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6268 - val_loss: 0.5951\n",
      "Epoch 40/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6480 - val_loss: 0.5915\n",
      "Epoch 41/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6313 - val_loss: 0.5874\n",
      "Epoch 42/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6538 - val_loss: 0.5844\n",
      "Epoch 43/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6315 - val_loss: 0.5798\n",
      "Epoch 44/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.6503 - val_loss: 0.5755\n",
      "Epoch 45/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.6492 - val_loss: 0.5739\n",
      "Epoch 46/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.6409 - val_loss: 0.5724\n",
      "Epoch 47/600\n",
      "8/8 [==============================] - 0s 32ms/step - loss: 0.6071 - val_loss: 0.5708\n",
      "Epoch 48/600\n",
      "8/8 [==============================] - 0s 18ms/step - loss: 0.6250 - val_loss: 0.5673\n",
      "Epoch 49/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6024 - val_loss: 0.5634\n",
      "Epoch 50/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.6272 - val_loss: 0.5611\n",
      "Epoch 51/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.6273 - val_loss: 0.5579\n",
      "Epoch 52/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.6152 - val_loss: 0.5554\n",
      "Epoch 53/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.6052 - val_loss: 0.5551\n",
      "Epoch 54/600\n",
      "8/8 [==============================] - 0s 21ms/step - loss: 0.6260 - val_loss: 0.5526\n",
      "Epoch 55/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.6321 - val_loss: 0.5554\n",
      "Epoch 56/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6275 - val_loss: 0.5549\n",
      "Epoch 57/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.6081 - val_loss: 0.5529\n",
      "Epoch 58/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5985 - val_loss: 0.5511\n",
      "Epoch 59/600\n",
      "8/8 [==============================] - 0s 17ms/step - loss: 0.5963 - val_loss: 0.5505\n",
      "Epoch 60/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.5967 - val_loss: 0.5492\n",
      "Epoch 61/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.6161 - val_loss: 0.5461\n",
      "Epoch 62/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6034 - val_loss: 0.5437\n",
      "Epoch 63/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5712 - val_loss: 0.5392\n",
      "Epoch 64/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.6196 - val_loss: 0.5342\n",
      "Epoch 65/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5867 - val_loss: 0.5298\n",
      "Epoch 66/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5808 - val_loss: 0.5254\n",
      "Epoch 67/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5987 - val_loss: 0.5222\n",
      "Epoch 68/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.6131 - val_loss: 0.5194\n",
      "Epoch 69/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5727 - val_loss: 0.5163\n",
      "Epoch 70/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5923 - val_loss: 0.5137\n",
      "Epoch 71/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5651 - val_loss: 0.5117\n",
      "Epoch 72/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5809 - val_loss: 0.5083\n",
      "Epoch 73/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5886 - val_loss: 0.5055\n",
      "Epoch 74/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5532 - val_loss: 0.5034\n",
      "Epoch 75/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5712 - val_loss: 0.5005\n",
      "Epoch 76/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5400 - val_loss: 0.4986\n",
      "Epoch 77/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5904 - val_loss: 0.4960\n",
      "Epoch 78/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5700 - val_loss: 0.4928\n",
      "Epoch 79/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5705 - val_loss: 0.4913\n",
      "Epoch 80/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5693 - val_loss: 0.4895\n",
      "Epoch 81/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5663 - val_loss: 0.4898\n",
      "Epoch 82/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5442 - val_loss: 0.4886\n",
      "Epoch 83/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5928 - val_loss: 0.4873\n",
      "Epoch 84/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5408 - val_loss: 0.4855\n",
      "Epoch 85/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5597 - val_loss: 0.4831\n",
      "Epoch 86/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5624 - val_loss: 0.4840\n",
      "Epoch 87/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.5661 - val_loss: 0.4814\n",
      "Epoch 88/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5807 - val_loss: 0.4795\n",
      "Epoch 89/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5203 - val_loss: 0.4746\n",
      "Epoch 90/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5674 - val_loss: 0.4700\n",
      "Epoch 91/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5426 - val_loss: 0.4665\n",
      "Epoch 92/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5826 - val_loss: 0.4635\n",
      "Epoch 93/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5409 - val_loss: 0.4607\n",
      "Epoch 94/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5532 - val_loss: 0.4578\n",
      "Epoch 95/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5482 - val_loss: 0.4546\n",
      "Epoch 96/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5553 - val_loss: 0.4518\n",
      "Epoch 97/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5429 - val_loss: 0.4495\n",
      "Epoch 98/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5502 - val_loss: 0.4479\n",
      "Epoch 99/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5510 - val_loss: 0.4464\n",
      "Epoch 100/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5510 - val_loss: 0.4448\n",
      "Epoch 101/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5583 - val_loss: 0.4431\n",
      "Epoch 102/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5242 - val_loss: 0.4414\n",
      "Epoch 103/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5289 - val_loss: 0.4399\n",
      "Epoch 104/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.5730 - val_loss: 0.4425\n",
      "Epoch 105/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5650 - val_loss: 0.4448\n",
      "Epoch 106/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5339 - val_loss: 0.4459\n",
      "Epoch 107/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5674 - val_loss: 0.4466\n",
      "Epoch 108/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5244 - val_loss: 0.4449\n",
      "Epoch 109/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5495 - val_loss: 0.4435\n",
      "Epoch 110/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5338 - val_loss: 0.4420\n",
      "Epoch 111/600\n",
      "8/8 [==============================] - 0s 26ms/step - loss: 0.5558 - val_loss: 0.4404\n",
      "Epoch 112/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5591 - val_loss: 0.4386\n",
      "Epoch 113/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5579 - val_loss: 0.4376\n",
      "Epoch 114/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5169 - val_loss: 0.4354\n",
      "Epoch 115/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5270 - val_loss: 0.4335\n",
      "Epoch 116/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5344 - val_loss: 0.4313\n",
      "Epoch 117/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5313 - val_loss: 0.4288\n",
      "Epoch 118/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4920 - val_loss: 0.4251\n",
      "Epoch 119/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5318 - val_loss: 0.4216\n",
      "Epoch 120/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4948 - val_loss: 0.4183\n",
      "Epoch 121/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5509 - val_loss: 0.4176\n",
      "Epoch 122/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5036 - val_loss: 0.4208\n",
      "Epoch 123/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5207 - val_loss: 0.4237\n",
      "Epoch 124/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5537 - val_loss: 0.4242\n",
      "Epoch 125/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5612 - val_loss: 0.4242\n",
      "Epoch 126/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5227 - val_loss: 0.4235\n",
      "Epoch 127/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4884 - val_loss: 0.4220\n",
      "Epoch 128/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5199 - val_loss: 0.4214\n",
      "Epoch 129/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5686 - val_loss: 0.4217\n",
      "Epoch 130/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5325 - val_loss: 0.4215\n",
      "Epoch 131/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5074 - val_loss: 0.4183\n",
      "Epoch 132/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5253 - val_loss: 0.4166\n",
      "Epoch 133/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5067 - val_loss: 0.4164\n",
      "Epoch 134/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4954 - val_loss: 0.4140\n",
      "Epoch 135/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5084 - val_loss: 0.4106\n",
      "Epoch 136/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5285 - val_loss: 0.4088\n",
      "Epoch 137/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5274 - val_loss: 0.4089\n",
      "Epoch 138/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5221 - val_loss: 0.4096\n",
      "Epoch 139/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4903 - val_loss: 0.4075\n",
      "Epoch 140/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5192 - val_loss: 0.4046\n",
      "Epoch 141/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5038 - val_loss: 0.4034\n",
      "Epoch 142/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.5207 - val_loss: 0.4032\n",
      "Epoch 143/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.5145 - val_loss: 0.4023\n",
      "Epoch 144/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4982 - val_loss: 0.4016\n",
      "Epoch 145/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5041 - val_loss: 0.4007\n",
      "Epoch 146/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5240 - val_loss: 0.4004\n",
      "Epoch 147/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4962 - val_loss: 0.3987\n",
      "Epoch 148/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5067 - val_loss: 0.3966\n",
      "Epoch 149/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4933 - val_loss: 0.3949\n",
      "Epoch 150/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.5515 - val_loss: 0.3946\n",
      "Epoch 151/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4955 - val_loss: 0.3938\n",
      "Epoch 152/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4880 - val_loss: 0.3923\n",
      "Epoch 153/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4716 - val_loss: 0.3905\n",
      "Epoch 154/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5032 - val_loss: 0.3883\n",
      "Epoch 155/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4863 - val_loss: 0.3873\n",
      "Epoch 156/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5175 - val_loss: 0.3899\n",
      "Epoch 157/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4816 - val_loss: 0.3903\n",
      "Epoch 158/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4797 - val_loss: 0.3890\n",
      "Epoch 159/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4907 - val_loss: 0.3870\n",
      "Epoch 160/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5011 - val_loss: 0.3849\n",
      "Epoch 161/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5113 - val_loss: 0.3839\n",
      "Epoch 162/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5026 - val_loss: 0.3827\n",
      "Epoch 163/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4711 - val_loss: 0.3824\n",
      "Epoch 164/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4783 - val_loss: 0.3822\n",
      "Epoch 165/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5183 - val_loss: 0.3825\n",
      "Epoch 166/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4842 - val_loss: 0.3818\n",
      "Epoch 167/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5073 - val_loss: 0.3807\n",
      "Epoch 168/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5138 - val_loss: 0.3806\n",
      "Epoch 169/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5117 - val_loss: 0.3802\n",
      "Epoch 170/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4623 - val_loss: 0.3801\n",
      "Epoch 171/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4800 - val_loss: 0.3797\n",
      "Epoch 172/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4838 - val_loss: 0.3789\n",
      "Epoch 173/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4912 - val_loss: 0.3777\n",
      "Epoch 174/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5011 - val_loss: 0.3764\n",
      "Epoch 175/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4815 - val_loss: 0.3751\n",
      "Epoch 176/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4664 - val_loss: 0.3742\n",
      "Epoch 177/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4856 - val_loss: 0.3742\n",
      "Epoch 178/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4779 - val_loss: 0.3741\n",
      "Epoch 179/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4659 - val_loss: 0.3733\n",
      "Epoch 180/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5128 - val_loss: 0.3732\n",
      "Epoch 181/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4847 - val_loss: 0.3741\n",
      "Epoch 182/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4927 - val_loss: 0.3743\n",
      "Epoch 183/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5314 - val_loss: 0.3755\n",
      "Epoch 184/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4575 - val_loss: 0.3760\n",
      "Epoch 185/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4939 - val_loss: 0.3766\n",
      "Epoch 186/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4638 - val_loss: 0.3760\n",
      "Epoch 187/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4704 - val_loss: 0.3747\n",
      "Epoch 188/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4877 - val_loss: 0.3742\n",
      "Epoch 189/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5200 - val_loss: 0.3729\n",
      "Epoch 190/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4732 - val_loss: 0.3715\n",
      "Epoch 191/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4976 - val_loss: 0.3706\n",
      "Epoch 192/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5063 - val_loss: 0.3707\n",
      "Epoch 193/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4857 - val_loss: 0.3700\n",
      "Epoch 194/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4674 - val_loss: 0.3686\n",
      "Epoch 195/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5221 - val_loss: 0.3680\n",
      "Epoch 196/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4488 - val_loss: 0.3670\n",
      "Epoch 197/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5184 - val_loss: 0.3667\n",
      "Epoch 198/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4972 - val_loss: 0.3663\n",
      "Epoch 199/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4711 - val_loss: 0.3674\n",
      "Epoch 200/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4950 - val_loss: 0.3684\n",
      "Epoch 201/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4956 - val_loss: 0.3706\n",
      "Epoch 202/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4562 - val_loss: 0.3711\n",
      "Epoch 203/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4486 - val_loss: 0.3707\n",
      "Epoch 204/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4109 - val_loss: 0.3691\n",
      "Epoch 205/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4335 - val_loss: 0.3675\n",
      "Epoch 206/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5061 - val_loss: 0.3676\n",
      "Epoch 207/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4951 - val_loss: 0.3681\n",
      "Epoch 208/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4387 - val_loss: 0.3676\n",
      "Epoch 209/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4621 - val_loss: 0.3670\n",
      "Epoch 210/600\n",
      "8/8 [==============================] - 0s 20ms/step - loss: 0.4637 - val_loss: 0.3671\n",
      "Epoch 211/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4591 - val_loss: 0.3697\n",
      "Epoch 212/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4811 - val_loss: 0.3725\n",
      "Epoch 213/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4623 - val_loss: 0.3724\n",
      "Epoch 214/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4728 - val_loss: 0.3721\n",
      "Epoch 215/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.5077 - val_loss: 0.3712\n",
      "Epoch 216/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4475 - val_loss: 0.3704\n",
      "Epoch 217/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.5035 - val_loss: 0.3688\n",
      "Epoch 218/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4417 - val_loss: 0.3675\n",
      "Epoch 219/600\n",
      "8/8 [==============================] - 0s 25ms/step - loss: 0.5029 - val_loss: 0.3666\n",
      "Epoch 220/600\n",
      "8/8 [==============================] - 0s 22ms/step - loss: 0.4814 - val_loss: 0.3657\n",
      "Epoch 221/600\n",
      "8/8 [==============================] - 0s 19ms/step - loss: 0.4606 - val_loss: 0.3660\n",
      "Epoch 222/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4587 - val_loss: 0.3672\n",
      "Epoch 223/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.4829 - val_loss: 0.3683\n",
      "Epoch 224/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.5156 - val_loss: 0.3694\n",
      "Epoch 225/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4721 - val_loss: 0.3689\n",
      "Epoch 226/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.5298 - val_loss: 0.3675\n",
      "Epoch 227/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4643 - val_loss: 0.3669\n",
      "Epoch 228/600\n",
      "8/8 [==============================] - 0s 14ms/step - loss: 0.4562 - val_loss: 0.3658\n",
      "Epoch 229/600\n",
      "8/8 [==============================] - 0s 16ms/step - loss: 0.4403 - val_loss: 0.3640\n",
      "Epoch 230/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4548 - val_loss: 0.3634\n",
      "Epoch 231/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4776 - val_loss: 0.3638\n",
      "Epoch 232/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4589 - val_loss: 0.3637\n",
      "Epoch 233/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4707 - val_loss: 0.3658\n",
      "Epoch 234/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4676 - val_loss: 0.3658\n",
      "Epoch 235/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4840 - val_loss: 0.3670\n",
      "Epoch 236/600\n",
      "8/8 [==============================] - 0s 15ms/step - loss: 0.4468 - val_loss: 0.3666\n",
      "Epoch 237/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4937 - val_loss: 0.3659\n",
      "Epoch 238/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4620 - val_loss: 0.3671\n",
      "Epoch 239/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4430 - val_loss: 0.3673\n",
      "Epoch 240/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4352 - val_loss: 0.3657\n",
      "Epoch 241/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4723 - val_loss: 0.3639\n",
      "Epoch 242/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4300 - val_loss: 0.3632\n",
      "Epoch 243/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4890 - val_loss: 0.3614\n",
      "Epoch 244/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4908 - val_loss: 0.3591\n",
      "Epoch 245/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4896 - val_loss: 0.3583\n",
      "Epoch 246/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4731 - val_loss: 0.3581\n",
      "Epoch 247/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4412 - val_loss: 0.3589\n",
      "Epoch 248/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4802 - val_loss: 0.3590\n",
      "Epoch 249/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4488 - val_loss: 0.3590\n",
      "Epoch 250/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4671 - val_loss: 0.3596\n",
      "Epoch 251/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4701 - val_loss: 0.3598\n",
      "Epoch 252/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5020 - val_loss: 0.3607\n",
      "Epoch 253/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5152 - val_loss: 0.3614\n",
      "Epoch 254/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4500 - val_loss: 0.3667\n",
      "Epoch 255/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4358 - val_loss: 0.3677\n",
      "Epoch 256/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4176 - val_loss: 0.3654\n",
      "Epoch 257/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4961 - val_loss: 0.3633\n",
      "Epoch 258/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4391 - val_loss: 0.3619\n",
      "Epoch 259/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4985 - val_loss: 0.3612\n",
      "Epoch 260/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4639 - val_loss: 0.3606\n",
      "Epoch 261/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4566 - val_loss: 0.3600\n",
      "Epoch 262/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4523 - val_loss: 0.3589\n",
      "Epoch 263/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4411 - val_loss: 0.3584\n",
      "Epoch 264/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4433 - val_loss: 0.3570\n",
      "Epoch 265/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4393 - val_loss: 0.3555\n",
      "Epoch 266/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4412 - val_loss: 0.3538\n",
      "Epoch 267/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4314 - val_loss: 0.3521\n",
      "Epoch 268/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4674 - val_loss: 0.3511\n",
      "Epoch 269/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4568 - val_loss: 0.3523\n",
      "Epoch 270/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4772 - val_loss: 0.3537\n",
      "Epoch 271/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4389 - val_loss: 0.3533\n",
      "Epoch 272/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.5001 - val_loss: 0.3552\n",
      "Epoch 273/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4862 - val_loss: 0.3576\n",
      "Epoch 274/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4979 - val_loss: 0.3587\n",
      "Epoch 275/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4367 - val_loss: 0.3635\n",
      "Epoch 276/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4667 - val_loss: 0.3633\n",
      "Epoch 277/600\n",
      "8/8 [==============================] - 0s 13ms/step - loss: 0.4454 - val_loss: 0.3624\n",
      "Epoch 278/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4430 - val_loss: 0.3602\n",
      "Epoch 279/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4112 - val_loss: 0.3580\n",
      "Epoch 280/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4736 - val_loss: 0.3569\n",
      "Epoch 281/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4447 - val_loss: 0.3561\n",
      "Epoch 282/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4387 - val_loss: 0.3557\n",
      "Epoch 283/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4527 - val_loss: 0.3557\n",
      "Epoch 284/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4118 - val_loss: 0.3552\n",
      "Epoch 285/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4435 - val_loss: 0.3544\n",
      "Epoch 286/600\n",
      "8/8 [==============================] - 0s 12ms/step - loss: 0.4801 - val_loss: 0.3569\n",
      "Epoch 287/600\n",
      "8/8 [==============================] - 0s 11ms/step - loss: 0.4544 - val_loss: 0.3577\n",
      "Epoch 288/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4247 - val_loss: 0.3568\n",
      "Epoch 289/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4187 - val_loss: 0.3557\n",
      "Epoch 290/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4800 - val_loss: 0.3551\n",
      "Epoch 291/600\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.4678 - val_loss: 0.3548\n",
      "Epoch 292/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4912 - val_loss: 0.3550\n",
      "Epoch 293/600\n",
      "8/8 [==============================] - 0s 10ms/step - loss: 0.4889 - val_loss: 0.3571\n",
      "Epoch 293: early stopping\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c83a9d3040>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Entrenamos el modelo utilizando Early Stopping\n",
    "model.fit(\n",
    "    x=X_train,\n",
    "    y=y_train, \n",
    "    validation_data=(X_test, y_test),\n",
    "    epochs=600,\n",
    "    callbacks=[early_stop]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizamos las pérdidas\n",
    "model_loss = pd.DataFrame(model.history.history)\n",
    "model_loss.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluación del modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importamos los métodos necesarios\n",
    "from sklearn.metrics import classification_report, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3/3 [==============================] - 0s 4ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.88      0.78      0.83        37\n",
      "           1       0.81      0.90      0.85        39\n",
      "\n",
      "    accuracy                           0.84        76\n",
      "   macro avg       0.85      0.84      0.84        76\n",
      "weighted avg       0.85      0.84      0.84        76\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Realizamos una predicción sobre el conjunto de test\n",
    "predictions = np.round(model.predict(X_test))\n",
    "\n",
    "# Generamos un reporte de clasificación comparando los valores reales con las predicciones\n",
    "print(classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[29  8]\n",
      " [ 4 35]]\n"
     ]
    }
   ],
   "source": [
    "# Generamos una matriz de confusión para analizar los errores. Retorna una matriz donde se muestra:\n",
    "#  ---------------------\n",
    "# |   | acierto | error |   osea:\n",
    "# |---+---------+-------|        [\n",
    "# | 1 |    n    |   n   |          [0 en ambos, 0 real y predicción 1],\n",
    "# |---+---------+-------|          [1 real y predicción 0, 1 en ambos]\n",
    "# | 0 |    n    |   n   |        ]\n",
    "#  ---------------------\n",
    "\n",
    "print(confusion_matrix(y_test, predictions))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicción de nuevos datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 53ms/step\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "      <th>target</th>\n",
       "      <th>PREDICCIÓN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>145</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>0</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>55</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>178</td>\n",
       "      <td>0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>58</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>1</td>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>55</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>140</td>\n",
       "      <td>0</td>\n",
       "      <td>153</td>\n",
       "      <td>0</td>\n",
       "      <td>1.3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>173</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>172</td>\n",
       "      <td>1</td>\n",
       "      <td>162</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>59</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>150</td>\n",
       "      <td>1</td>\n",
       "      <td>174</td>\n",
       "      <td>0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>52</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>1</td>\n",
       "      <td>160</td>\n",
       "      <td>0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "      <td>139</td>\n",
       "      <td>0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>54</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "      <td>171</td>\n",
       "      <td>0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    age  sex  cp  trestbps  restecg  thalach  exang  oldpeak  slope  ca  thal  \\\n",
       "0    60    1   3       145        0      150      0      2.3      0   0     1   \n",
       "1    42    1   2       130        1      187      0      3.5      0   0     2   \n",
       "2    40    0   1       130        0      172      0      1.4      2   0     2   \n",
       "3    55    1   1       120        1      178      0      0.8      2   0     2   \n",
       "4    62    0   0       120        1      163      1      0.6      2   0     2   \n",
       "5    58    1   0       140        1      148      0      0.4      1   0     1   \n",
       "6    55    0   1       140        0      153      0      1.3      1   0     2   \n",
       "7    34    1   1       120        1      173      0      0.0      2   0     3   \n",
       "8    50    1   2       172        1      162      0      0.5      2   0     3   \n",
       "9    59    1   2       150        1      174      0      1.6      2   0     2   \n",
       "10   52    1   0       140        1      160      0      1.2      2   0     2   \n",
       "11   42    0   2       130        1      139      0      0.2      2   0     2   \n",
       "12   54    1   1       130        1      171      0      0.6      2   0     2   \n",
       "\n",
       "    target  PREDICCIÓN  \n",
       "0      NaN         1.0  \n",
       "1      NaN         1.0  \n",
       "2      NaN         1.0  \n",
       "3      NaN         1.0  \n",
       "4      NaN         1.0  \n",
       "5      NaN         1.0  \n",
       "6      NaN         1.0  \n",
       "7      NaN         1.0  \n",
       "8      NaN         1.0  \n",
       "9      NaN         1.0  \n",
       "10     NaN         1.0  \n",
       "11     NaN         1.0  \n",
       "12     NaN         1.0  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pred = pd.read_excel('./datasets/Heart_diagnosis.xlsx', sheet_name='Datos_Predecir')\n",
    "df_pred = df_pred[df.columns]\n",
    "X_pred = df_pred.drop('target', axis=1)\n",
    "X_pred = scaler.transform(X_pred)\n",
    "resultado = np.round(model.predict(X_pred))\n",
    "df_pred['PREDICCIÓN'] = pd.DataFrame(resultado)\n",
    "df_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Guardar modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('modelo_pred_heart.h5')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.2 64-bit",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.2"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "26de051ba29f2982a8de78e945f0abaf191376122a1563185a90213a26c5da77"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}