Complete Tensorflow 2 and Keras Deep Learning Bootcamp - JP/03-ANNs/02-Keras-Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Keras TF 2.0 - Classification Project\n",
    "\n",
    "There are 3 models that we will be demostrating.\n",
    "\n",
    "+ [Example One: Choosing too many epochs and overfitting!](#overfit)\n",
    "+ [Example Two: Early Stopping](#earlystop)\n",
    "+ [Example Three: Adding in DropOut Layers](#dropout)\n",
    "\n",
    "----\n",
    "\n",
    "## The Data\n",
    "\n",
    "### Breast cancer wisconsin (diagnostic) dataset\n",
    "--------------------------------------------\n",
    "\n",
    "**Data Set Characteristics:**\n",
    "\n",
    "    :Number of Instances: 569\n",
    "\n",
    "    :Number of Attributes: 30 numeric, predictive attributes and the class\n",
    "\n",
    "    :Attribute Information:\n",
    "        - radius (mean of distances from center to points on the perimeter)\n",
    "        - texture (standard deviation of gray-scale values)\n",
    "        - perimeter\n",
    "        - area\n",
    "        - smoothness (local variation in radius lengths)\n",
    "        - compactness (perimeter^2 / area - 1.0)\n",
    "        - concavity (severity of concave portions of the contour)\n",
    "        - concave points (number of concave portions of the contour)\n",
    "        - symmetry \n",
    "        - fractal dimension (\"coastline approximation\" - 1)\n",
    "\n",
    "        The mean, standard error, and \"worst\" or largest (mean of the three\n",
    "        largest values) of these features were computed for each image,\n",
    "        resulting in 30 features.  For instance, field 3 is Mean Radius, field\n",
    "        13 is Radius SE, field 23 is Worst Radius.\n",
    "\n",
    "        - class:\n",
    "                - WDBC-Malignant\n",
    "                - WDBC-Benign\n",
    "\n",
    "    :Summary Statistics:\n",
    "\n",
    "    ===================================== ====== ======\n",
    "                                           Min    Max\n",
    "    ===================================== ====== ======\n",
    "    radius (mean):                        6.981  28.11\n",
    "    texture (mean):                       9.71   39.28\n",
    "    perimeter (mean):                     43.79  188.5\n",
    "    area (mean):                          143.5  2501.0\n",
    "    smoothness (mean):                    0.053  0.163\n",
    "    compactness (mean):                   0.019  0.345\n",
    "    concavity (mean):                     0.0    0.427\n",
    "    concave points (mean):                0.0    0.201\n",
    "    symmetry (mean):                      0.106  0.304\n",
    "    fractal dimension (mean):             0.05   0.097\n",
    "    radius (standard error):              0.112  2.873\n",
    "    texture (standard error):             0.36   4.885\n",
    "    perimeter (standard error):           0.757  21.98\n",
    "    area (standard error):                6.802  542.2\n",
    "    smoothness (standard error):          0.002  0.031\n",
    "    compactness (standard error):         0.002  0.135\n",
    "    concavity (standard error):           0.0    0.396\n",
    "    concave points (standard error):      0.0    0.053\n",
    "    symmetry (standard error):            0.008  0.079\n",
    "    fractal dimension (standard error):   0.001  0.03\n",
    "    radius (worst):                       7.93   36.04\n",
    "    texture (worst):                      12.02  49.54\n",
    "    perimeter (worst):                    50.41  251.2\n",
    "    area (worst):                         185.2  4254.0\n",
    "    smoothness (worst):                   0.071  0.223\n",
    "    compactness (worst):                  0.027  1.058\n",
    "    concavity (worst):                    0.0    1.252\n",
    "    concave points (worst):               0.0    0.291\n",
    "    symmetry (worst):                     0.156  0.664\n",
    "    fractal dimension (worst):            0.055  0.208\n",
    "    ===================================== ====== ======\n",
    "\n",
    "    :Missing Attribute Values: None\n",
    "\n",
    "    :Class Distribution: 212 - Malignant, 357 - Benign\n",
    "\n",
    "    :Creator:  Dr. William H. Wolberg, W. Nick Street, Olvi L. Mangasarian\n",
    "\n",
    "    :Donor: Nick Street\n",
    "\n",
    "    :Date: November, 1995\n",
    "\n",
    "This is a copy of UCI ML Breast Cancer Wisconsin (Diagnostic) datasets.\n",
    "https://goo.gl/U2Uwz2\n",
    "\n",
    "Features are computed from a digitized image of a fine needle\n",
    "aspirate (FNA) of a breast mass.  They describe\n",
    "characteristics of the cell nuclei present in the image.\n",
    "\n",
    "Separating plane described above was obtained using\n",
    "Multisurface Method-Tree (MSM-T) [K. P. Bennett, \"Decision Tree\n",
    "Construction Via Linear Programming.\" Proceedings of the 4th\n",
    "Midwest Artificial Intelligence and Cognitive Science Society,\n",
    "pp. 97-101, 1992], a classification method which uses linear\n",
    "programming to construct a decision tree.  Relevant features\n",
    "were selected using an exhaustive search in the space of 1-4\n",
    "features and 1-3 separating planes.\n",
    "\n",
    "The actual linear program used to obtain the separating plane\n",
    "in the 3-dimensional space is that described in:\n",
    "[K. P. Bennett and O. L. Mangasarian: \"Robust Linear\n",
    "Programming Discrimination of Two Linearly Inseparable Sets\",\n",
    "Optimization Methods and Software 1, 1992, 23-34].\n",
    "\n",
    "This database is also available through the UW CS ftp server:\n",
    "\n",
    "ftp ftp.cs.wisc.edu\n",
    "cd math-prog/cpo-dataset/machine-learn/WDBC/\n",
    "\n",
    ".. topic:: References\n",
    "\n",
    "   - W.N. Street, W.H. Wolberg and O.L. Mangasarian. Nuclear feature extraction \n",
    "     for breast tumor diagnosis. IS&T/SPIE 1993 International Symposium on \n",
    "     Electronic Imaging: Science and Technology, volume 1905, pages 861-870,\n",
    "     San Jose, CA, 1993.\n",
    "   - O.L. Mangasarian, W.N. Street and W.H. Wolberg. Breast cancer diagnosis and \n",
    "     prognosis via linear programming. Operations Research, 43(4), pages 570-577, \n",
    "     July-August 1995.\n",
    "   - W.H. Wolberg, W.N. Street, and O.L. Mangasarian. Machine learning techniques\n",
    "     to diagnose breast cancer from fine-needle aspirates. Cancer Letters 77 (1994) \n",
    "     163-171."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/cancer_classification.csv')"
   ]
  },
  {
   "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>mean radius</th>\n",
       "      <th>mean texture</th>\n",
       "      <th>mean perimeter</th>\n",
       "      <th>mean area</th>\n",
       "      <th>mean smoothness</th>\n",
       "      <th>mean compactness</th>\n",
       "      <th>mean concavity</th>\n",
       "      <th>mean concave points</th>\n",
       "      <th>mean symmetry</th>\n",
       "      <th>mean fractal dimension</th>\n",
       "      <th>...</th>\n",
       "      <th>worst texture</th>\n",
       "      <th>worst perimeter</th>\n",
       "      <th>worst area</th>\n",
       "      <th>worst smoothness</th>\n",
       "      <th>worst compactness</th>\n",
       "      <th>worst concavity</th>\n",
       "      <th>worst concave points</th>\n",
       "      <th>worst symmetry</th>\n",
       "      <th>worst fractal dimension</th>\n",
       "      <th>benign_0__mal_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17.99</td>\n",
       "      <td>10.38</td>\n",
       "      <td>122.8</td>\n",
       "      <td>1001.0</td>\n",
       "      <td>0.11840</td>\n",
       "      <td>0.27760</td>\n",
       "      <td>0.3001</td>\n",
       "      <td>0.14710</td>\n",
       "      <td>0.2419</td>\n",
       "      <td>0.07871</td>\n",
       "      <td>...</td>\n",
       "      <td>17.33</td>\n",
       "      <td>184.6</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>0.1622</td>\n",
       "      <td>0.6656</td>\n",
       "      <td>0.7119</td>\n",
       "      <td>0.2654</td>\n",
       "      <td>0.4601</td>\n",
       "      <td>0.11890</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20.57</td>\n",
       "      <td>17.77</td>\n",
       "      <td>132.9</td>\n",
       "      <td>1326.0</td>\n",
       "      <td>0.08474</td>\n",
       "      <td>0.07864</td>\n",
       "      <td>0.0869</td>\n",
       "      <td>0.07017</td>\n",
       "      <td>0.1812</td>\n",
       "      <td>0.05667</td>\n",
       "      <td>...</td>\n",
       "      <td>23.41</td>\n",
       "      <td>158.8</td>\n",
       "      <td>1956.0</td>\n",
       "      <td>0.1238</td>\n",
       "      <td>0.1866</td>\n",
       "      <td>0.2416</td>\n",
       "      <td>0.1860</td>\n",
       "      <td>0.2750</td>\n",
       "      <td>0.08902</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean radius  mean texture  mean perimeter  mean area  mean smoothness  \\\n",
       "0        17.99         10.38           122.8     1001.0          0.11840   \n",
       "1        20.57         17.77           132.9     1326.0          0.08474   \n",
       "\n",
       "   mean compactness  mean concavity  mean concave points  mean symmetry  \\\n",
       "0           0.27760          0.3001              0.14710         0.2419   \n",
       "1           0.07864          0.0869              0.07017         0.1812   \n",
       "\n",
       "   mean fractal dimension  ...  worst texture  worst perimeter  worst area  \\\n",
       "0                 0.07871  ...          17.33            184.6      2019.0   \n",
       "1                 0.05667  ...          23.41            158.8      1956.0   \n",
       "\n",
       "   worst smoothness  worst compactness  worst concavity  worst concave points  \\\n",
       "0            0.1622             0.6656           0.7119                0.2654   \n",
       "1            0.1238             0.1866           0.2416                0.1860   \n",
       "\n",
       "   worst symmetry  worst fractal dimension  benign_0__mal_1  \n",
       "0          0.4601                  0.11890                0  \n",
       "1          0.2750                  0.08902                0  \n",
       "\n",
       "[2 rows x 31 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 569 entries, 0 to 568\n",
      "Data columns (total 31 columns):\n",
      " #   Column                   Non-Null Count  Dtype  \n",
      "---  ------                   --------------  -----  \n",
      " 0   mean radius              569 non-null    float64\n",
      " 1   mean texture             569 non-null    float64\n",
      " 2   mean perimeter           569 non-null    float64\n",
      " 3   mean area                569 non-null    float64\n",
      " 4   mean smoothness          569 non-null    float64\n",
      " 5   mean compactness         569 non-null    float64\n",
      " 6   mean concavity           569 non-null    float64\n",
      " 7   mean concave points      569 non-null    float64\n",
      " 8   mean symmetry            569 non-null    float64\n",
      " 9   mean fractal dimension   569 non-null    float64\n",
      " 10  radius error             569 non-null    float64\n",
      " 11  texture error            569 non-null    float64\n",
      " 12  perimeter error          569 non-null    float64\n",
      " 13  area error               569 non-null    float64\n",
      " 14  smoothness error         569 non-null    float64\n",
      " 15  compactness error        569 non-null    float64\n",
      " 16  concavity error          569 non-null    float64\n",
      " 17  concave points error     569 non-null    float64\n",
      " 18  symmetry error           569 non-null    float64\n",
      " 19  fractal dimension error  569 non-null    float64\n",
      " 20  worst radius             569 non-null    float64\n",
      " 21  worst texture            569 non-null    float64\n",
      " 22  worst perimeter          569 non-null    float64\n",
      " 23  worst area               569 non-null    float64\n",
      " 24  worst smoothness         569 non-null    float64\n",
      " 25  worst compactness        569 non-null    float64\n",
      " 26  worst concavity          569 non-null    float64\n",
      " 27  worst concave points     569 non-null    float64\n",
      " 28  worst symmetry           569 non-null    float64\n",
      " 29  worst fractal dimension  569 non-null    float64\n",
      " 30  benign_0__mal_1          569 non-null    int64  \n",
      "dtypes: float64(30), int64(1)\n",
      "memory usage: 137.9 KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>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>mean radius</th>\n",
       "      <td>569.0</td>\n",
       "      <td>14.127292</td>\n",
       "      <td>3.524049</td>\n",
       "      <td>6.981000</td>\n",
       "      <td>11.700000</td>\n",
       "      <td>13.370000</td>\n",
       "      <td>15.780000</td>\n",
       "      <td>28.11000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean texture</th>\n",
       "      <td>569.0</td>\n",
       "      <td>19.289649</td>\n",
       "      <td>4.301036</td>\n",
       "      <td>9.710000</td>\n",
       "      <td>16.170000</td>\n",
       "      <td>18.840000</td>\n",
       "      <td>21.800000</td>\n",
       "      <td>39.28000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean perimeter</th>\n",
       "      <td>569.0</td>\n",
       "      <td>91.969033</td>\n",
       "      <td>24.298981</td>\n",
       "      <td>43.790000</td>\n",
       "      <td>75.170000</td>\n",
       "      <td>86.240000</td>\n",
       "      <td>104.100000</td>\n",
       "      <td>188.50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean area</th>\n",
       "      <td>569.0</td>\n",
       "      <td>654.889104</td>\n",
       "      <td>351.914129</td>\n",
       "      <td>143.500000</td>\n",
       "      <td>420.300000</td>\n",
       "      <td>551.100000</td>\n",
       "      <td>782.700000</td>\n",
       "      <td>2501.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean smoothness</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.096360</td>\n",
       "      <td>0.014064</td>\n",
       "      <td>0.052630</td>\n",
       "      <td>0.086370</td>\n",
       "      <td>0.095870</td>\n",
       "      <td>0.105300</td>\n",
       "      <td>0.16340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean compactness</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.104341</td>\n",
       "      <td>0.052813</td>\n",
       "      <td>0.019380</td>\n",
       "      <td>0.064920</td>\n",
       "      <td>0.092630</td>\n",
       "      <td>0.130400</td>\n",
       "      <td>0.34540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean concavity</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.088799</td>\n",
       "      <td>0.079720</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029560</td>\n",
       "      <td>0.061540</td>\n",
       "      <td>0.130700</td>\n",
       "      <td>0.42680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean concave points</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.048919</td>\n",
       "      <td>0.038803</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.020310</td>\n",
       "      <td>0.033500</td>\n",
       "      <td>0.074000</td>\n",
       "      <td>0.20120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean symmetry</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.181162</td>\n",
       "      <td>0.027414</td>\n",
       "      <td>0.106000</td>\n",
       "      <td>0.161900</td>\n",
       "      <td>0.179200</td>\n",
       "      <td>0.195700</td>\n",
       "      <td>0.30400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean fractal dimension</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.062798</td>\n",
       "      <td>0.007060</td>\n",
       "      <td>0.049960</td>\n",
       "      <td>0.057700</td>\n",
       "      <td>0.061540</td>\n",
       "      <td>0.066120</td>\n",
       "      <td>0.09744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>radius error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.405172</td>\n",
       "      <td>0.277313</td>\n",
       "      <td>0.111500</td>\n",
       "      <td>0.232400</td>\n",
       "      <td>0.324200</td>\n",
       "      <td>0.478900</td>\n",
       "      <td>2.87300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>texture error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>1.216853</td>\n",
       "      <td>0.551648</td>\n",
       "      <td>0.360200</td>\n",
       "      <td>0.833900</td>\n",
       "      <td>1.108000</td>\n",
       "      <td>1.474000</td>\n",
       "      <td>4.88500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perimeter error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>2.866059</td>\n",
       "      <td>2.021855</td>\n",
       "      <td>0.757000</td>\n",
       "      <td>1.606000</td>\n",
       "      <td>2.287000</td>\n",
       "      <td>3.357000</td>\n",
       "      <td>21.98000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>area error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>40.337079</td>\n",
       "      <td>45.491006</td>\n",
       "      <td>6.802000</td>\n",
       "      <td>17.850000</td>\n",
       "      <td>24.530000</td>\n",
       "      <td>45.190000</td>\n",
       "      <td>542.20000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smoothness error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.007041</td>\n",
       "      <td>0.003003</td>\n",
       "      <td>0.001713</td>\n",
       "      <td>0.005169</td>\n",
       "      <td>0.006380</td>\n",
       "      <td>0.008146</td>\n",
       "      <td>0.03113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>compactness error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.025478</td>\n",
       "      <td>0.017908</td>\n",
       "      <td>0.002252</td>\n",
       "      <td>0.013080</td>\n",
       "      <td>0.020450</td>\n",
       "      <td>0.032450</td>\n",
       "      <td>0.13540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>concavity error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.031894</td>\n",
       "      <td>0.030186</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.015090</td>\n",
       "      <td>0.025890</td>\n",
       "      <td>0.042050</td>\n",
       "      <td>0.39600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>concave points error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.011796</td>\n",
       "      <td>0.006170</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007638</td>\n",
       "      <td>0.010930</td>\n",
       "      <td>0.014710</td>\n",
       "      <td>0.05279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>symmetry error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.020542</td>\n",
       "      <td>0.008266</td>\n",
       "      <td>0.007882</td>\n",
       "      <td>0.015160</td>\n",
       "      <td>0.018730</td>\n",
       "      <td>0.023480</td>\n",
       "      <td>0.07895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fractal dimension error</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.003795</td>\n",
       "      <td>0.002646</td>\n",
       "      <td>0.000895</td>\n",
       "      <td>0.002248</td>\n",
       "      <td>0.003187</td>\n",
       "      <td>0.004558</td>\n",
       "      <td>0.02984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst radius</th>\n",
       "      <td>569.0</td>\n",
       "      <td>16.269190</td>\n",
       "      <td>4.833242</td>\n",
       "      <td>7.930000</td>\n",
       "      <td>13.010000</td>\n",
       "      <td>14.970000</td>\n",
       "      <td>18.790000</td>\n",
       "      <td>36.04000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst texture</th>\n",
       "      <td>569.0</td>\n",
       "      <td>25.677223</td>\n",
       "      <td>6.146258</td>\n",
       "      <td>12.020000</td>\n",
       "      <td>21.080000</td>\n",
       "      <td>25.410000</td>\n",
       "      <td>29.720000</td>\n",
       "      <td>49.54000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst perimeter</th>\n",
       "      <td>569.0</td>\n",
       "      <td>107.261213</td>\n",
       "      <td>33.602542</td>\n",
       "      <td>50.410000</td>\n",
       "      <td>84.110000</td>\n",
       "      <td>97.660000</td>\n",
       "      <td>125.400000</td>\n",
       "      <td>251.20000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst area</th>\n",
       "      <td>569.0</td>\n",
       "      <td>880.583128</td>\n",
       "      <td>569.356993</td>\n",
       "      <td>185.200000</td>\n",
       "      <td>515.300000</td>\n",
       "      <td>686.500000</td>\n",
       "      <td>1084.000000</td>\n",
       "      <td>4254.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst smoothness</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.132369</td>\n",
       "      <td>0.022832</td>\n",
       "      <td>0.071170</td>\n",
       "      <td>0.116600</td>\n",
       "      <td>0.131300</td>\n",
       "      <td>0.146000</td>\n",
       "      <td>0.22260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst compactness</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.254265</td>\n",
       "      <td>0.157336</td>\n",
       "      <td>0.027290</td>\n",
       "      <td>0.147200</td>\n",
       "      <td>0.211900</td>\n",
       "      <td>0.339100</td>\n",
       "      <td>1.05800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst concavity</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.272188</td>\n",
       "      <td>0.208624</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.114500</td>\n",
       "      <td>0.226700</td>\n",
       "      <td>0.382900</td>\n",
       "      <td>1.25200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst concave points</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.114606</td>\n",
       "      <td>0.065732</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.064930</td>\n",
       "      <td>0.099930</td>\n",
       "      <td>0.161400</td>\n",
       "      <td>0.29100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst symmetry</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.290076</td>\n",
       "      <td>0.061867</td>\n",
       "      <td>0.156500</td>\n",
       "      <td>0.250400</td>\n",
       "      <td>0.282200</td>\n",
       "      <td>0.317900</td>\n",
       "      <td>0.66380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worst fractal dimension</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.083946</td>\n",
       "      <td>0.018061</td>\n",
       "      <td>0.055040</td>\n",
       "      <td>0.071460</td>\n",
       "      <td>0.080040</td>\n",
       "      <td>0.092080</td>\n",
       "      <td>0.20750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>benign_0__mal_1</th>\n",
       "      <td>569.0</td>\n",
       "      <td>0.627417</td>\n",
       "      <td>0.483918</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         count        mean         std         min  \\\n",
       "mean radius              569.0   14.127292    3.524049    6.981000   \n",
       "mean texture             569.0   19.289649    4.301036    9.710000   \n",
       "mean perimeter           569.0   91.969033   24.298981   43.790000   \n",
       "mean area                569.0  654.889104  351.914129  143.500000   \n",
       "mean smoothness          569.0    0.096360    0.014064    0.052630   \n",
       "mean compactness         569.0    0.104341    0.052813    0.019380   \n",
       "mean concavity           569.0    0.088799    0.079720    0.000000   \n",
       "mean concave points      569.0    0.048919    0.038803    0.000000   \n",
       "mean symmetry            569.0    0.181162    0.027414    0.106000   \n",
       "mean fractal dimension   569.0    0.062798    0.007060    0.049960   \n",
       "radius error             569.0    0.405172    0.277313    0.111500   \n",
       "texture error            569.0    1.216853    0.551648    0.360200   \n",
       "perimeter error          569.0    2.866059    2.021855    0.757000   \n",
       "area error               569.0   40.337079   45.491006    6.802000   \n",
       "smoothness error         569.0    0.007041    0.003003    0.001713   \n",
       "compactness error        569.0    0.025478    0.017908    0.002252   \n",
       "concavity error          569.0    0.031894    0.030186    0.000000   \n",
       "concave points error     569.0    0.011796    0.006170    0.000000   \n",
       "symmetry error           569.0    0.020542    0.008266    0.007882   \n",
       "fractal dimension error  569.0    0.003795    0.002646    0.000895   \n",
       "worst radius             569.0   16.269190    4.833242    7.930000   \n",
       "worst texture            569.0   25.677223    6.146258   12.020000   \n",
       "worst perimeter          569.0  107.261213   33.602542   50.410000   \n",
       "worst area               569.0  880.583128  569.356993  185.200000   \n",
       "worst smoothness         569.0    0.132369    0.022832    0.071170   \n",
       "worst compactness        569.0    0.254265    0.157336    0.027290   \n",
       "worst concavity          569.0    0.272188    0.208624    0.000000   \n",
       "worst concave points     569.0    0.114606    0.065732    0.000000   \n",
       "worst symmetry           569.0    0.290076    0.061867    0.156500   \n",
       "worst fractal dimension  569.0    0.083946    0.018061    0.055040   \n",
       "benign_0__mal_1          569.0    0.627417    0.483918    0.000000   \n",
       "\n",
       "                                25%         50%          75%         max  \n",
       "mean radius               11.700000   13.370000    15.780000    28.11000  \n",
       "mean texture              16.170000   18.840000    21.800000    39.28000  \n",
       "mean perimeter            75.170000   86.240000   104.100000   188.50000  \n",
       "mean area                420.300000  551.100000   782.700000  2501.00000  \n",
       "mean smoothness            0.086370    0.095870     0.105300     0.16340  \n",
       "mean compactness           0.064920    0.092630     0.130400     0.34540  \n",
       "mean concavity             0.029560    0.061540     0.130700     0.42680  \n",
       "mean concave points        0.020310    0.033500     0.074000     0.20120  \n",
       "mean symmetry              0.161900    0.179200     0.195700     0.30400  \n",
       "mean fractal dimension     0.057700    0.061540     0.066120     0.09744  \n",
       "radius error               0.232400    0.324200     0.478900     2.87300  \n",
       "texture error              0.833900    1.108000     1.474000     4.88500  \n",
       "perimeter error            1.606000    2.287000     3.357000    21.98000  \n",
       "area error                17.850000   24.530000    45.190000   542.20000  \n",
       "smoothness error           0.005169    0.006380     0.008146     0.03113  \n",
       "compactness error          0.013080    0.020450     0.032450     0.13540  \n",
       "concavity error            0.015090    0.025890     0.042050     0.39600  \n",
       "concave points error       0.007638    0.010930     0.014710     0.05279  \n",
       "symmetry error             0.015160    0.018730     0.023480     0.07895  \n",
       "fractal dimension error    0.002248    0.003187     0.004558     0.02984  \n",
       "worst radius              13.010000   14.970000    18.790000    36.04000  \n",
       "worst texture             21.080000   25.410000    29.720000    49.54000  \n",
       "worst perimeter           84.110000   97.660000   125.400000   251.20000  \n",
       "worst area               515.300000  686.500000  1084.000000  4254.00000  \n",
       "worst smoothness           0.116600    0.131300     0.146000     0.22260  \n",
       "worst compactness          0.147200    0.211900     0.339100     1.05800  \n",
       "worst concavity            0.114500    0.226700     0.382900     1.25200  \n",
       "worst concave points       0.064930    0.099930     0.161400     0.29100  \n",
       "worst symmetry             0.250400    0.282200     0.317900     0.66380  \n",
       "worst fractal dimension    0.071460    0.080040     0.092080     0.20750  \n",
       "benign_0__mal_1            0.000000    1.000000     1.000000     1.00000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=df,x='benign_0__mal_1');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Check the correlation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "benign_0__mal_1            1.000000\n",
       "smoothness error           0.067016\n",
       "mean fractal dimension     0.012838\n",
       "texture error              0.008303\n",
       "symmetry error             0.006522\n",
       "fractal dimension error   -0.077972\n",
       "concavity error           -0.253730\n",
       "compactness error         -0.292999\n",
       "worst fractal dimension   -0.323872\n",
       "mean symmetry             -0.330499\n",
       "mean smoothness           -0.358560\n",
       "concave points error      -0.408042\n",
       "mean texture              -0.415185\n",
       "worst symmetry            -0.416294\n",
       "worst smoothness          -0.421465\n",
       "worst texture             -0.456903\n",
       "area error                -0.548236\n",
       "perimeter error           -0.556141\n",
       "radius error              -0.567134\n",
       "worst compactness         -0.590998\n",
       "mean compactness          -0.596534\n",
       "worst concavity           -0.659610\n",
       "mean concavity            -0.696360\n",
       "mean area                 -0.708984\n",
       "mean radius               -0.730029\n",
       "worst area                -0.733825\n",
       "mean perimeter            -0.742636\n",
       "worst radius              -0.776454\n",
       "mean concave points       -0.776614\n",
       "worst perimeter           -0.782914\n",
       "worst concave points      -0.793566\n",
       "Name: benign_0__mal_1, dtype: float64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()['benign_0__mal_1'].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "df.corr()['benign_0__mal_1'].sort_values()[:-1].plot(kind='bar');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "sns.heatmap(df.corr());"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop('benign_0__mal_1', axis=1)\n",
    "y = df['benign_0__mal_1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=101)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Scaling Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Modelling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(426, 30)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape # we have 30 features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.add(Dense(30, activation='relu'))\n",
    "model.add(Dense(15, activation='relu'))\n",
    "\n",
    "# Binary Classification => Sigmoid\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training the Model \n",
    "\n",
    "# <a name='overfit'>Example One: Choosing too many epochs and overfitting!</a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "14/14 [==============================] - 0s 34ms/step - loss: 0.6765 - val_loss: 0.6549\n",
      "Epoch 2/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6360 - val_loss: 0.6144\n",
      "Epoch 3/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.5950 - val_loss: 0.5718\n",
      "Epoch 4/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.5503 - val_loss: 0.5261\n",
      "Epoch 5/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.5079 - val_loss: 0.4840\n",
      "Epoch 6/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.4640 - val_loss: 0.4394\n",
      "Epoch 7/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.4211 - val_loss: 0.3976\n",
      "Epoch 8/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3816 - val_loss: 0.3587\n",
      "Epoch 9/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3477 - val_loss: 0.3245\n",
      "Epoch 10/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.3151 - val_loss: 0.2907\n",
      "Epoch 11/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2861 - val_loss: 0.2624\n",
      "Epoch 12/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2616 - val_loss: 0.2374\n",
      "Epoch 13/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2385 - val_loss: 0.2154\n",
      "Epoch 14/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2172 - val_loss: 0.1967\n",
      "Epoch 15/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2009 - val_loss: 0.1798\n",
      "Epoch 16/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1881 - val_loss: 0.1759\n",
      "Epoch 17/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1748 - val_loss: 0.1567\n",
      "Epoch 18/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1636 - val_loss: 0.1503\n",
      "Epoch 19/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1541 - val_loss: 0.1449\n",
      "Epoch 20/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1483 - val_loss: 0.1399\n",
      "Epoch 21/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1403 - val_loss: 0.1340\n",
      "Epoch 22/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1333 - val_loss: 0.1277\n",
      "Epoch 23/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1288 - val_loss: 0.1245\n",
      "Epoch 24/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1289 - val_loss: 0.1200\n",
      "Epoch 25/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1179 - val_loss: 0.1221\n",
      "Epoch 26/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1108 - val_loss: 0.1153\n",
      "Epoch 27/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1073 - val_loss: 0.1142\n",
      "Epoch 28/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1044 - val_loss: 0.1126\n",
      "Epoch 29/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1011 - val_loss: 0.1099\n",
      "Epoch 30/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0965 - val_loss: 0.1098\n",
      "Epoch 31/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0942 - val_loss: 0.1082\n",
      "Epoch 32/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0923 - val_loss: 0.1094\n",
      "Epoch 33/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0881 - val_loss: 0.1062\n",
      "Epoch 34/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0869 - val_loss: 0.1077\n",
      "Epoch 35/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0844 - val_loss: 0.1073\n",
      "Epoch 36/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0858 - val_loss: 0.1063\n",
      "Epoch 37/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0798 - val_loss: 0.1055\n",
      "Epoch 38/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0777 - val_loss: 0.1062\n",
      "Epoch 39/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0770 - val_loss: 0.1039\n",
      "Epoch 40/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0762 - val_loss: 0.1041\n",
      "Epoch 41/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0748 - val_loss: 0.1056\n",
      "Epoch 42/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0758 - val_loss: 0.1032\n",
      "Epoch 43/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0704 - val_loss: 0.1108\n",
      "Epoch 44/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0717 - val_loss: 0.1015\n",
      "Epoch 45/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0707 - val_loss: 0.1041\n",
      "Epoch 46/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0681 - val_loss: 0.1031\n",
      "Epoch 47/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0691 - val_loss: 0.1049\n",
      "Epoch 48/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0683 - val_loss: 0.1099\n",
      "Epoch 49/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0671 - val_loss: 0.1019\n",
      "Epoch 50/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0679 - val_loss: 0.1043\n",
      "Epoch 51/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0657 - val_loss: 0.1114\n",
      "Epoch 52/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0686 - val_loss: 0.1036\n",
      "Epoch 53/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0630 - val_loss: 0.1081\n",
      "Epoch 54/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0626 - val_loss: 0.1042\n",
      "Epoch 55/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0636 - val_loss: 0.1076\n",
      "Epoch 56/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0638 - val_loss: 0.1020\n",
      "Epoch 57/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0592 - val_loss: 0.1180\n",
      "Epoch 58/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0619 - val_loss: 0.1064\n",
      "Epoch 59/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0596 - val_loss: 0.1092\n",
      "Epoch 60/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0630 - val_loss: 0.1043\n",
      "Epoch 61/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0654 - val_loss: 0.1072\n",
      "Epoch 62/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0609 - val_loss: 0.1076\n",
      "Epoch 63/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0580 - val_loss: 0.1068\n",
      "Epoch 64/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0584 - val_loss: 0.1079\n",
      "Epoch 65/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0599 - val_loss: 0.1053\n",
      "Epoch 66/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0567 - val_loss: 0.1065\n",
      "Epoch 67/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0555 - val_loss: 0.1094\n",
      "Epoch 68/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0560 - val_loss: 0.1085\n",
      "Epoch 69/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0586 - val_loss: 0.1064\n",
      "Epoch 70/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0550 - val_loss: 0.1148\n",
      "Epoch 71/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0544 - val_loss: 0.1054\n",
      "Epoch 72/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0569 - val_loss: 0.1101\n",
      "Epoch 73/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0536 - val_loss: 0.1066\n",
      "Epoch 74/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0540 - val_loss: 0.1221\n",
      "Epoch 75/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0544 - val_loss: 0.1098\n",
      "Epoch 76/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0541 - val_loss: 0.1088\n",
      "Epoch 77/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0536 - val_loss: 0.1137\n",
      "Epoch 78/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0538 - val_loss: 0.1098\n",
      "Epoch 79/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0530 - val_loss: 0.1122\n",
      "Epoch 80/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0540 - val_loss: 0.1079\n",
      "Epoch 81/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0521 - val_loss: 0.1142\n",
      "Epoch 82/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0510 - val_loss: 0.1126\n",
      "Epoch 83/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 9ms/step - loss: 0.0514 - val_loss: 0.1120\n",
      "Epoch 84/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0518 - val_loss: 0.1195\n",
      "Epoch 85/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0535 - val_loss: 0.1091\n",
      "Epoch 86/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0564 - val_loss: 0.1240\n",
      "Epoch 87/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0531 - val_loss: 0.1158\n",
      "Epoch 88/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0505 - val_loss: 0.1172\n",
      "Epoch 89/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0509 - val_loss: 0.1106\n",
      "Epoch 90/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0515 - val_loss: 0.1188\n",
      "Epoch 91/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0493 - val_loss: 0.1120\n",
      "Epoch 92/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0515 - val_loss: 0.1175\n",
      "Epoch 93/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0501 - val_loss: 0.1186\n",
      "Epoch 94/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0491 - val_loss: 0.1129\n",
      "Epoch 95/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0507 - val_loss: 0.1179\n",
      "Epoch 96/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0502 - val_loss: 0.1189\n",
      "Epoch 97/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0490 - val_loss: 0.1186\n",
      "Epoch 98/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0484 - val_loss: 0.1171\n",
      "Epoch 99/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0499 - val_loss: 0.1191\n",
      "Epoch 100/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0493 - val_loss: 0.1144\n",
      "Epoch 101/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0486 - val_loss: 0.1177\n",
      "Epoch 102/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0536 - val_loss: 0.1167\n",
      "Epoch 103/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0520 - val_loss: 0.1237\n",
      "Epoch 104/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0526 - val_loss: 0.1172\n",
      "Epoch 105/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0564 - val_loss: 0.1222\n",
      "Epoch 106/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0473 - val_loss: 0.1130\n",
      "Epoch 107/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0509 - val_loss: 0.1235\n",
      "Epoch 108/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0463 - val_loss: 0.1204\n",
      "Epoch 109/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0483 - val_loss: 0.1227\n",
      "Epoch 110/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0474 - val_loss: 0.1199\n",
      "Epoch 111/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0466 - val_loss: 0.1193\n",
      "Epoch 112/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0457 - val_loss: 0.1235\n",
      "Epoch 113/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0473 - val_loss: 0.1259\n",
      "Epoch 114/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0482 - val_loss: 0.1257\n",
      "Epoch 115/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0456 - val_loss: 0.1227\n",
      "Epoch 116/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0447 - val_loss: 0.1216\n",
      "Epoch 117/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0463 - val_loss: 0.1204\n",
      "Epoch 118/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0466 - val_loss: 0.1223\n",
      "Epoch 119/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0527 - val_loss: 0.1343\n",
      "Epoch 120/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0550 - val_loss: 0.1169\n",
      "Epoch 121/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0457 - val_loss: 0.1323\n",
      "Epoch 122/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0451 - val_loss: 0.1233\n",
      "Epoch 123/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0454 - val_loss: 0.1196\n",
      "Epoch 124/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0447 - val_loss: 0.1307\n",
      "Epoch 125/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0436 - val_loss: 0.1245\n",
      "Epoch 126/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0435 - val_loss: 0.1332\n",
      "Epoch 127/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0444 - val_loss: 0.1287\n",
      "Epoch 128/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0480 - val_loss: 0.1270\n",
      "Epoch 129/600\n",
      "14/14 [==============================] - 0s 8ms/step - loss: 0.0438 - val_loss: 0.1324\n",
      "Epoch 130/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0443 - val_loss: 0.1255\n",
      "Epoch 131/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0428 - val_loss: 0.1249\n",
      "Epoch 132/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0425 - val_loss: 0.1307\n",
      "Epoch 133/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0437 - val_loss: 0.1255\n",
      "Epoch 134/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0443 - val_loss: 0.1393\n",
      "Epoch 135/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0436 - val_loss: 0.1235\n",
      "Epoch 136/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0424 - val_loss: 0.1252\n",
      "Epoch 137/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0458 - val_loss: 0.1315\n",
      "Epoch 138/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0435 - val_loss: 0.1279\n",
      "Epoch 139/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0472 - val_loss: 0.1207\n",
      "Epoch 140/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0453 - val_loss: 0.1394\n",
      "Epoch 141/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0426 - val_loss: 0.1272\n",
      "Epoch 142/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0422 - val_loss: 0.1402\n",
      "Epoch 143/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0457 - val_loss: 0.1219\n",
      "Epoch 144/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0444 - val_loss: 0.1349\n",
      "Epoch 145/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0416 - val_loss: 0.1241\n",
      "Epoch 146/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0422 - val_loss: 0.1292\n",
      "Epoch 147/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0402 - val_loss: 0.1337\n",
      "Epoch 148/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0411 - val_loss: 0.1327\n",
      "Epoch 149/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0401 - val_loss: 0.1292\n",
      "Epoch 150/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0400 - val_loss: 0.1347\n",
      "Epoch 151/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0409 - val_loss: 0.1284\n",
      "Epoch 152/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0399 - val_loss: 0.1277\n",
      "Epoch 153/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0404 - val_loss: 0.1329\n",
      "Epoch 154/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0403 - val_loss: 0.1324\n",
      "Epoch 155/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0408 - val_loss: 0.1301\n",
      "Epoch 156/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0414 - val_loss: 0.1308\n",
      "Epoch 157/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0408 - val_loss: 0.1334\n",
      "Epoch 158/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0457 - val_loss: 0.1308\n",
      "Epoch 159/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0482 - val_loss: 0.1345\n",
      "Epoch 160/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0449 - val_loss: 0.1239\n",
      "Epoch 161/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0417 - val_loss: 0.1353\n",
      "Epoch 162/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0382 - val_loss: 0.1310\n",
      "Epoch 163/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0382 - val_loss: 0.1312\n",
      "Epoch 164/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0421 - val_loss: 0.1278\n",
      "Epoch 165/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0421 - val_loss: 0.1384\n",
      "Epoch 166/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0422 - val_loss: 0.1300\n",
      "Epoch 167/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0403 - val_loss: 0.1317\n",
      "Epoch 168/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0386 - val_loss: 0.1375\n",
      "Epoch 169/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0379 - val_loss: 0.1266\n",
      "Epoch 170/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0386 - val_loss: 0.1389\n",
      "Epoch 171/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0372 - val_loss: 0.1314\n",
      "Epoch 172/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0375 - val_loss: 0.1322\n",
      "Epoch 173/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0384 - val_loss: 0.1353\n",
      "Epoch 174/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0377 - val_loss: 0.1323\n",
      "Epoch 175/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0379 - val_loss: 0.1386\n",
      "Epoch 176/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0399 - val_loss: 0.1295\n",
      "Epoch 177/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0381 - val_loss: 0.1338\n",
      "Epoch 178/600\n",
      "14/14 [==============================] - 0s 1ms/step - loss: 0.0368 - val_loss: 0.1310\n",
      "Epoch 179/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0371 - val_loss: 0.1320\n",
      "Epoch 180/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0376 - val_loss: 0.1349\n",
      "Epoch 181/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0388 - val_loss: 0.1264\n",
      "Epoch 182/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0382 - val_loss: 0.1310\n",
      "Epoch 183/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0376 - val_loss: 0.1498\n",
      "Epoch 184/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0372 - val_loss: 0.1373\n",
      "Epoch 185/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0362 - val_loss: 0.1320\n",
      "Epoch 186/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0350 - val_loss: 0.1401\n",
      "Epoch 187/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0378 - val_loss: 0.1351\n",
      "Epoch 188/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0354 - val_loss: 0.1383\n",
      "Epoch 189/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0394 - val_loss: 0.1464\n",
      "Epoch 190/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0374 - val_loss: 0.1321\n",
      "Epoch 191/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0382 - val_loss: 0.1400\n",
      "Epoch 192/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0352 - val_loss: 0.1305\n",
      "Epoch 193/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0359 - val_loss: 0.1429\n",
      "Epoch 194/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0389 - val_loss: 0.1334\n",
      "Epoch 195/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0350 - val_loss: 0.1387\n",
      "Epoch 196/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0355 - val_loss: 0.1444\n",
      "Epoch 197/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0345 - val_loss: 0.1366\n",
      "Epoch 198/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0341 - val_loss: 0.1351\n",
      "Epoch 199/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0382 - val_loss: 0.1496\n",
      "Epoch 200/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0379 - val_loss: 0.1309\n",
      "Epoch 201/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0346 - val_loss: 0.1457\n",
      "Epoch 202/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0339 - val_loss: 0.1352\n",
      "Epoch 203/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0354 - val_loss: 0.1339\n",
      "Epoch 204/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0349 - val_loss: 0.1431\n",
      "Epoch 205/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0337 - val_loss: 0.1328\n",
      "Epoch 206/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0336 - val_loss: 0.1495\n",
      "Epoch 207/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0342 - val_loss: 0.1360\n",
      "Epoch 208/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0352 - val_loss: 0.1417\n",
      "Epoch 209/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0345 - val_loss: 0.1425\n",
      "Epoch 210/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0333 - val_loss: 0.1473\n",
      "Epoch 211/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0330 - val_loss: 0.1360\n",
      "Epoch 212/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0350 - val_loss: 0.1363\n",
      "Epoch 213/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0335 - val_loss: 0.1381\n",
      "Epoch 214/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0334 - val_loss: 0.1476\n",
      "Epoch 215/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0337 - val_loss: 0.1450\n",
      "Epoch 216/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0321 - val_loss: 0.1420\n",
      "Epoch 217/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0338 - val_loss: 0.1382\n",
      "Epoch 218/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0330 - val_loss: 0.1438\n",
      "Epoch 219/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0331 - val_loss: 0.1535\n",
      "Epoch 220/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0343 - val_loss: 0.1333\n",
      "Epoch 221/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0407 - val_loss: 0.1591\n",
      "Epoch 222/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0378 - val_loss: 0.1442\n",
      "Epoch 223/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0348 - val_loss: 0.1523\n",
      "Epoch 224/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0336 - val_loss: 0.1329\n",
      "Epoch 225/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0353 - val_loss: 0.1458\n",
      "Epoch 226/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0313 - val_loss: 0.1435\n",
      "Epoch 227/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0320 - val_loss: 0.1470\n",
      "Epoch 228/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0312 - val_loss: 0.1461\n",
      "Epoch 229/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0308 - val_loss: 0.1444\n",
      "Epoch 230/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0306 - val_loss: 0.1414\n",
      "Epoch 231/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0318 - val_loss: 0.1521\n",
      "Epoch 232/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0371 - val_loss: 0.1448\n",
      "Epoch 233/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0307 - val_loss: 0.1489\n",
      "Epoch 234/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0319 - val_loss: 0.1483\n",
      "Epoch 235/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0317 - val_loss: 0.1498\n",
      "Epoch 236/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0300 - val_loss: 0.1453\n",
      "Epoch 237/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0307 - val_loss: 0.1577\n",
      "Epoch 238/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0331 - val_loss: 0.1475\n",
      "Epoch 239/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0306 - val_loss: 0.1588\n",
      "Epoch 240/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0308 - val_loss: 0.1485\n",
      "Epoch 241/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0292 - val_loss: 0.1462\n",
      "Epoch 242/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0295 - val_loss: 0.1479\n",
      "Epoch 243/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0303 - val_loss: 0.1503\n",
      "Epoch 244/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0296 - val_loss: 0.1501\n",
      "Epoch 245/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0298 - val_loss: 0.1437\n",
      "Epoch 246/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0338 - val_loss: 0.1666\n",
      "Epoch 247/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0296 - val_loss: 0.1454\n",
      "Epoch 248/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0291 - val_loss: 0.1622\n",
      "Epoch 249/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0301 - val_loss: 0.1409\n",
      "Epoch 250/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0305 - val_loss: 0.1538\n",
      "Epoch 251/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0279 - val_loss: 0.1474\n",
      "Epoch 252/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0301 - val_loss: 0.1599\n",
      "Epoch 253/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0274 - val_loss: 0.1447\n",
      "Epoch 254/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0295 - val_loss: 0.1570\n",
      "Epoch 255/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0292 - val_loss: 0.1474\n",
      "Epoch 256/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0307 - val_loss: 0.1752\n",
      "Epoch 257/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0316 - val_loss: 0.1456\n",
      "Epoch 258/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0289 - val_loss: 0.1483\n",
      "Epoch 259/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0270 - val_loss: 0.1507\n",
      "Epoch 260/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0295 - val_loss: 0.1434\n",
      "Epoch 261/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0293 - val_loss: 0.1571\n",
      "Epoch 262/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0291 - val_loss: 0.1418\n",
      "Epoch 263/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0328 - val_loss: 0.1544\n",
      "Epoch 264/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0414 - val_loss: 0.1469\n",
      "Epoch 265/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0275 - val_loss: 0.1591\n",
      "Epoch 266/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0263 - val_loss: 0.1429\n",
      "Epoch 267/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0282 - val_loss: 0.1518\n",
      "Epoch 268/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0270 - val_loss: 0.1505\n",
      "Epoch 269/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0263 - val_loss: 0.1634\n",
      "Epoch 270/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0308 - val_loss: 0.1670\n",
      "Epoch 271/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0272 - val_loss: 0.1503\n",
      "Epoch 272/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0261 - val_loss: 0.1616\n",
      "Epoch 273/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0269 - val_loss: 0.1587\n",
      "Epoch 274/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0263 - val_loss: 0.1525\n",
      "Epoch 275/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0263 - val_loss: 0.1505\n",
      "Epoch 276/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0267 - val_loss: 0.1639\n",
      "Epoch 277/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0325 - val_loss: 0.1514\n",
      "Epoch 278/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0492 - val_loss: 0.1443\n",
      "Epoch 279/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0310 - val_loss: 0.1619\n",
      "Epoch 280/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0254 - val_loss: 0.1475\n",
      "Epoch 281/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0257 - val_loss: 0.1591\n",
      "Epoch 282/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0269 - val_loss: 0.1542\n",
      "Epoch 283/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0291 - val_loss: 0.1547\n",
      "Epoch 284/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0271 - val_loss: 0.1659\n",
      "Epoch 285/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0274 - val_loss: 0.1462\n",
      "Epoch 286/600\n",
      "14/14 [==============================] - ETA: 0s - loss: 0.002 - 0s 2ms/step - loss: 0.0341 - val_loss: 0.1744\n",
      "Epoch 287/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0243 - val_loss: 0.1496\n",
      "Epoch 288/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0304 - val_loss: 0.1708\n",
      "Epoch 289/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0285 - val_loss: 0.1443\n",
      "Epoch 290/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0338 - val_loss: 0.1610\n",
      "Epoch 291/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0274 - val_loss: 0.1491\n",
      "Epoch 292/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0254 - val_loss: 0.1625\n",
      "Epoch 293/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0306 - val_loss: 0.1490\n",
      "Epoch 294/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0278 - val_loss: 0.1744\n",
      "Epoch 295/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0260 - val_loss: 0.1504\n",
      "Epoch 296/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0236 - val_loss: 0.1621\n",
      "Epoch 297/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0258 - val_loss: 0.1478\n",
      "Epoch 298/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0251 - val_loss: 0.1716\n",
      "Epoch 299/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0255 - val_loss: 0.1442\n",
      "Epoch 300/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0293 - val_loss: 0.1674\n",
      "Epoch 301/600\n",
      "14/14 [==============================] - 0s 6ms/step - loss: 0.0278 - val_loss: 0.1548\n",
      "Epoch 302/600\n",
      "14/14 [==============================] - 0s 8ms/step - loss: 0.0329 - val_loss: 0.1510\n",
      "Epoch 303/600\n",
      "14/14 [==============================] - 0s 6ms/step - loss: 0.0301 - val_loss: 0.1437\n",
      "Epoch 304/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0239 - val_loss: 0.1525\n",
      "Epoch 305/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0247 - val_loss: 0.1555\n",
      "Epoch 306/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0248 - val_loss: 0.1516\n",
      "Epoch 307/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0252 - val_loss: 0.1547\n",
      "Epoch 308/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0266 - val_loss: 0.1505\n",
      "Epoch 309/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0296 - val_loss: 0.1611\n",
      "Epoch 310/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0254 - val_loss: 0.1597\n",
      "Epoch 311/600\n",
      "14/14 [==============================] - 0s 1ms/step - loss: 0.0228 - val_loss: 0.1573\n",
      "Epoch 312/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0227 - val_loss: 0.1619\n",
      "Epoch 313/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0237 - val_loss: 0.1566\n",
      "Epoch 314/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0243 - val_loss: 0.1731\n",
      "Epoch 315/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0274 - val_loss: 0.1565\n",
      "Epoch 316/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0310 - val_loss: 0.1528\n",
      "Epoch 317/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0236 - val_loss: 0.1569\n",
      "Epoch 318/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0224 - val_loss: 0.1668\n",
      "Epoch 319/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0220 - val_loss: 0.1716\n",
      "Epoch 320/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0217 - val_loss: 0.1623\n",
      "Epoch 321/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0219 - val_loss: 0.1599\n",
      "Epoch 322/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0227 - val_loss: 0.1682\n",
      "Epoch 323/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0224 - val_loss: 0.1547\n",
      "Epoch 324/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0227 - val_loss: 0.1698\n",
      "Epoch 325/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0224 - val_loss: 0.1654\n",
      "Epoch 326/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0224 - val_loss: 0.1609\n",
      "Epoch 327/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0236 - val_loss: 0.1603\n",
      "Epoch 328/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0213 - val_loss: 0.1667\n",
      "Epoch 329/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0230 - val_loss: 0.1833\n",
      "Epoch 330/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0281 - val_loss: 0.1538\n",
      "Epoch 331/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0257 - val_loss: 0.1735\n",
      "Epoch 332/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0285 - val_loss: 0.1504\n",
      "Epoch 333/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0272 - val_loss: 0.1650\n",
      "Epoch 334/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0216 - val_loss: 0.1625\n",
      "Epoch 335/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0217 - val_loss: 0.1695\n",
      "Epoch 336/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0234 - val_loss: 0.1581\n",
      "Epoch 337/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0225 - val_loss: 0.1650\n",
      "Epoch 338/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0211 - val_loss: 0.1600\n",
      "Epoch 339/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0202 - val_loss: 0.1654\n",
      "Epoch 340/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0203 - val_loss: 0.1737\n",
      "Epoch 341/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0267 - val_loss: 0.1923\n",
      "Epoch 342/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0208 - val_loss: 0.1560\n",
      "Epoch 343/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0253 - val_loss: 0.1712\n",
      "Epoch 344/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0222 - val_loss: 0.1624\n",
      "Epoch 345/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0195 - val_loss: 0.1696\n",
      "Epoch 346/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0202 - val_loss: 0.1595\n",
      "Epoch 347/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0201 - val_loss: 0.1745\n",
      "Epoch 348/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0200 - val_loss: 0.1795\n",
      "Epoch 349/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0199 - val_loss: 0.1697\n",
      "Epoch 350/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0196 - val_loss: 0.1685\n",
      "Epoch 351/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0221 - val_loss: 0.1832\n",
      "Epoch 352/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0235 - val_loss: 0.1608\n",
      "Epoch 353/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0214 - val_loss: 0.1920\n",
      "Epoch 354/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0198 - val_loss: 0.1643\n",
      "Epoch 355/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0288 - val_loss: 0.1841\n",
      "Epoch 356/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0265 - val_loss: 0.1515\n",
      "Epoch 357/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0270 - val_loss: 0.1845\n",
      "Epoch 358/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0207 - val_loss: 0.1668\n",
      "Epoch 359/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0182 - val_loss: 0.1617\n",
      "Epoch 360/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0236 - val_loss: 0.1746\n",
      "Epoch 361/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0199 - val_loss: 0.1687\n",
      "Epoch 362/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0194 - val_loss: 0.1772\n",
      "Epoch 363/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0208 - val_loss: 0.1747\n",
      "Epoch 364/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0263 - val_loss: 0.1956\n",
      "Epoch 365/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0235 - val_loss: 0.1532\n",
      "Epoch 366/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0204 - val_loss: 0.1804\n",
      "Epoch 367/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0187 - val_loss: 0.1727\n",
      "Epoch 368/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0198 - val_loss: 0.1721\n",
      "Epoch 369/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0228 - val_loss: 0.1865\n",
      "Epoch 370/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0207 - val_loss: 0.1808\n",
      "Epoch 371/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0193 - val_loss: 0.1848\n",
      "Epoch 372/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0199 - val_loss: 0.1710\n",
      "Epoch 373/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0182 - val_loss: 0.1755\n",
      "Epoch 374/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0178 - val_loss: 0.1895\n",
      "Epoch 375/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0191 - val_loss: 0.1653\n",
      "Epoch 376/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0194 - val_loss: 0.1963\n",
      "Epoch 377/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0205 - val_loss: 0.1831\n",
      "Epoch 378/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0175 - val_loss: 0.1792\n",
      "Epoch 379/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0212 - val_loss: 0.1919\n",
      "Epoch 380/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0258 - val_loss: 0.1670\n",
      "Epoch 381/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0192 - val_loss: 0.1804\n",
      "Epoch 382/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0178 - val_loss: 0.1842\n",
      "Epoch 383/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0234 - val_loss: 0.1893\n",
      "Epoch 384/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0212 - val_loss: 0.1755\n",
      "Epoch 385/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0209 - val_loss: 0.1732\n",
      "Epoch 386/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0212 - val_loss: 0.1767\n",
      "Epoch 387/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0205 - val_loss: 0.1968\n",
      "Epoch 388/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0175 - val_loss: 0.1820\n",
      "Epoch 389/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0166 - val_loss: 0.1828\n",
      "Epoch 390/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0165 - val_loss: 0.1753\n",
      "Epoch 391/600\n",
      "14/14 [==============================] - ETA: 0s - loss: 0.001 - 0s 3ms/step - loss: 0.0161 - val_loss: 0.1916\n",
      "Epoch 392/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0164 - val_loss: 0.1770\n",
      "Epoch 393/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0250 - val_loss: 0.2088\n",
      "Epoch 394/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0194 - val_loss: 0.1837\n",
      "Epoch 395/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0183 - val_loss: 0.1773\n",
      "Epoch 396/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0188 - val_loss: 0.1967\n",
      "Epoch 397/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0336 - val_loss: 0.1674\n",
      "Epoch 398/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0192 - val_loss: 0.1917\n",
      "Epoch 399/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0163 - val_loss: 0.1783\n",
      "Epoch 400/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0159 - val_loss: 0.1884\n",
      "Epoch 401/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0166 - val_loss: 0.1844\n",
      "Epoch 402/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0168 - val_loss: 0.1952\n",
      "Epoch 403/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0159 - val_loss: 0.1927\n",
      "Epoch 404/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0151 - val_loss: 0.1831\n",
      "Epoch 405/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0155 - val_loss: 0.1919\n",
      "Epoch 406/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0156 - val_loss: 0.1871\n",
      "Epoch 407/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0148 - val_loss: 0.1923\n",
      "Epoch 408/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0151 - val_loss: 0.1900\n",
      "Epoch 409/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0186 - val_loss: 0.1821\n",
      "Epoch 410/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0168 - val_loss: 0.1868\n",
      "Epoch 411/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0158 - val_loss: 0.2007\n",
      "Epoch 412/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0170 - val_loss: 0.1867\n",
      "Epoch 413/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0155 - val_loss: 0.1897\n",
      "Epoch 414/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0151 - val_loss: 0.1858\n",
      "Epoch 415/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0179 - val_loss: 0.2025\n",
      "Epoch 416/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0165 - val_loss: 0.1907\n",
      "Epoch 417/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0158 - val_loss: 0.1912\n",
      "Epoch 418/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0161 - val_loss: 0.2023\n",
      "Epoch 419/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0151 - val_loss: 0.1804\n",
      "Epoch 420/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0150 - val_loss: 0.1948\n",
      "Epoch 421/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0159 - val_loss: 0.2041\n",
      "Epoch 422/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0161 - val_loss: 0.1836\n",
      "Epoch 423/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0154 - val_loss: 0.2301\n",
      "Epoch 424/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0181 - val_loss: 0.1843\n",
      "Epoch 425/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0143 - val_loss: 0.1979\n",
      "Epoch 426/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0138 - val_loss: 0.1879\n",
      "Epoch 427/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0151 - val_loss: 0.1966\n",
      "Epoch 428/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0137 - val_loss: 0.1904\n",
      "Epoch 429/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0136 - val_loss: 0.1960\n",
      "Epoch 430/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0132 - val_loss: 0.1899\n",
      "Epoch 431/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0137 - val_loss: 0.1961\n",
      "Epoch 432/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0160 - val_loss: 0.2001\n",
      "Epoch 433/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0178 - val_loss: 0.2218\n",
      "Epoch 434/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0165 - val_loss: 0.1775\n",
      "Epoch 435/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0145 - val_loss: 0.2055\n",
      "Epoch 436/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0139 - val_loss: 0.1896\n",
      "Epoch 437/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0174 - val_loss: 0.2254\n",
      "Epoch 438/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0161 - val_loss: 0.1975\n",
      "Epoch 439/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0130 - val_loss: 0.1969\n",
      "Epoch 440/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0129 - val_loss: 0.1962\n",
      "Epoch 441/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0133 - val_loss: 0.1995\n",
      "Epoch 442/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0130 - val_loss: 0.2010\n",
      "Epoch 443/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0159 - val_loss: 0.1953\n",
      "Epoch 444/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0147 - val_loss: 0.2035\n",
      "Epoch 445/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0122 - val_loss: 0.1870\n",
      "Epoch 446/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0132 - val_loss: 0.2156\n",
      "Epoch 447/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0154 - val_loss: 0.1921\n",
      "Epoch 448/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0140 - val_loss: 0.2133\n",
      "Epoch 449/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0132 - val_loss: 0.2017\n",
      "Epoch 450/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0137 - val_loss: 0.2012\n",
      "Epoch 451/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0125 - val_loss: 0.1984\n",
      "Epoch 452/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0123 - val_loss: 0.2133\n",
      "Epoch 453/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0134 - val_loss: 0.2059\n",
      "Epoch 454/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0136 - val_loss: 0.1971\n",
      "Epoch 455/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0131 - val_loss: 0.2059\n",
      "Epoch 456/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0131 - val_loss: 0.2077\n",
      "Epoch 457/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0122 - val_loss: 0.2152\n",
      "Epoch 458/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0136 - val_loss: 0.1992\n",
      "Epoch 459/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0121 - val_loss: 0.2210\n",
      "Epoch 460/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0129 - val_loss: 0.2081\n",
      "Epoch 461/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0157 - val_loss: 0.2096\n",
      "Epoch 462/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0133 - val_loss: 0.2101\n",
      "Epoch 463/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0131 - val_loss: 0.2147\n",
      "Epoch 464/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0135 - val_loss: 0.2025\n",
      "Epoch 465/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0122 - val_loss: 0.2173\n",
      "Epoch 466/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0113 - val_loss: 0.2076\n",
      "Epoch 467/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0130 - val_loss: 0.2070\n",
      "Epoch 468/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0122 - val_loss: 0.2159\n",
      "Epoch 469/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0116 - val_loss: 0.2135\n",
      "Epoch 470/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0115 - val_loss: 0.2170\n",
      "Epoch 471/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0121 - val_loss: 0.2078\n",
      "Epoch 472/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0126 - val_loss: 0.2253\n",
      "Epoch 473/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0132 - val_loss: 0.2084\n",
      "Epoch 474/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0113 - val_loss: 0.2242\n",
      "Epoch 475/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0130 - val_loss: 0.2123\n",
      "Epoch 476/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0139 - val_loss: 0.2016\n",
      "Epoch 477/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0118 - val_loss: 0.2338\n",
      "Epoch 478/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0178 - val_loss: 0.1968\n",
      "Epoch 479/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0130 - val_loss: 0.2447\n",
      "Epoch 480/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0112 - val_loss: 0.2088\n",
      "Epoch 481/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0135 - val_loss: 0.2508\n",
      "Epoch 482/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0126 - val_loss: 0.2032\n",
      "Epoch 483/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0110 - val_loss: 0.2393\n",
      "Epoch 484/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0104 - val_loss: 0.2052\n",
      "Epoch 485/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0133 - val_loss: 0.2207\n",
      "Epoch 486/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0116 - val_loss: 0.2284\n",
      "Epoch 487/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0154 - val_loss: 0.2045\n",
      "Epoch 488/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0170 - val_loss: 0.2898\n",
      "Epoch 489/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0215 - val_loss: 0.2125\n",
      "Epoch 490/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0189 - val_loss: 0.2195\n",
      "Epoch 491/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0169 - val_loss: 0.2358\n",
      "Epoch 492/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0125 - val_loss: 0.2230\n",
      "Epoch 493/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0103 - val_loss: 0.2167\n",
      "Epoch 494/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0129 - val_loss: 0.2482\n",
      "Epoch 495/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0131 - val_loss: 0.2213\n",
      "Epoch 496/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0118 - val_loss: 0.2401\n",
      "Epoch 497/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0105 - val_loss: 0.2166\n",
      "Epoch 498/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0192 - val_loss: 0.2681\n",
      "Epoch 499/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0168 - val_loss: 0.2049\n",
      "Epoch 500/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0135 - val_loss: 0.2210\n",
      "Epoch 501/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0128 - val_loss: 0.2288\n",
      "Epoch 502/600\n",
      "14/14 [==============================] - ETA: 0s - loss: 0.010 - 0s 3ms/step - loss: 0.0108 - val_loss: 0.2163\n",
      "Epoch 503/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0092 - val_loss: 0.2356\n",
      "Epoch 504/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0096 - val_loss: 0.2280\n",
      "Epoch 505/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0101 - val_loss: 0.2254\n",
      "Epoch 506/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0146 - val_loss: 0.3215\n",
      "Epoch 507/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0223 - val_loss: 0.2048\n",
      "Epoch 508/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0112 - val_loss: 0.2442\n",
      "Epoch 509/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0100 - val_loss: 0.2347\n",
      "Epoch 510/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0102 - val_loss: 0.2334\n",
      "Epoch 511/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0093 - val_loss: 0.2332\n",
      "Epoch 512/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0093 - val_loss: 0.2323\n",
      "Epoch 513/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0104 - val_loss: 0.2229\n",
      "Epoch 514/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0088 - val_loss: 0.2343\n",
      "Epoch 515/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0090 - val_loss: 0.2279\n",
      "Epoch 516/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0091 - val_loss: 0.2340\n",
      "Epoch 517/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0121 - val_loss: 0.2236\n",
      "Epoch 518/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0106 - val_loss: 0.2314\n",
      "Epoch 519/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0088 - val_loss: 0.2332\n",
      "Epoch 520/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0099 - val_loss: 0.2243\n",
      "Epoch 521/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0101 - val_loss: 0.2382\n",
      "Epoch 522/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0094 - val_loss: 0.2272\n",
      "Epoch 523/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0097 - val_loss: 0.2379\n",
      "Epoch 524/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0095 - val_loss: 0.2406\n",
      "Epoch 525/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0089 - val_loss: 0.2373\n",
      "Epoch 526/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0086 - val_loss: 0.2379\n",
      "Epoch 527/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0085 - val_loss: 0.2310\n",
      "Epoch 528/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0085 - val_loss: 0.2372\n",
      "Epoch 529/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0088 - val_loss: 0.2538\n",
      "Epoch 530/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0108 - val_loss: 0.2173\n",
      "Epoch 531/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0096 - val_loss: 0.2503\n",
      "Epoch 532/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0112 - val_loss: 0.2175\n",
      "Epoch 533/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0091 - val_loss: 0.2565\n",
      "Epoch 534/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0081 - val_loss: 0.2200\n",
      "Epoch 535/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0081 - val_loss: 0.2641\n",
      "Epoch 536/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0110 - val_loss: 0.2480\n",
      "Epoch 537/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0086 - val_loss: 0.2555\n",
      "Epoch 538/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0090 - val_loss: 0.2215\n",
      "Epoch 539/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0096 - val_loss: 0.2652\n",
      "Epoch 540/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0113 - val_loss: 0.2387\n",
      "Epoch 541/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0082 - val_loss: 0.2473\n",
      "Epoch 542/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0096 - val_loss: 0.2619\n",
      "Epoch 543/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0094 - val_loss: 0.2289\n",
      "Epoch 544/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0081 - val_loss: 0.2640\n",
      "Epoch 545/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0093 - val_loss: 0.2433\n",
      "Epoch 546/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0089 - val_loss: 0.2409\n",
      "Epoch 547/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0079 - val_loss: 0.2396\n",
      "Epoch 548/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0082 - val_loss: 0.2455\n",
      "Epoch 549/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0077 - val_loss: 0.2580\n",
      "Epoch 550/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0084 - val_loss: 0.2332\n",
      "Epoch 551/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0081 - val_loss: 0.2754\n",
      "Epoch 552/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0080 - val_loss: 0.2339\n",
      "Epoch 553/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0092 - val_loss: 0.2819\n",
      "Epoch 554/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0090 - val_loss: 0.2363\n",
      "Epoch 555/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0084 - val_loss: 0.2797\n",
      "Epoch 556/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0080 - val_loss: 0.2376\n",
      "Epoch 557/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0099 - val_loss: 0.2845\n",
      "Epoch 558/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0089 - val_loss: 0.2424\n",
      "Epoch 559/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0078 - val_loss: 0.2596\n",
      "Epoch 560/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0084 - val_loss: 0.2483\n",
      "Epoch 561/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0064 - val_loss: 0.2702\n",
      "Epoch 562/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0073 - val_loss: 0.2415\n",
      "Epoch 563/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0090 - val_loss: 0.3074\n",
      "Epoch 564/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0170 - val_loss: 0.2263\n",
      "Epoch 565/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0107 - val_loss: 0.2729\n",
      "Epoch 566/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0101 - val_loss: 0.2746\n",
      "Epoch 567/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0092 - val_loss: 0.2609\n",
      "Epoch 568/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0088 - val_loss: 0.2724\n",
      "Epoch 569/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0078 - val_loss: 0.2602\n",
      "Epoch 570/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0073 - val_loss: 0.2496\n",
      "Epoch 571/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0076 - val_loss: 0.2557\n",
      "Epoch 572/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0074 - val_loss: 0.2537\n",
      "Epoch 573/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0070 - val_loss: 0.2611\n",
      "Epoch 574/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0064 - val_loss: 0.2603\n",
      "Epoch 575/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0065 - val_loss: 0.2647\n",
      "Epoch 576/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0068 - val_loss: 0.2734\n",
      "Epoch 577/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0071 - val_loss: 0.2620\n",
      "Epoch 578/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0065 - val_loss: 0.2684\n",
      "Epoch 579/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0063 - val_loss: 0.2691\n",
      "Epoch 580/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0062 - val_loss: 0.2699\n",
      "Epoch 581/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0062 - val_loss: 0.2707\n",
      "Epoch 582/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0064 - val_loss: 0.2780\n",
      "Epoch 583/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0068 - val_loss: 0.2622\n",
      "Epoch 584/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0068 - val_loss: 0.2703\n",
      "Epoch 585/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0076 - val_loss: 0.3144\n",
      "Epoch 586/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0086 - val_loss: 0.2503\n",
      "Epoch 587/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0075 - val_loss: 0.2855\n",
      "Epoch 588/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0071 - val_loss: 0.2648\n",
      "Epoch 589/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0077 - val_loss: 0.2632\n",
      "Epoch 590/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0066 - val_loss: 0.2711\n",
      "Epoch 591/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0086 - val_loss: 0.2557\n",
      "Epoch 592/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0086 - val_loss: 0.2919\n",
      "Epoch 593/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0058 - val_loss: 0.2564\n",
      "Epoch 594/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0065 - val_loss: 0.2640\n",
      "Epoch 595/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0063 - val_loss: 0.2703\n",
      "Epoch 596/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0057 - val_loss: 0.2657\n",
      "Epoch 597/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0056 - val_loss: 0.2758\n",
      "Epoch 598/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0055 - val_loss: 0.2728\n",
      "Epoch 599/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0056 - val_loss: 0.2684\n",
      "Epoch 600/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0059 - val_loss: 0.2636\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x44feb975b0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x=X_train, y=y_train, \n",
    "             epochs=600,\n",
    "             validation_data=(X_test, y_test),\n",
    "             verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "losses = pd.DataFrame(model.history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "losses.plot();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the perfect example of overfitting. We can see that initally, loss is starting to decrease on validation.\n",
    "\n",
    "However it get spiked up after a certain point in time of few epoches."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <a name='earlystop'>Example Two: Early Stopping</a>\n",
    "\n",
    "We obviously trained too much! Let's use early stopping to track the val_loss and stop training once it begins increasing too much!\n",
    "\n",
    "+ **`monitor`**: Quantity to be monitored. Which we are monitoring such as `val_loss` or `accuracy`.\n",
    "+ **`mode`**: if we are monitoring loss, we want to minimize it. So we use `min`. For the case of accuracy, we want to maximize it. So we use `max`\n",
    "+ **` patience`**: Number of epochs with no improvement after which training will be stopped. It is a hard stopping."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Dense(30, activation='relu'))\n",
    "model.add(Dense(15, activation='relu'))\n",
    "\n",
    "# Binary Classification => Sigmoid\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  help(EarlyStopping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "early_stopping = EarlyStopping(monitor='val_loss', \n",
    "                                                  mode='min',\n",
    "                                                  patience=25,\n",
    "                                                  verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "14/14 [==============================] - 0s 29ms/step - loss: 0.7097 - val_loss: 0.6804\n",
      "Epoch 2/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.6748 - val_loss: 0.6584\n",
      "Epoch 3/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6517 - val_loss: 0.6366\n",
      "Epoch 4/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.6283 - val_loss: 0.6116\n",
      "Epoch 5/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6014 - val_loss: 0.5769\n",
      "Epoch 6/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.5620 - val_loss: 0.5379\n",
      "Epoch 7/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.5218 - val_loss: 0.4925\n",
      "Epoch 8/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.4784 - val_loss: 0.4477\n",
      "Epoch 9/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4327 - val_loss: 0.4062\n",
      "Epoch 10/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.3929 - val_loss: 0.3613\n",
      "Epoch 11/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3528 - val_loss: 0.3259\n",
      "Epoch 12/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3208 - val_loss: 0.2937\n",
      "Epoch 13/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.2935 - val_loss: 0.2657\n",
      "Epoch 14/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2790 - val_loss: 0.2418\n",
      "Epoch 15/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2485 - val_loss: 0.2291\n",
      "Epoch 16/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2337 - val_loss: 0.2079\n",
      "Epoch 17/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2187 - val_loss: 0.1992\n",
      "Epoch 18/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2064 - val_loss: 0.1838\n",
      "Epoch 19/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1959 - val_loss: 0.1757\n",
      "Epoch 20/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1841 - val_loss: 0.1637\n",
      "Epoch 21/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1765 - val_loss: 0.1585\n",
      "Epoch 22/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1724 - val_loss: 0.1507\n",
      "Epoch 23/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1610 - val_loss: 0.1437\n",
      "Epoch 24/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1604 - val_loss: 0.1471\n",
      "Epoch 25/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1524 - val_loss: 0.1353\n",
      "Epoch 26/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1460 - val_loss: 0.1395\n",
      "Epoch 27/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1377 - val_loss: 0.1282\n",
      "Epoch 28/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1335 - val_loss: 0.1325\n",
      "Epoch 29/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1295 - val_loss: 0.1263\n",
      "Epoch 30/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1259 - val_loss: 0.1204\n",
      "Epoch 31/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1240 - val_loss: 0.1197\n",
      "Epoch 32/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1166 - val_loss: 0.1195\n",
      "Epoch 33/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1128 - val_loss: 0.1137\n",
      "Epoch 34/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1109 - val_loss: 0.1200\n",
      "Epoch 35/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1043 - val_loss: 0.1105\n",
      "Epoch 36/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1038 - val_loss: 0.1130\n",
      "Epoch 37/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1007 - val_loss: 0.1107\n",
      "Epoch 38/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0975 - val_loss: 0.1087\n",
      "Epoch 39/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0954 - val_loss: 0.1069\n",
      "Epoch 40/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0928 - val_loss: 0.1117\n",
      "Epoch 41/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0913 - val_loss: 0.1077\n",
      "Epoch 42/600\n",
      "14/14 [==============================] - 0s 6ms/step - loss: 0.0891 - val_loss: 0.1052\n",
      "Epoch 43/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0863 - val_loss: 0.1069\n",
      "Epoch 44/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0843 - val_loss: 0.1036\n",
      "Epoch 45/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0825 - val_loss: 0.1073\n",
      "Epoch 46/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0822 - val_loss: 0.1047\n",
      "Epoch 47/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0815 - val_loss: 0.1049\n",
      "Epoch 48/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0777 - val_loss: 0.1010\n",
      "Epoch 49/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0780 - val_loss: 0.1064\n",
      "Epoch 50/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0751 - val_loss: 0.1021\n",
      "Epoch 51/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0739 - val_loss: 0.1052\n",
      "Epoch 52/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0726 - val_loss: 0.1018\n",
      "Epoch 53/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0730 - val_loss: 0.1027\n",
      "Epoch 54/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0707 - val_loss: 0.1069\n",
      "Epoch 55/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0725 - val_loss: 0.1047\n",
      "Epoch 56/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0730 - val_loss: 0.1014\n",
      "Epoch 57/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0680 - val_loss: 0.1057\n",
      "Epoch 58/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0675 - val_loss: 0.1061\n",
      "Epoch 59/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0683 - val_loss: 0.1052\n",
      "Epoch 60/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0662 - val_loss: 0.1063\n",
      "Epoch 61/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0656 - val_loss: 0.1043\n",
      "Epoch 62/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0686 - val_loss: 0.1037\n",
      "Epoch 63/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0647 - val_loss: 0.1036\n",
      "Epoch 64/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0651 - val_loss: 0.1077\n",
      "Epoch 65/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0621 - val_loss: 0.1022\n",
      "Epoch 66/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0628 - val_loss: 0.1076\n",
      "Epoch 67/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0614 - val_loss: 0.1067\n",
      "Epoch 68/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0612 - val_loss: 0.1052\n",
      "Epoch 69/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0611 - val_loss: 0.1057\n",
      "Epoch 70/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0594 - val_loss: 0.1087\n",
      "Epoch 71/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0594 - val_loss: 0.1069\n",
      "Epoch 72/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0612 - val_loss: 0.1088\n",
      "Epoch 73/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0609 - val_loss: 0.1049\n",
      "Epoch 00073: early stopping\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x44ff267e80>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x=X_train, y=y_train,\n",
    "             validation_data=(X_test, y_test),\n",
    "             epochs=600,\n",
    "             callbacks=[early_stopping],\n",
    "             verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "losses = pd.DataFrame(model.history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "losses.plot();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the plot that we want to see. Both loss and val_loss are beginning to decrease."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <a name='dropout'>Example Three: Adding in DropOut Layers</a>\n",
    "\n",
    "**Early Stopping**\n",
    "\n",
    "- Keras can automatically stop training based on a loss condition on the validation data passed during the model.fit call.\n",
    "\n",
    "**Dropout Layers**\n",
    "\n",
    "- can be added to layers to \"turn off\" neurons during training to prevent overfitting.\n",
    "\n",
    "- each dropout layer will \"drop\" a user-defined percentage of neuron units in the previous layer every batch.\n",
    "\n",
    "**We can define by `model.add(Dropout(0 ~ 1))`: values can between 0 and 1.**\n",
    "\n",
    "0 percent , 1 mean 100% of neurons."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Dense(30, activation='relu'))\n",
    "model.add(Dropout(0.5)) # 50% of randomly selected neurons for each batch (1 entire epoch triaining) will be truned off\n",
    "\n",
    "model.add(Dense(15, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# in combination with early stopping callbacks\n",
    "early_stopping = EarlyStopping(monitor='val_loss',\n",
    "                                                  mode='min',\n",
    "                                                  patience=25,\n",
    "                                                  verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/600\n",
      "14/14 [==============================] - 0s 18ms/step - loss: 0.6931 - val_loss: 0.6675\n",
      "Epoch 2/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6685 - val_loss: 0.6503\n",
      "Epoch 3/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.6499 - val_loss: 0.6289\n",
      "Epoch 4/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6359 - val_loss: 0.6076\n",
      "Epoch 5/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.6001 - val_loss: 0.5792\n",
      "Epoch 6/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.5845 - val_loss: 0.5454\n",
      "Epoch 7/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.5740 - val_loss: 0.5215\n",
      "Epoch 8/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.5369 - val_loss: 0.4953\n",
      "Epoch 9/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.5524 - val_loss: 0.4779\n",
      "Epoch 10/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.5163 - val_loss: 0.4585\n",
      "Epoch 11/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4919 - val_loss: 0.4314\n",
      "Epoch 12/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4822 - val_loss: 0.4135\n",
      "Epoch 13/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4806 - val_loss: 0.3951\n",
      "Epoch 14/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4563 - val_loss: 0.3765\n",
      "Epoch 15/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4546 - val_loss: 0.3615\n",
      "Epoch 16/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4189 - val_loss: 0.3440\n",
      "Epoch 17/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4206 - val_loss: 0.3272\n",
      "Epoch 18/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4341 - val_loss: 0.3165\n",
      "Epoch 19/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.4212 - val_loss: 0.3085\n",
      "Epoch 20/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3699 - val_loss: 0.2965\n",
      "Epoch 21/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.4038 - val_loss: 0.2845\n",
      "Epoch 22/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3712 - val_loss: 0.2736\n",
      "Epoch 23/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3459 - val_loss: 0.2604\n",
      "Epoch 24/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3580 - val_loss: 0.2525\n",
      "Epoch 25/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3494 - val_loss: 0.2405\n",
      "Epoch 26/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3133 - val_loss: 0.2317\n",
      "Epoch 27/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3244 - val_loss: 0.2248\n",
      "Epoch 28/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3573 - val_loss: 0.2218\n",
      "Epoch 29/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.3455 - val_loss: 0.2147\n",
      "Epoch 30/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.2943 - val_loss: 0.2067\n",
      "Epoch 31/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2890 - val_loss: 0.1967\n",
      "Epoch 32/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2974 - val_loss: 0.1902\n",
      "Epoch 33/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.2775 - val_loss: 0.1879\n",
      "Epoch 34/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.2795 - val_loss: 0.1793\n",
      "Epoch 35/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.2691 - val_loss: 0.1773\n",
      "Epoch 36/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.3213 - val_loss: 0.1706\n",
      "Epoch 37/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2731 - val_loss: 0.1739\n",
      "Epoch 38/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2531 - val_loss: 0.1663\n",
      "Epoch 39/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2325 - val_loss: 0.1586\n",
      "Epoch 40/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2615 - val_loss: 0.1614\n",
      "Epoch 41/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2492 - val_loss: 0.1684\n",
      "Epoch 42/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2250 - val_loss: 0.1502\n",
      "Epoch 43/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2361 - val_loss: 0.1469\n",
      "Epoch 44/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2033 - val_loss: 0.1429\n",
      "Epoch 45/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2040 - val_loss: 0.1382\n",
      "Epoch 46/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2157 - val_loss: 0.1324\n",
      "Epoch 47/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2200 - val_loss: 0.1363\n",
      "Epoch 48/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2220 - val_loss: 0.1326\n",
      "Epoch 49/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1798 - val_loss: 0.1230\n",
      "Epoch 50/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.2125 - val_loss: 0.1245\n",
      "Epoch 51/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.2003 - val_loss: 0.1283\n",
      "Epoch 52/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1853 - val_loss: 0.1170\n",
      "Epoch 53/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1820 - val_loss: 0.1143\n",
      "Epoch 54/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1882 - val_loss: 0.1223\n",
      "Epoch 55/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1830 - val_loss: 0.1130\n",
      "Epoch 56/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1823 - val_loss: 0.1132\n",
      "Epoch 57/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1812 - val_loss: 0.1136\n",
      "Epoch 58/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1941 - val_loss: 0.1155\n",
      "Epoch 59/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1601 - val_loss: 0.1112\n",
      "Epoch 60/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1516 - val_loss: 0.1104\n",
      "Epoch 61/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1378 - val_loss: 0.1049\n",
      "Epoch 62/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1410 - val_loss: 0.1066\n",
      "Epoch 63/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1487 - val_loss: 0.1038\n",
      "Epoch 64/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1658 - val_loss: 0.1030\n",
      "Epoch 65/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1483 - val_loss: 0.1036\n",
      "Epoch 66/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1161 - val_loss: 0.0947\n",
      "Epoch 67/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1412 - val_loss: 0.0955\n",
      "Epoch 68/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1633 - val_loss: 0.1095\n",
      "Epoch 69/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1495 - val_loss: 0.0970\n",
      "Epoch 70/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1562 - val_loss: 0.1001\n",
      "Epoch 71/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1327 - val_loss: 0.0971\n",
      "Epoch 72/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1338 - val_loss: 0.0993\n",
      "Epoch 73/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1426 - val_loss: 0.0958\n",
      "Epoch 74/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1404 - val_loss: 0.0973\n",
      "Epoch 75/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1246 - val_loss: 0.1029\n",
      "Epoch 76/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1501 - val_loss: 0.0922\n",
      "Epoch 77/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1553 - val_loss: 0.0911\n",
      "Epoch 78/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1331 - val_loss: 0.1053\n",
      "Epoch 79/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1507 - val_loss: 0.1011\n",
      "Epoch 80/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1229 - val_loss: 0.0976\n",
      "Epoch 81/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1301 - val_loss: 0.1035\n",
      "Epoch 82/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1209 - val_loss: 0.1093\n",
      "Epoch 83/600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1343 - val_loss: 0.0935\n",
      "Epoch 84/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1232 - val_loss: 0.0917\n",
      "Epoch 85/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1213 - val_loss: 0.0982\n",
      "Epoch 86/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1328 - val_loss: 0.1003\n",
      "Epoch 87/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1236 - val_loss: 0.0999\n",
      "Epoch 88/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1423 - val_loss: 0.0990\n",
      "Epoch 89/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1288 - val_loss: 0.0992\n",
      "Epoch 90/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1269 - val_loss: 0.0962\n",
      "Epoch 91/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1141 - val_loss: 0.1039\n",
      "Epoch 92/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1048 - val_loss: 0.0928\n",
      "Epoch 93/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1062 - val_loss: 0.0933\n",
      "Epoch 94/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1125 - val_loss: 0.0916\n",
      "Epoch 95/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1112 - val_loss: 0.0902\n",
      "Epoch 96/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1167 - val_loss: 0.0950\n",
      "Epoch 97/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1250 - val_loss: 0.0997\n",
      "Epoch 98/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1116 - val_loss: 0.0878\n",
      "Epoch 99/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1173 - val_loss: 0.0855\n",
      "Epoch 100/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1123 - val_loss: 0.0916\n",
      "Epoch 101/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1034 - val_loss: 0.0853\n",
      "Epoch 102/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0980 - val_loss: 0.0910\n",
      "Epoch 103/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1149 - val_loss: 0.0895\n",
      "Epoch 104/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1044 - val_loss: 0.0845\n",
      "Epoch 105/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1119 - val_loss: 0.1021\n",
      "Epoch 106/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1042 - val_loss: 0.0880\n",
      "Epoch 107/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1145 - val_loss: 0.0937\n",
      "Epoch 108/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.1262 - val_loss: 0.1050\n",
      "Epoch 109/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1076 - val_loss: 0.0995\n",
      "Epoch 110/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1092 - val_loss: 0.0905\n",
      "Epoch 111/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0972 - val_loss: 0.0960\n",
      "Epoch 112/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.1096 - val_loss: 0.0889\n",
      "Epoch 113/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.1023 - val_loss: 0.0923\n",
      "Epoch 114/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0934 - val_loss: 0.0884\n",
      "Epoch 115/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0885 - val_loss: 0.0930\n",
      "Epoch 116/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.1098 - val_loss: 0.0963\n",
      "Epoch 117/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0860 - val_loss: 0.0873\n",
      "Epoch 118/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0871 - val_loss: 0.0905\n",
      "Epoch 119/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1095 - val_loss: 0.0899\n",
      "Epoch 120/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.1007 - val_loss: 0.1007\n",
      "Epoch 121/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1195 - val_loss: 0.0935\n",
      "Epoch 122/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0928 - val_loss: 0.0803\n",
      "Epoch 123/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0902 - val_loss: 0.0976\n",
      "Epoch 124/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1146 - val_loss: 0.0921\n",
      "Epoch 125/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0857 - val_loss: 0.0934\n",
      "Epoch 126/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0956 - val_loss: 0.0973\n",
      "Epoch 127/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0991 - val_loss: 0.1019\n",
      "Epoch 128/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0917 - val_loss: 0.0870\n",
      "Epoch 129/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0865 - val_loss: 0.0908\n",
      "Epoch 130/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0798 - val_loss: 0.0947\n",
      "Epoch 131/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0876 - val_loss: 0.1040\n",
      "Epoch 132/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0829 - val_loss: 0.1075\n",
      "Epoch 133/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0954 - val_loss: 0.0861\n",
      "Epoch 134/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0940 - val_loss: 0.0970\n",
      "Epoch 135/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.1026 - val_loss: 0.0872\n",
      "Epoch 136/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0825 - val_loss: 0.0859\n",
      "Epoch 137/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0700 - val_loss: 0.0944\n",
      "Epoch 138/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0847 - val_loss: 0.0953\n",
      "Epoch 139/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0840 - val_loss: 0.0939\n",
      "Epoch 140/600\n",
      "14/14 [==============================] - 0s 2ms/step - loss: 0.0901 - val_loss: 0.0960\n",
      "Epoch 141/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0880 - val_loss: 0.0882\n",
      "Epoch 142/600\n",
      "14/14 [==============================] - 0s 6ms/step - loss: 0.0833 - val_loss: 0.0935\n",
      "Epoch 143/600\n",
      "14/14 [==============================] - 0s 4ms/step - loss: 0.0859 - val_loss: 0.0898\n",
      "Epoch 144/600\n",
      "14/14 [==============================] - 0s 6ms/step - loss: 0.0841 - val_loss: 0.0871\n",
      "Epoch 145/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0893 - val_loss: 0.0915\n",
      "Epoch 146/600\n",
      "14/14 [==============================] - 0s 3ms/step - loss: 0.0936 - val_loss: 0.0963\n",
      "Epoch 147/600\n",
      "14/14 [==============================] - 0s 5ms/step - loss: 0.0841 - val_loss: 0.0947\n",
      "Epoch 00147: early stopping\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x4480fe0ca0>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x=X_train, y=y_train,\n",
    "             validation_data=(X_test, y_test),\n",
    "             callbacks=[early_stopping],\n",
    "             epochs=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "losses = pd.DataFrame(model.history.history)\n",
    "\n",
    "losses.plot();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is great. Both loss and val_loss are flattening out at the same time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "------\n",
    "-----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1],\n",
       "       [1],\n",
       "       [1],\n",
       "       [0],\n",
       "       [1]])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = model.predict_classes(X_test)\n",
    "\n",
    "predictions[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.98      0.96        55\n",
      "           1       0.99      0.97      0.98        88\n",
      "\n",
      "    accuracy                           0.97       143\n",
      "   macro avg       0.97      0.97      0.97       143\n",
      "weighted avg       0.97      0.97      0.97       143\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[54,  1],\n",
       "       [ 3, 85]], dtype=int64)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_test, predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(confusion_matrix(y_test, predictions), annot=True, cbar=False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-----"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv-datascience",
   "language": "python",
   "name": "venv-datascience"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}