Neural Network - Price Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using Theano backend.\n",
      "WARNING (theano.configdefaults): g++ not available, if using conda: `conda install m2w64-toolchain`\n",
      "WARNING (theano.configdefaults): g++ not detected ! Theano will be unable to execute optimized C-implementations (for both CPU and GPU) and will default to Python implementations. Performance will be severely degraded. To remove this warning, set Theano flags cxx to an empty string.\n",
      "C:\\Users\\Mayur\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow\n",
    "import keras\n",
    "#from keras.models import Sequential\n",
    "#from keras.layers import Dense\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.cross_validation import train_test_split\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 16829 entries, 0 to 16828\n",
      "Data columns (total 13 columns):\n",
      "date              16829 non-null object\n",
      "serial_number     16829 non-null object\n",
      "model             16829 non-null object\n",
      "capacity_bytes    16829 non-null int64\n",
      "failure           16829 non-null object\n",
      "smart_1_raw       16829 non-null float64\n",
      "smart_5_raw       16829 non-null float64\n",
      "smart_9_raw       16829 non-null int64\n",
      "smart_194_raw     16829 non-null int64\n",
      "smart_197_raw     16829 non-null float64\n",
      "Temp              16829 non-null float64\n",
      "Duration          16829 non-null int64\n",
      "RPM               16829 non-null int64\n",
      "dtypes: float64(4), int64(5), object(4)\n",
      "memory usage: 1.7+ MB\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv('Modified.csv')\n",
    "df.head()\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 16829 entries, 0 to 16828\n",
      "Data columns (total 13 columns):\n",
      "date              16829 non-null object\n",
      "serial_number     16829 non-null object\n",
      "model             16829 non-null object\n",
      "capacity_bytes    16829 non-null int64\n",
      "failure           16829 non-null object\n",
      "smart_1_raw       16829 non-null float64\n",
      "smart_5_raw       16829 non-null float64\n",
      "smart_9_raw       16829 non-null int64\n",
      "smart_194_raw     16829 non-null int64\n",
      "smart_197_raw     16829 non-null float64\n",
      "Temp              16829 non-null float64\n",
      "Duration          16829 non-null int64\n",
      "RPM               16829 non-null int64\n",
      "dtypes: float64(4), int64(5), object(4)\n",
      "memory usage: 1.7+ MB\n"
     ]
    }
   ],
   "source": [
    "#df = df.drop(df.index[range(20000,1048575)])\n",
    "df.info()        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def handle_non_numerical_data(df):\n",
    "    columns = df.columns.values\n",
    "    \n",
    "    for column in columns:\n",
    "        text_digit_vals = {} \n",
    "        def convert_to_int(val):\n",
    "            return text_digit_vals[val]\n",
    "        \n",
    "        if df[column].dtype != np.int64 and df[column].dtype != np.float64:\n",
    "            column_contents = df[column].values.tolist()\n",
    "            unique_elements = set(column_contents)\n",
    "            x=0\n",
    "            for unique in unique_elements:\n",
    "                if unique not in text_digit_vals:\n",
    "                    text_digit_vals[unique] = x\n",
    "                    x = x + 1\n",
    "            \n",
    "            df[column] = list(map(convert_to_int, df[column]))\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = handle_non_numerical_data(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>serial_number</th>\n",
       "      <th>model</th>\n",
       "      <th>capacity_bytes</th>\n",
       "      <th>failure</th>\n",
       "      <th>smart_1_raw</th>\n",
       "      <th>smart_5_raw</th>\n",
       "      <th>smart_9_raw</th>\n",
       "      <th>smart_194_raw</th>\n",
       "      <th>smart_197_raw</th>\n",
       "      <th>Temp</th>\n",
       "      <th>Duration</th>\n",
       "      <th>RPM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>79</td>\n",
       "      <td>3386</td>\n",
       "      <td>5</td>\n",
       "      <td>3000592982016</td>\n",
       "      <td>0</td>\n",
       "      <td>12688.97</td>\n",
       "      <td>3306.02</td>\n",
       "      <td>17244</td>\n",
       "      <td>21</td>\n",
       "      <td>90.32</td>\n",
       "      <td>24.183211</td>\n",
       "      <td>81</td>\n",
       "      <td>7109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>42</td>\n",
       "      <td>9192</td>\n",
       "      <td>1</td>\n",
       "      <td>2000398934016</td>\n",
       "      <td>0</td>\n",
       "      <td>24791.93</td>\n",
       "      <td>3084.01</td>\n",
       "      <td>24839</td>\n",
       "      <td>30</td>\n",
       "      <td>99.51</td>\n",
       "      <td>20.146430</td>\n",
       "      <td>115</td>\n",
       "      <td>5286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>89</td>\n",
       "      <td>7211</td>\n",
       "      <td>5</td>\n",
       "      <td>3000592982016</td>\n",
       "      <td>0</td>\n",
       "      <td>25327.22</td>\n",
       "      <td>2665.91</td>\n",
       "      <td>8364</td>\n",
       "      <td>26</td>\n",
       "      <td>111.38</td>\n",
       "      <td>20.521995</td>\n",
       "      <td>138</td>\n",
       "      <td>9631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>212</td>\n",
       "      <td>2133</td>\n",
       "      <td>1</td>\n",
       "      <td>2000398934016</td>\n",
       "      <td>0</td>\n",
       "      <td>24893.94</td>\n",
       "      <td>2698.33</td>\n",
       "      <td>17562</td>\n",
       "      <td>26</td>\n",
       "      <td>116.91</td>\n",
       "      <td>22.560980</td>\n",
       "      <td>102</td>\n",
       "      <td>4488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>174</td>\n",
       "      <td>9797</td>\n",
       "      <td>5</td>\n",
       "      <td>3000592982016</td>\n",
       "      <td>0</td>\n",
       "      <td>18711.09</td>\n",
       "      <td>2815.64</td>\n",
       "      <td>16956</td>\n",
       "      <td>23</td>\n",
       "      <td>118.83</td>\n",
       "      <td>33.413798</td>\n",
       "      <td>88</td>\n",
       "      <td>11193</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   date  serial_number  model  capacity_bytes  failure  smart_1_raw  \\\n",
       "0    79           3386      5   3000592982016        0     12688.97   \n",
       "1    42           9192      1   2000398934016        0     24791.93   \n",
       "2    89           7211      5   3000592982016        0     25327.22   \n",
       "3   212           2133      1   2000398934016        0     24893.94   \n",
       "4   174           9797      5   3000592982016        0     18711.09   \n",
       "\n",
       "   smart_5_raw  smart_9_raw  smart_194_raw  smart_197_raw       Temp  \\\n",
       "0      3306.02        17244             21          90.32  24.183211   \n",
       "1      3084.01        24839             30          99.51  20.146430   \n",
       "2      2665.91         8364             26         111.38  20.521995   \n",
       "3      2698.33        17562             26         116.91  22.560980   \n",
       "4      2815.64        16956             23         118.83  33.413798   \n",
       "\n",
       "   Duration    RPM  \n",
       "0        81   7109  \n",
       "1       115   5286  \n",
       "2       138   9631  \n",
       "3       102   4488  \n",
       "4        88  11193  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def remove_outlier(df_in, col_name):\n",
    "    q1 = df_in[col_name].quantile(0.25)\n",
    "    q3 = df_in[col_name].quantile(0.75)\n",
    "    iqr = q3-q1\n",
    "    fence_low  = q1-1.5*iqr\n",
    "    fence_high = q3+1.5*iqr\n",
    "    df_out = df_in.loc[(df_in[col_name] > fence_low) & (df_in[col_name] < fence_high)]\n",
    "    return df_out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#df = remove_outlier(df, 'FAREAMT_BC')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = df.drop(['date','serial_number','model'], axis=1)\n",
    "y = df['failure']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'MinMaxScaler' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-11-1546a7233d17>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mscaler\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mMinMaxScaler\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mX\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mscaler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit_transform\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'MinMaxScaler' is not defined"
     ]
    }
   ],
   "source": [
    "scaler = MinMaxScaler()\n",
    "X = scaler.fit_transform(X)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y = df['FAREAMT_BC'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=123)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(600, input_dim=9, kernel_initializer='normal', activation='relu'))\n",
    "model.add(Dense(400, kernel_initializer='normal', activation='relu'))\n",
    "model.add(Dense(1, kernel_initializer='normal'))\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 158us/step - loss: 412172.7208\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 2s 138us/step - loss: 123312.2438\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 120135.2893\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 2s 118us/step - loss: 119800.9205\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 119719.4186\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 119515.7509\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 119487.6004\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 2s 114us/step - loss: 119604.9010\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 2s 118us/step - loss: 119416.6129\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 119274.5683\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 2s 116us/step - loss: 119218.2293\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 2s 117us/step - loss: 119190.8928\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 119223.1686\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 119457.9521\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 119281.1713\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 2s 123us/step - loss: 119130.2757\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 2s 135us/step - loss: 119181.8387\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 2s 128us/step - loss: 119433.3249\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 118964.9438\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 2s 143us/step - loss: 118839.4680\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 2s 124us/step - loss: 118942.6919\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 2s 137us/step - loss: 119156.6337\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 2s 126us/step - loss: 119118.5150\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 118790.6041\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 119098.5048\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 2s 134us/step - loss: 118761.8771\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 2s 132us/step - loss: 118981.5498\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 2s 138us/step - loss: 118785.7038\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 2s 128us/step - loss: 118990.2184\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 2s 133us/step - loss: 118593.0086\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 154us/step - loss: 118417.1552\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 2s 138us/step - loss: 118895.5623\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 2s 149us/step - loss: 118629.2299\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 2s 131us/step - loss: 118631.4274\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 2s 139us/step - loss: 119037.0615\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 118876.0201\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 118691.4449\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 2s 135us/step - loss: 118640.2828\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 2s 126us/step - loss: 118556.5650\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 2s 139us/step - loss: 118844.7490\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 2s 130us/step - loss: 118766.9679\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 2s 134us/step - loss: 118708.9748\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 118939.7637\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 2s 130us/step - loss: 118350.6335\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 118558.0769\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 2s 124us/step - loss: 118491.0418\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 118472.1787\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 118386.1957\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 118512.1287\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 2s 132us/step - loss: 118484.0221\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 154us/step - loss: 118135.3487\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 118432.3541\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 117886.2062\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 2s 133us/step - loss: 117779.4871\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 2s 139us/step - loss: 117706.6899\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 2s 137us/step - loss: 117897.7066\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 117194.1479\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 2s 134us/step - loss: 116679.6803\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 2s 124us/step - loss: 116357.8938\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 2s 142us/step - loss: 115985.7786\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 2s 135us/step - loss: 115348.3155\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 114197.0428\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 2s 132us/step - loss: 113648.9792\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 2s 144us/step - loss: 112882.3137\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 2s 132us/step - loss: 112394.0135\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 2s 144us/step - loss: 111740.8169\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 2s 131us/step - loss: 110923.1666\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 2s 133us/step - loss: 110516.5091\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 2s 139us/step - loss: 110152.0051\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 2s 131us/step - loss: 109756.6524\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 2s 144us/step - loss: 109727.7381\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 2s 135us/step - loss: 109566.3960\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 2s 140us/step - loss: 109145.0903\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 109171.5832\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 2s 123us/step - loss: 108775.9864\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 108934.3291\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 2s 136us/step - loss: 108396.9029\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 2s 132us/step - loss: 108533.9991\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 2s 140us/step - loss: 108273.7466\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 2s 140us/step - loss: 108388.1876\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 2s 142us/step - loss: 108186.3919\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 108269.6726\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 108059.9969\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 2s 138us/step - loss: 107869.5216\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 107927.3293\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 2s 138us/step - loss: 107960.4992\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 108036.0157\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 2s 129us/step - loss: 107697.9279\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 107192.5641\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 107318.0409\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 107072.4877\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 106944.7767\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 106724.2908\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 2s 131us/step - loss: 107445.1197\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 106754.5943\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 107125.7819\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 2s 123us/step - loss: 107100.4279\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 106488.9766\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 106582.7409\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 105859.9789\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 2s 129us/step - loss: 106061.0248\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 106200.6513\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 105963.8370\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 106062.6561\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 105771.9468\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 105681.6975\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 105249.8273\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 105311.2636\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 2s 128us/step - loss: 105483.6483\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 2s 142us/step - loss: 105169.9622\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 2s 116us/step - loss: 105414.2532\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 105061.8240\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 105250.6469\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 2s 110us/step - loss: 104721.5693\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 104967.8168\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 104505.5189\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 104556.3179\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 104292.2554\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 2s 109us/step - loss: 104109.0958\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 104314.5621\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 104243.6174\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 2s 146us/step - loss: 103781.7614\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 2s 131us/step - loss: 103659.5993\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 2s 142us/step - loss: 103573.8259\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 103414.5264\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 103255.3702\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 103647.1007\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 2s 118us/step - loss: 103416.8120\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 2s 110us/step - loss: 102947.7047\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 102925.0601\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 2s 117us/step - loss: 102807.4663\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 102694.1052\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 2s 116us/step - loss: 102517.9722\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 102393.1564\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 102522.1773\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 102487.2527\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 2s 117us/step - loss: 101857.3433\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 2s 134us/step - loss: 101913.6043\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 101796.1474\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 101575.0889\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 2s 110us/step - loss: 101509.4278\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 2s 124us/step - loss: 101440.6724\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 2s 128us/step - loss: 100947.2565\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 100791.5348\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 2s 127us/step - loss: 100896.8876\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 2s 141us/step - loss: 100418.0275\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 2s 117us/step - loss: 100408.1909\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 100296.9931\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 2s 130us/step - loss: 100154.5282\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 2s 140us/step - loss: 100055.5643\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 2s 128us/step - loss: 99925.4549\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 2s 113us/step - loss: 99486.5731\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 99298.7906\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 99287.4743\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 98852.9263\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 98575.3166\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 98551.3031\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 98721.8053\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 98469.6225\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 97937.1004\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 97968.7993\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 97929.7298\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 2s 113us/step - loss: 97763.3598\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 97599.6601\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 97450.6524\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 97266.8999\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 96782.7244\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 97122.6187\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 96483.4047\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 96385.6043\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 96243.4360\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 96115.4878\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 95911.8322\n",
      "Epoch 174/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 2s 104us/step - loss: 95900.9336\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 2s 113us/step - loss: 95311.0859\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 95698.9969\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 95475.4041\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 95045.7292\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 95384.1591\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 94368.2134\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 94288.8089\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 93958.7710\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 93727.1947\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 93909.4442\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 93415.5313\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 93138.0139\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 93066.6271\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 93111.0952\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 92886.7576\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 92269.5081\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 92330.6389\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 91701.7353\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 91510.8809\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 90951.2078\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 90452.1135\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 90328.1302\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 90356.4823\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 89421.2523\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 89440.2307\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 89015.4219\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 88982.4734\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 88640.8964\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 87813.2548\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 87974.5239\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 87703.2232\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 86983.6017\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 2s 109us/step - loss: 86897.0702\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 86225.9770\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 85924.2423\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 85385.9378\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 85508.3329\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 85321.4927\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 84805.8807\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 84225.9713\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 84036.6596\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 83947.4008\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 83875.6135\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 83258.1083\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 82593.8081\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 82459.7186\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 82196.5431\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 82413.3826\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 82009.6857\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 81833.0353\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 2s 109us/step - loss: 81224.7029\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 80487.6567\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 80752.0566\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 80902.2856\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 80512.4645\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 2s 113us/step - loss: 80213.5408\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 79676.5519\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 79384.6864\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 79098.4972\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 78985.9060\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 78599.3754\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 78503.3620\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 78004.0937\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 78247.3967\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 77013.4643\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 76988.5735\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 77040.7804\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 76932.7809\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 76491.2481\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 76807.7233\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 75927.9357\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 76196.4992\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 75688.9984\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 75010.9115\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 75356.8148\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 74859.5349\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 74936.2215\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 74489.4155\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 73589.7386\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 74143.4409\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 73430.0097\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 73495.1374\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 72853.4207\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 2s 113us/step - loss: 72516.7971\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 72850.7241\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 72543.6742\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 72443.1342\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 71651.0408\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 71414.1122\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 72032.8886\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 71186.8239\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 71151.7905\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 70812.3073\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 70931.5110\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 70591.6452\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 70344.3071\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 70005.0011\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 70102.7755\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 69853.0030\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 69427.0998\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 68927.0007\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 68958.8560\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 69041.5744\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 68956.1519\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 68219.6714\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 68664.0584\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 68735.4295\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 68036.6794\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 67787.3614\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 67636.6243\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 67957.1245\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 67707.7672\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 67590.0991\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 2s 109us/step - loss: 67686.7314\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 66850.2632\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 67305.0632\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 66649.3339\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 66588.0612\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 2s 116us/step - loss: 66281.0942\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 65673.2880\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 66269.2898\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 65728.2582\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 65531.5390\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 2s 109us/step - loss: 65929.8739\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 65372.4852\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 64853.3887\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 65385.1783\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 65086.2410\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 64839.3939\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 64973.6188\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 64940.7444\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 64611.3839\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 64534.0090\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 64094.0176\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 64195.7961\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 64297.9957\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 64176.4529\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 64014.8562\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 63535.4245\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 63436.9854\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 63330.7239\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 63291.1579\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 63704.5872\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 63184.5714\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 62694.8028\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 62764.0552\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 62377.8241\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 2s 115us/step - loss: 63160.1370\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 2s 108us/step - loss: 62549.8620\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 62328.1366\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 62496.2169\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 62056.2845\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 62779.9622\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 61793.6426\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 61873.0290\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 61718.6650\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 61928.8599\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 61313.1992\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 61336.4500\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 61179.1154\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60871.1581\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 61171.4306\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60953.2876\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 61297.8656\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60698.7649\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60915.9913\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 2s 117us/step - loss: 60356.9174\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 2s 114us/step - loss: 60805.2523\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60312.6946\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60531.7901\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 60173.6246\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 60073.3416\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 60249.0617\n",
      "Epoch 348/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 2s 103us/step - loss: 59845.6825\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 59861.6808\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 59975.6842\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 59702.0939\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 59139.8354\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 59872.5055\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 59208.3546\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 59172.1595\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 2s 114us/step - loss: 59274.8307\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 59814.8123\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 58734.6571\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 59197.4431\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 58852.6048\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 58560.1173\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 58555.2011\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 58761.1950\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 58494.5929\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 58046.4020\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 58320.6604\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 58194.8055\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 57821.2661\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 57867.4050\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57677.0910\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 58603.0001\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 57758.9353\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 58190.5839\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57633.0487\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57580.4222\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57452.8743\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 57468.2470\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57957.9556\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 57381.5731\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 57161.4109\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56824.2048\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 56990.2829\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 56507.8886\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 56731.7630\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 56709.3320\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56753.1876\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 56500.2567\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 56556.0541\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56294.6863\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56086.7700\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 56189.7304\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 56012.4734\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56269.1657\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56206.8933\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 55730.7954\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 56456.4127\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 55840.0266\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 56066.9184\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 55457.0508\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 56050.6621\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 55765.9339\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 55621.1723\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 55203.9102\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 55314.6461\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 55289.9659\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 55376.6038\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 55588.8749\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 55009.1947\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 55101.1257\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 54873.9432\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 54742.6874\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 54614.5038\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 54949.0086\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 54731.3088\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 54629.9385\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 54624.7068\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 55020.5880\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 55131.4912\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 54220.9003\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 54080.7026\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 54500.0429\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 54225.3714\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 2s 112us/step - loss: 54130.9965\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 54196.1472\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 53556.6655\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 54106.9726\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 53802.4506\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53744.3552\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53281.5215\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 54469.0837\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53919.3573\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53990.6827\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 53352.6656\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53121.7674\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53204.9286\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 52946.7718\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53365.6137\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53148.0366\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 52802.9022\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53033.4340\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53406.9427\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53077.8236\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 52465.0221\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 53149.9347\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 53263.6506\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 52502.1342\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 52716.7524\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 52819.4406\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 52974.6777\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 52362.1451\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 52163.7681\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 51942.6112\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 2s 114us/step - loss: 51930.0210\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 52062.3509\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 52728.3494\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 2s 105us/step - loss: 52040.8416\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 2s 103us/step - loss: 51960.9836\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 2s 104us/step - loss: 51904.8315\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 2s 111us/step - loss: 52038.9140\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 2s 106us/step - loss: 51894.2938\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 51648.8917\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 2s 107us/step - loss: 51123.1790\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 2s 137us/step - loss: 52083.6016\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 51652.0085\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 51298.2302\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 51439.8912\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 51665.0027\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 51301.7063\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 51068.6511\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 51188.9558\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50727.3314\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 51895.5372\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 50850.1930\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 2s 118us/step - loss: 50833.9769\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 51055.5590\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 51470.1841\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 2s 118us/step - loss: 50985.1940\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 51173.7590\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 51392.1778\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 51307.9036\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 50119.8078\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 50426.3786\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50458.4243\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50634.0393\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 50273.2315\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50952.8396\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50376.7092\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 50327.7515\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50828.2739\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 50445.4288\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 2s 119us/step - loss: 50220.7047\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 49848.1807\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 2s 122us/step - loss: 49940.8287\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 2s 125us/step - loss: 49828.6075\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 49566.5680\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 2s 120us/step - loss: 49925.8264\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 2s 126us/step - loss: 49916.2042\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 2s 126us/step - loss: 49781.5124\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 49678.0024\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 2s 121us/step - loss: 49516.5655\n"
     ]
    }
   ],
   "source": [
    "modeldata = model.fit(x_train, y_train, epochs=500, batch_size=50)\n",
    "model.save(\"model_4.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#from keras.models import load_model\n",
    "#model = load_model('model_3.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1844/1844 [==============================] - 0s 217us/step\n"
     ]
    }
   ],
   "source": [
    "results = model.evaluate(x_test, y_test, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "64672.446854663773"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 51777.7659\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 51300.1416\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 50857.9492\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 50530.0842\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 50555.9959\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 51258.4495\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 50324.6427\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 50275.7624\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 50914.9767\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 51061.0822\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 50189.2449\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 50324.8074\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 50293.1118\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 50793.4980\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 50384.6295\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49791.1426\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 50085.3895\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 50086.3132\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 50079.4594\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 49935.7862\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 50486.0246\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 50292.1375\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49911.3476\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 49752.4147\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 49557.5880\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 49926.7135\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 49304.4384\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 50092.7908\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 50091.0016\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49229.6088\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49363.9713\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 49387.8650\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 49033.5723\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 49525.9237\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 49194.7019\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49141.1881\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 49287.2861\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 48980.3644\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 49180.2868\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 49017.2446\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 48281.1649\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 49858.6522\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 48554.3394\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 48716.9211\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 48459.1157\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 48339.6329\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 48631.1590\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 48457.6741\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 48644.7527\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 47945.8119\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 48337.9659\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 48346.0569\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 48535.6566\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 47878.8654\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 48584.1557\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 48347.5323\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 48350.0144\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 47936.2414\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 47861.1538\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 47462.0667\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 47586.7617\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 47992.9605\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 47799.6142\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 48004.4850\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 47305.6148\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 47110.2421\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 47443.7378\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 47530.7850\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 47585.7577\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 47721.6180\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 47322.2800\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 47202.4275\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 47155.6070\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 46973.3060\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 46885.8050\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 46899.8413\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 46931.4109\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 46883.8811\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 47638.4854\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 46570.7262\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 46400.4447\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 47152.6621\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 46272.7745\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 46741.4687\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 46941.9445\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 46706.9034\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 46418.6156\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 46307.5992\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 46201.7264\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 46138.7825\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 46215.5380\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 46472.2709\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 46062.1806\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 46587.7964\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 45711.7446\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 46700.6564\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 45804.3719\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 45884.2977\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 46325.7453\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 46017.8895\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 46606.3352\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 45666.1053\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 45966.6478\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 45894.6364\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 45284.7407\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 45990.0447\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 45653.3341\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 46111.9256\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 45623.7229\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 45094.4846\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 45628.3872\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 45891.9920\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 45659.1594\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 45619.1498\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 45432.9909\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 44943.3624\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 45278.1975\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 45112.2620\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 44783.1503\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 45520.2688\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 44754.1664\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 45143.0880\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 44905.6592\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 44884.8632\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 44730.6937\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 45004.6255\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 44404.2927\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 44945.5911\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 44465.9203\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 44781.8997\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 44463.8207\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 44716.1326\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 44521.5886\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 44812.0149\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 44344.6446\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 44137.3494\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 44853.1345\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 44266.2418\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 44125.9887\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 44148.0902\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 44509.5326\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 44290.1574\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43898.4074\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 43951.6175\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 43968.7182\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 43798.0100\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 44137.1149\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 44158.4066\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 44075.4237\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 43682.4278\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43307.1404\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 43816.6101\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 44115.4819\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 43985.1564\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 43797.1651\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 43014.1012\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 43316.9351\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43390.4573\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 43201.8890\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43433.4807\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43665.3808\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 43383.7394\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 43067.1258\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 43493.3947\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43201.0674\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 43976.0083\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 43442.5185\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 43264.5178\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 43375.4337\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 43200.7323\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 43146.0063\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 42957.1589\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42789.0926\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 43315.9319\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 42843.8456\n",
      "Epoch 176/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 191us/step - loss: 43083.9884\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42754.1404\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 43080.7419\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42840.3594\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 43131.8942\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42935.4463\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 42422.5405\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 42537.8233\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 42634.3299\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 42782.6987\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42456.7954\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 42586.9480\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 42272.3687\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 42568.5431\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 42493.5861\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 42550.9486\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 42766.9993\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 42482.6761\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 42630.5883\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 42268.2567\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 42063.4064\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 41976.5000\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 42453.3203\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 41833.9717\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 42244.5739\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 42091.5002\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 41445.6514\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 41927.0832\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 42539.3089\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 41537.4496\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 41945.6956\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 41611.7194\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 41446.8147\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 41734.2652\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 41812.6544\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 41840.6835\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 41454.2365\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 41208.7821\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 42011.2444\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 41749.3423\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 41140.8284\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 41455.8433\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 41063.5935\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 41635.0338\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 41386.6605\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 41243.6329\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 41826.6782\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 41455.0337\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 40893.2387\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 40979.0841\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 41055.1804\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 41355.7929\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 41113.6560\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 41050.6839\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 41048.2114\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 41027.5518\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 41134.0759\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 41119.1579\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 40640.2409\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 41240.1045\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 40868.1493\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 40919.2735\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 41398.1389\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 40649.6236\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 39939.6140\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 40534.2086\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 40194.1714\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40780.8943\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40166.2287\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40174.4014\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 40487.4245\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 40680.4368\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40290.8089\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 40660.3382\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 39982.0003\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40219.1424\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40165.1460\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 40780.4423\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 40304.4351\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 40400.4768\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 40417.4188\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 39705.9745\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 39663.9065\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 39889.6410\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 40386.5152\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 39723.1095\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 39768.9568\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 40195.7144\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 39614.0575\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 40017.5344\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 39566.6072\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 39977.8380\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 39407.7595\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 39792.7113\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 39739.3810\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 39342.6003\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 39481.7055\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 39767.5946\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 39752.9146\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 39395.9204\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 39329.2409\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 39226.9511\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 39108.2250\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 39423.0494\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 38933.6814\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 38918.1509\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 39329.9815\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 39277.7995\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 38836.1810\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38806.4990\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 39097.9282\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 38798.9464\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 39003.5992\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 38643.3199\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 38861.3831\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38812.1212\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 38846.4940\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 38767.9783\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 38907.7977\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 38930.5729\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38908.0187\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 38633.9311\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 38820.5536\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38575.3799\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 38644.5736\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 38353.9965\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 38498.2061\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38226.1823\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38448.9028\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38301.7791\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 38471.6407\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 38371.8352\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 38510.9654\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38177.8659\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38359.8179\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 38196.9040\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38012.3780\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38423.0687\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38197.8499\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37985.4015\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 37744.6049\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 37900.8229\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37928.6957\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 38754.5204\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37745.9228\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 38072.8198\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 38410.9874\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 37794.7167\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37230.5785\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37938.8068\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37508.7124\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 37941.2438\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37558.8948\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37334.1191\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 37319.5001\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 37629.9228\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37594.9978\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 37827.8539\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37463.6185\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 37264.4380\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 37421.3056\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37636.3204\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37032.6399\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37467.1987\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37438.5523\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37316.3902\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 36941.6028\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37215.5172\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 37043.1318\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 37021.4453\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 36965.9474\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 37414.8737\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37023.1004\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 37206.8930\n",
      "Epoch 350/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 189us/step - loss: 37059.8878\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 37095.7639\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 37065.4707\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 36929.5271\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 36962.0020\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 36547.4675\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 37211.8020\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 37045.7761\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 36631.3947\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 36901.2025\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 37473.8921\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 36726.2048\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 36500.0263\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 36379.5177\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 36863.9805\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 36439.3444\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 36809.6923\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 36732.6725\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 36790.9153\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 36326.0639\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 36498.2854\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 36413.5422\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 36284.7051\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 5s 275us/step - loss: 36617.0595\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 36500.5274\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 36723.0516\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 35997.8116\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 36190.7857\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 36254.7131\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 35915.9520\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 35994.1472\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 36127.9388\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 36032.1075\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 36249.8210\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 36102.3872\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 36340.7071\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 36063.8339\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 35807.8855\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35833.9973\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 4s 239us/step - loss: 36037.6311\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 35429.9121\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 35888.6334\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 36001.8821\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 35640.2407\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 35835.2984\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 35826.0885\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 36013.1081\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 35562.8522\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 35740.4444\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 35910.1341\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35621.4358\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 35479.3254\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35709.2409\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 35434.4956\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 35291.2800\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 5s 275us/step - loss: 35430.1942\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 35598.9085\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 35341.9100\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 35162.8157\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35328.3493\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35345.2105\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 35151.9182\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 36002.8112\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 35645.4515\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34925.4201\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 35086.8780\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 35087.6168\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 35306.8469\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35291.5446\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 34870.3867\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34904.8735\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 35247.1580\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34903.8630\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 35105.0601\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 35383.0087\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 34933.0104\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 34705.7267\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 34561.4849\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 34692.9182\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34599.0938\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 34748.7935\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 34968.1169\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34425.1696\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 35207.3367\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34633.1196\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34649.3706\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34856.0079\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34680.5587\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 34344.3504\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 34945.9228\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34610.1596\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 4s 256us/step - loss: 34860.3652\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 34504.9847\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 34365.7883\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 34078.6175\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34052.2066\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 34885.0466\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34369.8499\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 34368.7667\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 34275.6032\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33975.2159\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 34322.1096\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 35046.0839\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 34310.2196\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 33992.7113\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 34096.3989\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 34434.1540\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34006.0043\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 34022.6921\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 34336.3810\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 4s 271us/step - loss: 33654.7439\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 33844.9889\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 34019.5419\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 5s 275us/step - loss: 34314.5911\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 33995.2793\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33806.5404\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33784.6159\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34001.0910\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 33680.8057\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 4s 263us/step - loss: 33784.8094\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 33563.1308\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34338.0673\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33782.3027\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 34037.7914\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 33507.2827\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 33860.8850\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33245.9445\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33830.2913\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 5s 274us/step - loss: 33784.3406\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 33478.6707\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33494.1116\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 33805.8443\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33589.2291\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 33815.7937\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33272.0175\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33147.9547\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33287.6039\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33648.6571\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 33422.6733\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33488.4257\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 33373.7510\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 33497.1492\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33026.0179\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33634.3963\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33106.7176\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 32957.4880\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33231.2086\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33594.7890\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 32788.7474\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 32965.4421\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 32929.3307\n",
      "1844/1844 [==============================] - 0s 57us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34897.6284\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34843.1789\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 35116.4090\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 34335.0732\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 34252.0118\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 34137.2853\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 34284.9775\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 33963.6443\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 34203.1870\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 33810.6370\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 34100.4656\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 33928.3456\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 4s 241us/step - loss: 33992.1204\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 4s 230us/step - loss: 34288.0317\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 34283.9660\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 33760.8295\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 33608.6030\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 33551.2025\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33499.2711\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 33435.1787\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 34161.7034\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33357.1325\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33475.7353\n",
      "Epoch 24/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 193us/step - loss: 33710.7100\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 33553.8546\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 5s 283us/step - loss: 33373.2820\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 33546.8681\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 33302.3399\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 33398.8230\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 33155.5454\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33402.9434\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33600.1211\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 33428.8034\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 33410.4133\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 33433.2757\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 33080.5136\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 33405.4843\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 32837.7366\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 32743.5752\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 32990.5324\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 33056.6145\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33032.3093\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 32755.1486\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 32812.8957\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 33425.7246\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 32723.5263\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 32884.8157\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 32943.7251\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 33293.8823\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 32507.2221\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 32828.2727\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 32532.6818\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 32556.4255\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 32851.8827\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 32608.9298\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 32461.5924\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 32646.3836\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 32510.6210\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 32067.9594\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 32367.8112\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 32588.4875\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 32301.7682\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 33114.4786\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 32187.0685\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 32111.6041\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 32002.7071\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 32223.6766\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 211us/step - loss: 32397.9979\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 4s 268us/step - loss: 32089.0012\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 4s 215us/step - loss: 32388.2890\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 31980.1142\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 31855.0628\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 32127.6378\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 32177.9868\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 32040.7301\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 32611.8151\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31964.8958\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 5s 281us/step - loss: 31838.1087\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 32188.2274\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31781.8311\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 31847.8230\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31802.0012\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31873.7356\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 32308.7171\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 31646.6821\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 31682.1782\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 31670.1034\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31629.2330\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 31749.3456\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 31255.7008\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31464.0769\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 31869.4689\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 31606.5356\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 5s 280us/step - loss: 31696.7832\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 31624.2349\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 31429.6632\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31365.5207\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 31397.4241\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31199.4460\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 5s 274us/step - loss: 31453.9614\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 31183.4643\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31336.9659\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 31686.1705\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31487.9973\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 31068.1312\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31381.6314\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 31533.6032\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 31253.6434\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 31083.9754\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 31270.4506\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 31287.2627\n",
      "Epoch 112/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 190us/step - loss: 31014.9682\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 31004.0096\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 30583.3696\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 31570.7753\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30777.3059\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 30853.5023\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30925.4900\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 30816.8725\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 30862.1132\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30934.9242\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 30936.5985\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 30894.4763\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 30661.5043\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30836.3300\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 30783.5668\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 30771.2181\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 30561.0382\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 5s 288us/step - loss: 30298.9806\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 31061.3573\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 30690.9339\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 31250.6664\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30373.7080\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 30467.2293\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 30335.8883\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30898.9401\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 30351.5755\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 30813.1672\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 30956.8408\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 30490.4549\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 4s 237us/step - loss: 30175.5560\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 30128.5985\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 30585.8203\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 30321.0040\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 30582.9213\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 30367.9951\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30112.3382\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 30389.4583\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 30474.9465\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 30287.5733\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 30075.0480\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 30191.7186\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 30402.3436\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 29863.1663\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 30208.0763\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 29947.4120\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 30315.5682\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29728.4646\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29879.3361\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 30659.2389\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 29827.1200\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29688.6546\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 29814.1490\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29768.4882\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 30107.6272\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 29999.0601\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29936.0848\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29790.5942\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 29739.9720\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29694.8660\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 29811.8548\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29718.3121\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 4s 247us/step - loss: 29758.8565\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 4s 220us/step - loss: 30192.9661\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 29211.4209\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 5s 293us/step - loss: 29757.6447\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 29490.0611\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29467.8561\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 29781.9545\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 29669.0704\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29485.8885\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29597.2774\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 29292.6503\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29499.5047\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 29596.4852\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 29435.2196\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29181.3712\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 29391.5658\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 29508.8999\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29118.1701\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29536.5396\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 29122.6695\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 29427.3393\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 29604.3124\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 4s 222us/step - loss: 29195.5495\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 4s 258us/step - loss: 29212.9125\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 29312.0771\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 4s 257us/step - loss: 28993.2332\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 4s 214us/step - loss: 28995.8303\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 29111.6304\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 29085.2142\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 28785.3281\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 29247.2715\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 4s 241us/step - loss: 29479.6206\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 4s 221us/step - loss: 29055.7050\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 28955.2983\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 28681.4910\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 28932.6676\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 29256.5706\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 28861.0598\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 28733.4337\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 29265.5000\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 28862.0237\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 28789.9820\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28924.8907\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 5s 289us/step - loss: 29114.2583\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 28552.8994\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 4s 254us/step - loss: 28907.2183\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 4s 239us/step - loss: 28581.2749\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 28744.8527\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 28479.3133\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 28685.3978\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 28566.3960\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 4s 238us/step - loss: 29060.4051\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 28442.9120\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 28663.7759\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28816.0163\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 28759.9914\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28450.9185\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 28481.4431\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 28538.3767\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 28703.6512\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 28645.7067\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 28249.8258\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28520.2700\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28478.7305\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 28090.4609\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28289.8662\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28657.3692\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 28166.9722\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 28299.9999\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28405.3465\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28386.0110\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 28217.3544\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 28295.8873\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 28150.0087\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 28234.1938\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 28148.2404\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 28256.3399\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 4s 221us/step - loss: 28169.6480\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 4s 251us/step - loss: 28057.1101\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28032.5347\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27840.3622\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 28274.6630\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 28004.3771\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 27990.5368\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 28184.0800\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28058.4707\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28334.9348\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 27904.0785\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 28300.6282\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27539.8283\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27959.2125\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27904.2989\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 27954.4777\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 27814.8260\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 28051.5809\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 28098.7379\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 27859.2327\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27698.3484\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 27467.6758\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27861.9523\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 27894.5347\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 27960.5741 0s - loss: \n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 27138.1801\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 27769.6895\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 27393.8004\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27622.0249\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 27716.6994\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27782.1015\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 27752.7610\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27482.7039\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27507.8958\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 27677.5405\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27438.0306\n",
      "Epoch 286/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 197us/step - loss: 27621.5254\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 5s 291us/step - loss: 27508.6361\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 27410.5939\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 27257.6230\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 27808.8674\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 27468.9780\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27516.0740\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 27367.8575\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27607.7451\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27301.5081\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 27420.1512\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 27136.6766\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 27210.0726\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27388.5309\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 27045.2641\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27662.8950\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 27185.3786\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27228.4275\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 27267.6710\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27730.6653\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 27410.4242\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 27143.9159\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26951.8908\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 5s 291us/step - loss: 27043.2720\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 27151.6539\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26800.0942\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 27681.5887\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 27098.1691\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 27010.6914\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 27299.6632\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26998.6565\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 26733.8126\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 27183.7469\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26680.1878\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 26991.4791\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 5s 278us/step - loss: 26677.1816\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 4s 239us/step - loss: 27028.9229\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26849.9875\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26898.1351\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 27083.8856\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26775.8951\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 27037.9634\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 27230.5208\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26899.3554\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26712.0961\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26451.6999\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 26417.6654\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26672.0927\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 6s 373us/step - loss: 26834.4572\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 27066.3410\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26559.2644\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26574.3082\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 26394.1707\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26639.3408\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 26385.3151 0s - los\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26987.8885\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26603.2307\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26755.2387\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 26435.4985\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26484.7630\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 26669.6053\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26412.1700\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 26420.5182\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 26562.0844\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26633.2058\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26698.7347\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26270.2054\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26378.9367\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26137.5277\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26559.7322\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26347.4555\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26393.9876\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26243.7342\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 26888.3158\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 26325.0070\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 26630.1356\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26224.7282\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26215.9561\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26400.0182\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 4s 239us/step - loss: 26357.2662\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 26110.9043\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26120.2657\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26211.1133\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25915.9627\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 26341.8616\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26175.2476\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26469.8386\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26336.3979\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25821.8710\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25646.0260\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26646.3628\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26524.1199\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26166.5775\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 26042.7477\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 25922.9638\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25986.8757\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 26210.0689\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25825.8037\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 25979.5919\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25685.0441\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26239.6259\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25668.0817\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 26144.0693\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 26158.0198\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25752.6786\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 26008.5295\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 25846.2502\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 25776.3340\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 26150.4581\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25921.0750\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25918.1875\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 26172.5007\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25926.1293\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25650.1010\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 25528.7413\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 5s 294us/step - loss: 25584.1726\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 26187.0402\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 5s 306us/step - loss: 25660.1314\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 25726.3511\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 4s 224us/step - loss: 25883.0425\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 4s 224us/step - loss: 25492.1074\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25632.8912\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 5s 293us/step - loss: 25722.1517\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 207us/step - loss: 25900.0728\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25598.7756\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 25420.7297\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25727.2362\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25717.1687\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25582.3768\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 25493.3851\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 25661.6376\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25126.1401\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25276.3538\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25719.9912\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25285.5500\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25620.1195\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25939.3528\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 25653.5938\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25357.0996\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25553.6360\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25606.9530\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25510.1052\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25442.2461\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25455.9710\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 25539.5224\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25580.4391\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25267.3077\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 24933.9231\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 5s 274us/step - loss: 25370.6215\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 25453.7105\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25474.3910\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25517.3893\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25082.7665\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25352.5093\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 25302.7516\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25311.3570\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25361.1215\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 4s 252us/step - loss: 25539.5747\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 25237.2924\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25001.9495\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25460.5941\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24940.9728\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25291.1945\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25281.8582\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25096.8876\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25079.4128\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25124.7353\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24903.7389\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25482.7888\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24914.9471\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25022.8255\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25042.4909\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24586.0997\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25086.8184\n",
      "Epoch 460/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 189us/step - loss: 25104.0976\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25460.1903\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 24662.5475\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24976.7755\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25010.9641\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 24950.3224\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24824.0815\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 25430.9133\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24878.4580\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 5s 271us/step - loss: 24845.3559 0s - loss: 24826.76\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 24878.5468\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 25019.6485\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 5s 331us/step - loss: 24743.5307\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 24726.6532\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 24902.8392\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 24653.3315\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24848.2655\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 24716.9134\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 4s 243us/step - loss: 24959.3846\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24641.9053\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24486.4138\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 24805.5902\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 25034.3422\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24722.5051\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 24911.1499 1s \n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25021.3000\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24488.6687\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24362.2815\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 24697.6701\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24452.3753\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25008.1777\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24530.3246\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24638.0212\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 24201.3983\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25014.1050\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24544.4138\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24746.9990\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 24411.9903\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24750.9185\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 24549.7711\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24335.5690\n",
      "1844/1844 [==============================] - 0s 50us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 26737.7401\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25998.4641\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 4s 217us/step - loss: 25514.4530\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 5s 271us/step - loss: 25781.3470\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25613.4468\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25643.4714\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25561.8140\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25545.9821\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25455.4871\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25625.5533\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25567.3373\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25562.8017\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25435.8423\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 25588.5493\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25355.7643\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25076.9950\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 25422.6242\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25735.6721\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 25636.0159\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25509.5770\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 25625.0130\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 24957.7003\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 25501.0065\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25401.9613\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25421.2910\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25815.1551\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 25118.6002\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24974.1948\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25413.8146\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 25397.9121\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 24820.3027\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 4s 214us/step - loss: 24977.3162\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 5s 319us/step - loss: 25277.5401\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 24805.7944\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 25407.5301\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 25083.0057\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 25200.9786\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 25069.3726\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 25127.9986\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 25180.5740\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 24933.6301\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 4s 236us/step - loss: 25360.8950\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 25206.8431\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 24777.9601\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 5s 301us/step - loss: 24802.8951\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 24697.7206\n",
      "Epoch 47/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 191us/step - loss: 24956.5927\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 25194.2177\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24898.7975\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24706.5378\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24724.7939\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24789.5225\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24910.7212\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24485.4483\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24722.2479\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24528.8836\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 24853.3571\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24400.0571\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 24633.8352\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24636.8972\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24680.6209\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24320.4218\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24442.3669\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24776.7443\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24798.5126\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24712.6709\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 25061.4719\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24331.4301\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24659.9454\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 6s 340us/step - loss: 24432.7470\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 24271.3212\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24236.3485\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24687.7018\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24508.3410\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24135.4191\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 24181.6657\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24292.3562\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 24306.1567\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 5s 272us/step - loss: 24498.9563\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 24160.2469\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - ETA: 0s - loss: 24649.143 - 3s 185us/step - loss: 24651.8988\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24118.2231\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24025.2907\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 24353.6926\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24033.1465\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 24540.6830\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 23979.6727\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 24312.7629\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 24384.7587\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24159.1999\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24598.7275\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24023.6165\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 24101.7299\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24133.4951\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 24025.8876\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24108.3124\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24442.0759\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24073.4924\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 24191.4912\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 24244.7638\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 24072.2894\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24020.0005\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23962.1212\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 23756.9066\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 5s 284us/step - loss: 24001.0415\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 23773.0588\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 5s 281us/step - loss: 24111.2613\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 24396.0856\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 23470.6942\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 23748.2203\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 24189.3209\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 24056.4642\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23851.1250\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 24428.0984\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23994.7790\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 23756.4280\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 23688.3505\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23814.8483\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 4s 212us/step - loss: 23623.9440\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 5s 293us/step - loss: 23421.3986\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 23891.3571\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23812.0058\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 23823.9368\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 23725.3267\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23743.8464\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23560.6751\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 23495.1225\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 23585.5876\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 23416.1567\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 23591.0957\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23335.3156\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 5s 331us/step - loss: 23757.0226\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 23727.3282\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 23935.8014\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23263.2975\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23524.6919\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 23667.4709\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 5s 273us/step - loss: 23411.4663\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23344.2385\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 23219.5187\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23589.2789\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23598.1013\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 22987.8824\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 23271.2086\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 23700.6313\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 23184.5011\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 23456.1100\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23314.8737\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23294.2434\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23308.3927\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 23303.2319\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 22965.1292\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 23073.4524\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 23553.2497\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 23021.0688\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23235.5194\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 23264.7313\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 22841.9072\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 23093.4611\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 23280.6738\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23211.3168\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 23219.0471\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 22997.9599\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 22922.3605\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 23509.5244\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 4s 257us/step - loss: 23299.0017\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 4s 240us/step - loss: 22539.9961\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 23165.8909\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 5s 312us/step - loss: 22927.9136\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 23197.9855\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 22624.8433\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 23133.1574\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 23155.2450\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 22979.0805\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 23009.9665\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 22949.4971\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 23301.1488\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 5s 283us/step - loss: 22621.1793\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 23080.1865\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 22571.0661\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22788.6800\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 22517.6716\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 22876.3171\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22523.8659\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 22820.3376\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 22705.6578\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 22497.3319\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22689.2758\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22474.4869\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 22681.6350\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 22311.1914\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 22796.0265\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 22686.4582\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 23082.9347\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22464.8485\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 22506.7678\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 22597.4437\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22481.0157\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 22286.3931\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 4s 222us/step - loss: 22557.3370\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 4s 267us/step - loss: 22615.2838\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 22497.5845\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 22606.0870\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22464.0200\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 22949.5252\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22317.0471\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22377.5461\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 22431.6869\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 21952.3064\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 22325.6757\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22455.2444\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 22125.7114\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 22144.3370\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 22358.8815\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 22607.9814\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 22438.2099\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22131.3665\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 22097.4409\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21922.9483\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 22099.3841\n",
      "Epoch 221/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 186us/step - loss: 22398.8694\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22087.4568\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 21817.3757\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 22363.6230\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22204.0358\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 22115.7995\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 22035.9155\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 22296.5522\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 22342.2446\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21819.5199\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 21910.9463\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 21986.6896\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 22196.5518\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 21799.5129\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 21966.3073\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21529.0687\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21836.7640\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 21940.3057\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 21775.1904\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 21976.4457\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 21783.9236\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 21558.0018\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 5s 278us/step - loss: 21897.7307\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 211us/step - loss: 21646.8795\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 21750.7286\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21552.6910\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 21456.5581\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 21649.0589\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 21927.4377\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21437.2188\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 21433.0220\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 21285.6199\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21766.2503\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21576.0892\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 5s 290us/step - loss: 21660.7144\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 4s 217us/step - loss: 21951.1840\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 21682.8173\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21354.3974\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21609.6024\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 21390.4925\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 21591.8979\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21538.0316\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 21100.4223\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 5s 307us/step - loss: 21772.5802\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 21270.8112\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 21492.1201\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21055.9364\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21309.3320\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 21217.2642\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 21370.3819\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 21283.1485\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 20815.7562\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 21304.2999\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 4s 269us/step - loss: 21269.8086\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 21332.9398\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 21128.0496\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 21151.4387\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 21149.4063\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 4s 242us/step - loss: 21096.3042\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 4s 252us/step - loss: 21336.3206\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 21342.6164\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 21018.1164\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21240.9156\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 21084.0584\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 5s 303us/step - loss: 20822.2943\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 20947.5636\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 21074.7777\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20953.0874\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20965.0747\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 21220.2582\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 5s 278us/step - loss: 20716.1706\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 20981.8980\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20968.0056\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 21097.5009\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 20710.3374\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20740.1294\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20728.3790\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20704.5126\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20968.0004\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20986.4232\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20900.2008\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20775.4449\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20791.3826\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 20601.8421\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20908.4687\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 21532.9406\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20764.8496\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 20101.6856\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20609.0459\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20729.4457\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 21043.1337\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20558.5119\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20569.8427\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 20559.8035\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 20408.0239\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 20734.5005\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20887.2151\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 20798.4462\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 20204.6158\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 5s 286us/step - loss: 20764.1830\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 3s 211us/step - loss: 20787.1529\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 20637.1985\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 20355.0470\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 20352.9656\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20289.2577\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 20556.7967\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20839.3120\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20201.8383\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 20681.7469\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20515.6657\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 20610.4724\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 19900.5920\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 5s 278us/step - loss: 20091.0399\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 20352.4660\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 20243.2081\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20383.9799\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 20115.5184\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20415.3793\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 20271.2292\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 19961.6535\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19966.3698\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 5s 325us/step - loss: 20113.3719\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 19950.5351\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 20120.7874\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 20094.1013\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 19873.0384\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20437.7959\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 20043.0625\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 19858.0392\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19836.3026\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20143.1896\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 19977.7478\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 19551.7195\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19629.1157\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19654.6271\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 20101.9795\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 20473.2857\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 4s 248us/step - loss: 19887.6245\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 19849.3923\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 19834.1380\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 5s 321us/step - loss: 19656.7743\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 19692.7615\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 4s 228us/step - loss: 19796.3509\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 19772.2488\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 19823.1772\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 5s 293us/step - loss: 19585.8752\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 19574.6422\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 4s 264us/step - loss: 19459.7790\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 4s 241us/step - loss: 19516.3442\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 19525.8283\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 19571.4151\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 19699.7764\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 19720.2611\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 19679.1331\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19808.5473\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19651.1871\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19248.6431\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19666.0656\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 19428.1203\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 19306.8027\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19591.3450\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19226.2793\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19492.2640\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 19606.5109\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 19537.0420\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 19363.7768\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 19299.3671\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 19200.7295\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19084.7360\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 19050.8763\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 20122.9529\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 18969.6324\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 19616.9213\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 19054.8926\n",
      "Epoch 395/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 184us/step - loss: 19195.4144\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19219.3339\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19085.0552\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 19074.4090\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 19014.1509\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 19493.0765\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 19182.4404\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18862.1024\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18691.7974\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18959.1193\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18795.9242\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 19127.4968\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 19243.4699\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19083.5640\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18925.4391\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18632.9491\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 5s 294us/step - loss: 19249.6575\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 18749.3073\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 5s 285us/step - loss: 18851.6093\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 18799.9354\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18835.5632\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 4s 240us/step - loss: 19243.7768\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 18921.2716\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18570.9871\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 18398.7570\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 18862.5473\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18795.5971\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18827.8270\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18666.6951\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18418.7109\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18760.8472\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18753.3684\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18734.8082\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18341.7162\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18493.9674\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18783.8914\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18585.3840\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18589.6655\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 18216.8128\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18716.3276\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 18431.7640\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 4s 230us/step - loss: 18832.0548\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18443.6127\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17993.2268\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18509.8768\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18274.3355\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 18742.9014\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 5s 288us/step - loss: 18011.4762\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 18488.8547\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18506.0727\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18374.8932\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17928.7036\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18641.8259\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18342.6473\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 18180.4588\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18459.0741\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18196.0499\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18150.4518\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18062.3501\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 5s 290us/step - loss: 18321.9718\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 18333.3738\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18446.5032\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 18046.1849\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 18321.1181\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17968.5019\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18212.6944\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18035.5104\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 17671.5392\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18098.8104\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17966.0840\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 18251.3898\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 18276.8927\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 5s 325us/step - loss: 18248.2953\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 17748.2008\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 18352.8364\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17815.0164\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 18634.1677\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 17561.5607\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18009.8942\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 17930.8160\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18033.0379\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17625.7020\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17805.5048\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17855.7859\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17655.3451\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18045.7988\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18173.4406\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 17720.0792\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17984.0718\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 17608.3809\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17579.3069\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 17652.1625\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 18005.4698\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 18132.4580\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17647.0310\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17492.5991\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17424.8996\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17799.1662\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 17292.2964\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17856.9542\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17604.6873\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17407.1502\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 17641.6601\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17424.8711\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17944.3015\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17229.4234\n",
      "1844/1844 [==============================] - 0s 41us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 19530.9858\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 19356.0817\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 19083.0331\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 18840.3858\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18992.9218\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18703.1106\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 18852.2407\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18935.1280\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18990.4109\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 18351.8780\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18687.8696\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18320.1485\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 18356.2943\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18294.2166\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18556.5736\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 18532.3879\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18195.4990\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18178.2805\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 18321.7618\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 18361.1754\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 18525.8744\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 18069.8369\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 18470.6427\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17830.4777\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 18161.8402\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 18401.1617\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17672.5159\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 17853.3155\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 5s 299us/step - loss: 17623.3839\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 17962.1077\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 17989.9550\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17776.2254\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17927.2032\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 17641.8025\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17990.4751\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17943.0555\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 18459.9167\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 17845.8469\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17482.4501\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17471.2990\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17778.8478\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17788.4597\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17628.1512\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17549.1114\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 5s 309us/step - loss: 17151.9959\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 17393.0894\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 17543.4491\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 17804.9708\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 4s 220us/step - loss: 17677.9957\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 4s 264us/step - loss: 17452.0114\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 17170.2144\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 17418.9140\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17511.9190\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17311.4757\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17700.1725\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 17525.7677\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 17475.3928\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17217.8477\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17208.3019\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 17728.6118\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 17252.8777\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 5s 321us/step - loss: 17246.4350\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 17147.6415\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 17043.6900\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 4s 249us/step - loss: 17101.3355\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 17197.1021\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 17074.9500\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 17252.1928\n",
      "Epoch 69/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 187us/step - loss: 17215.7886\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17404.1777\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 17177.4497\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 17032.1207\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 17260.8442\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 17124.7426\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 16660.9646\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 17006.6276\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 16798.4662\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 17171.1224\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 16937.6504\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 16913.6272\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 16997.6822\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 17002.0741\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 16398.8151\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16862.7427\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 16999.0821\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16421.6819\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 16732.4973\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 16734.0520\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 17101.0762\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16726.6013\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 16287.1811\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16677.9453\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16829.6676\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16309.9297\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16563.7211\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 16339.4123\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 16671.5325\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 4s 237us/step - loss: 16606.0170\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 16476.8220\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 16404.8152\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16565.3867\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16500.8282\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 16533.7906\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 16156.0604\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 16615.2263\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16754.7957\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 16341.8059\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 16778.4263\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16230.0069\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16334.9149\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16055.7518\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 16387.4816\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16266.7013\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 16132.5639\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 16426.6968\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 16036.6029\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16045.6269\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 16083.7111\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 16441.3578\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 16188.5030\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 15997.4835\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16288.6008\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15989.6338\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15894.5579\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 16088.0766\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15963.1833\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 16203.8532\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 16205.4640\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 16072.4875\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16208.5590\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 15728.9931\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15919.4990\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 15624.3347\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 16007.4061\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15947.2781\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 16023.7718\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 4s 251us/step - loss: 16019.9507\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 15943.6245\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 15894.7147\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 15697.7028\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 15820.1402\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 15823.3123\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 15667.4087\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15614.8085\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15956.1032\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15343.3754\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 15899.9653\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 15800.4237\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15672.2958\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 15446.3593\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15676.2281\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 15625.7198\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15495.3865 0s - loss: 15378.\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 15826.3955\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 15522.2029\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 15076.8161\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 15507.6394\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 15446.4931\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 15395.1501\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 15549.7581\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 15286.8298\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15380.9141\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 15391.9690\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15445.3717\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 15338.9053\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15084.8028\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 15159.7501\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15468.7545\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15437.3814\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 15049.7232\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 15185.0817\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 15438.8469\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15384.7099\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 15083.1973\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 15325.5437\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 15607.5101\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 4s 244us/step - loss: 14758.7048\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 15317.1829\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 15129.8249\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 14875.3956\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 15065.6335\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 15084.3208\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 15335.1700\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14989.2126\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 15266.9484\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 15230.4842\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14947.2023\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 15216.2091\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14924.7233\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 14751.2105\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14795.4525\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 15046.4888\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 15207.7645\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 14654.8532\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 14782.8098 0s - l\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 4s 254us/step - loss: 14935.7308\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14713.5267\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 15115.4449\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14732.2081\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14888.5433\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 5s 276us/step - loss: 14828.4440\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 207us/step - loss: 14626.1401\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 14998.7877\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14829.9232\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14878.9684\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14962.6707\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 14696.6492\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14770.9463\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 14776.6641\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 5s 298us/step - loss: 14671.0905\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 4s 212us/step - loss: 14721.0008\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 15248.8454\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 5s 303us/step - loss: 14430.8611\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 14401.3484\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14469.3298\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 14833.8843\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14824.8293\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14486.1975\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 14585.1698\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14436.7099\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 14460.7002\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 5s 277us/step - loss: 14553.9211\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 14556.4862\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 4s 268us/step - loss: 14491.5711\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 4s 237us/step - loss: 14599.9883\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14384.6999\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14130.3990\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14589.5826\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 14591.9127\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 14369.6813\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14216.2322\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14297.1192\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14334.4862\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 14428.1439\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14465.9835\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14619.7863\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14257.8484\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14242.0689\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14288.9282\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 5s 290us/step - loss: 14417.9844\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 14239.9361\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14170.5334\n",
      "Epoch 243/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 184us/step - loss: 14248.5018\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 14501.5020\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14580.4486\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 14704.9876\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14246.9045\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 14070.9084\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13941.4582\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14243.9254\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13974.4602\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13961.1152\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 14054.2984\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14028.9011\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 14125.2841\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 5s 323us/step - loss: 13808.2401\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 14198.6210\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13509.3332\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 13983.9982\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 13984.3483\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13847.1513\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 14084.0928\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 13864.7456\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13888.4847\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 4s 269us/step - loss: 13761.5402\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 13899.0127\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 13710.4231\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 13744.0649\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 14111.6138\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13971.8957\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 13633.7760\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 13539.1992\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 14146.3557\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13542.1383\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 13958.4255\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13789.4689\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 5s 327us/step - loss: 13812.9583\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 14087.6620\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 14018.8729\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13441.0871\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 13508.0126\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 14103.2305\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 13476.0033\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 13509.7116\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13373.6486\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 13599.7135\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 13835.9485\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13598.3933\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 13642.6035\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13481.7470\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 13539.2561\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 13489.0139\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 13649.3205\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13400.9776\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 13223.9699\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 13668.9751\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 13792.2936\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 4s 266us/step - loss: 13390.3016\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 4s 216us/step - loss: 13625.6078\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13513.0770\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 13400.9461\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13387.9163\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 13600.3365\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 13554.3107\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13454.7595\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 13354.4533\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 13019.4783\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13392.9874\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 13368.8949\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 13352.3489\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 13608.6606\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 13270.7983\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 13101.7385\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 4s 268us/step - loss: 13860.4478\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 4s 222us/step - loss: 13451.4781\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13300.7149\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12902.2944\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 13390.7871\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 13145.2663\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 13297.6006\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 13017.6759\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13150.8834\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13116.2405\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 13452.7366\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13040.2497\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 13376.1101\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12916.4154\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 13216.9150\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 13073.5675\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12867.4333\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13318.8460\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 13281.7562\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 4s 226us/step - loss: 13098.9953\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 12982.7512\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 13630.5172\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 13020.3975\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 13032.1552\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 4s 229us/step - loss: 13200.7259\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 12864.8908\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 12980.8376\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13062.0382\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12996.8684\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12751.8809\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13154.3247\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12906.7921\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12886.3913\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12862.2688\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 4s 233us/step - loss: 13471.2041\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12881.5675\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12814.6688\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12602.5906\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12736.6973\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12760.0132\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 13235.8051\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 12798.0502\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 12685.8692\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13113.0750\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12852.8770\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12853.8811\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12849.7616\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 13581.7991\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 12517.6735\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 12817.7699\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 12564.6803\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 13051.1294\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12572.9407\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12801.2754\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12561.4321\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12901.2501\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 12572.2121\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12765.9383\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12528.7748\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12687.1202\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12672.6558\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 12515.8398\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12577.6471\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12554.3665\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12642.6053\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12335.0295\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 12428.2410\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12609.5288\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12345.1472\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12367.7842\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 12622.7969\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 13195.9289\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12589.8313\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 12410.3058\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 12296.1001\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 207us/step - loss: 12579.0916\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12380.5937\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12336.2599\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12367.8479\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12652.8976\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12345.4945\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12435.9407\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12428.0924\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12549.2575\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12175.5941\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12278.6165\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 12060.6928\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 4s 213us/step - loss: 12296.3878\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12119.2508\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 12505.1872\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12255.2034\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12268.4820\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 12195.2481\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12469.7231\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12490.8746\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12017.6487\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12264.8427\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12260.4195\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12233.1622\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12074.9854\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12264.2234\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 12242.9041\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 4s 228us/step - loss: 12012.8236\n",
      "Epoch 417/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 210us/step - loss: 12288.1887\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 12457.8912\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12194.0789\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 4s 240us/step - loss: 12048.0875\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 6s 377us/step - loss: 12142.7031\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 12095.7327\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12012.4852\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12160.1935\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12201.2460\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11854.3898\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11891.9417\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12409.1902\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11862.4971\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11789.1867\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 12048.5413\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11759.6452\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12039.4441\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12012.2663\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12088.9390\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11874.1412\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12076.2703\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 4s 216us/step - loss: 11974.6898\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11816.0542\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 11833.7091\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 11854.7189\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11759.3781\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11919.2272\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 12263.6193\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11519.4583\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11802.3173\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12006.7735\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11931.5647\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11777.9929\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11695.9265\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11966.9438\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11868.7117\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 11710.9027\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12066.5528\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 12142.7909\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11492.6191\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11804.8710\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11418.1020\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12257.2932\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11404.4003\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11912.4239\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11765.8068\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11582.4686\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11716.0969\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11851.9382\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 11947.8247\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11576.1560\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11539.2131\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11849.6127\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11944.6944\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11544.6890\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11301.1437\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11488.8292\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11841.4262\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11682.7550\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 11430.2977\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11498.3722\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11581.3035\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 11490.5661\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 11619.9683\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11678.1899\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11466.9728\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11451.6712\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11677.5963\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11509.1208\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11498.1589\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11630.4573\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11788.4633\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11604.1519\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11402.6013\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11469.0693\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11444.3225\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11483.9697\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11381.4471\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11684.6362\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11323.3556\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11356.6496\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11442.2420\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11540.3446\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11180.3463\n",
      "1844/1844 [==============================] - 0s 49us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 13719.4990\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 13327.3767\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 13187.9886\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 12948.1622\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12884.7410\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12882.4508\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12697.3330\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 13053.3668\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 12879.1940\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 13553.4779\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12659.1870\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12299.2739\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12956.5502\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12723.1418\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12789.7243\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12624.5375\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 12690.9073\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 12367.1888\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 12576.4988\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12639.6363\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12374.6924\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12213.5816\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 12408.1908\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 12412.8654\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 12344.7503\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 12170.7438\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12025.7999\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 12377.5678\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 12702.5670\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12616.7667\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12045.5536\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12214.5556\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12079.1235\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12256.6181\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12306.7670\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 12169.5400\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12143.7278\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11837.7584\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12195.9945\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 12117.2143\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 12120.0194\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12241.8668\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 12219.7225\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11956.7940\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11975.4299\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 12216.4704\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 12003.3826\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 11796.7554\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11725.3983\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11971.6709\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 12142.1265\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 12083.6150\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11685.4138\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11859.8716\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 11990.1274\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11721.1390\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11562.9819\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11819.1955\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 12150.9542\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11527.4265\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11938.4525\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11955.8973\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11676.7708\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11608.2893\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11840.5970\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11774.4714\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11699.8432\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11728.4380\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11442.5634\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11545.2691\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11909.0845\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11789.7986\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11504.9999\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11844.7828\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11251.0447\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11470.8681\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11671.2739\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11587.9161\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11376.8298\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11619.8801\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11435.5213\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11320.8297\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11488.6052\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11241.9572\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11608.0476\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11480.7503\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11299.5886\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11997.1149\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11563.4114\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11413.3313\n",
      "Epoch 91/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 190us/step - loss: 11258.3413\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11347.6624\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11336.8642\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11571.3552\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 11425.3317\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11190.5765\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 4s 215us/step - loss: 11424.5380\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 11103.2641\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10980.3917\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11326.4517\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11159.8984\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11236.1057\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11265.8771\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11127.3030\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11015.3141\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11275.6605\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11249.7655\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 11372.3953\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11110.6617\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11138.8437\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11475.0960\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11299.3688\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11117.5515\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11048.6243\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11307.5003\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11084.7141\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11044.8405\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10975.8330\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11149.4743\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 11003.3472\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 10981.8598\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11258.7728\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10986.4135\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 11163.2514\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11418.0133\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11176.6659\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11343.3418\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 10713.1330\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10644.5110\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11109.7457\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10916.0501\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10946.9259\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 10680.5751\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10859.9250\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11015.8498\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10990.1766\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10973.0139\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10997.0684\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10786.1849\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10648.8981\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10919.8477\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10885.2013\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 11016.2870\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10713.7542\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10745.6424\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 11085.3956\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 11065.2883\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10811.0091\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10951.9600\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10730.8591\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10867.8295\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10528.2296\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10708.6106\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 11058.4694\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10558.9694\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10599.6599\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10606.7128\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 11133.9810\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10469.3040\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10748.9466\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 10433.3252\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 10588.8014\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10676.7492\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10294.2265\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10966.7917\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10516.5854\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10623.1720\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10867.7527\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10880.5564\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10527.3691\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 11386.6127\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10771.2108\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10412.6829\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10756.1999\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10525.7766\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10884.2193\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10375.2071\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 10339.7271\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10629.6153\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10277.2618\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10600.8883\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 10486.5523\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10777.5036\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10312.8819\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10309.5686\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 10453.7816\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 10897.9244\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10398.0195\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10481.9263\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10411.2030\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10326.5122\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10326.5640\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 10198.3871\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10868.9929\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10114.8611\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10153.0590\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10310.6091\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10412.6359\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10558.0879\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 10043.8372\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10257.6614\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 11046.2389\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 10187.1740\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10142.6150\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10152.3279\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 10205.5632\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 10406.6553\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10071.8449\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10356.5752\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10359.9846\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 10290.4740\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 10145.7905\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10490.6622\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9904.1077\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10413.4903\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 10283.7679\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9999.9189\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 10214.1155\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10138.0190\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9874.3498\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 10070.8748\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 10200.8626\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 10287.6462\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10022.8928\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10293.3428\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 10381.4565\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10192.4697\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9956.1624\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 10180.5292\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 4s 250us/step - loss: 10044.9053\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 10081.1157\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 10277.8985\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 9833.3084\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 9953.0708\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10013.9907\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 9881.4852\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 10167.5179\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9986.1736\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10254.2599\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9732.7291\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 10091.9894\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 9926.7169\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 10064.6299\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 9901.8501\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 4s 269us/step - loss: 10062.0870\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 10052.3153\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9783.0492\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 10139.4312\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 10071.0362\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 9780.8992\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 9919.6064\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 10127.0400\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9818.3300\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 10074.0334\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 9646.3956\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9886.9429\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9728.6208\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10021.8457\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9774.2331\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 4s 242us/step - loss: 9610.4077\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 9772.2698\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9879.0425\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10129.1462\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9784.7159\n",
      "Epoch 265/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 186us/step - loss: 9737.5973\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9614.0441\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9647.8086\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9701.1555\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 9820.5253\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9844.8097\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9254.0492\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9904.1835\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9623.2844\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9729.4347\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9791.0632\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9598.6674\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 9569.0030\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9572.7136\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9737.9884\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9684.1089\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9714.5002\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9967.7339\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 9668.2648\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9566.1851\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9525.8579\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 10003.1902\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9698.9321\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9654.9457\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 5s 302us/step - loss: 9573.8516\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 9510.9277\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 9595.7740\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 4s 223us/step - loss: 9489.8600\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9551.8315\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 4s 262us/step - loss: 9678.7976\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 207us/step - loss: 9503.1143\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9278.8050\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 5s 273us/step - loss: 9642.2526\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 9760.2622\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 4s 229us/step - loss: 9517.7296\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 9476.1513\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 9554.0390\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 10000.2134\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9247.1962\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9366.7874\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9826.1967\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9233.8294\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9332.0966\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9805.0682\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9578.6521\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9782.6950\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9264.8246\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9522.7753\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9345.4664\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9169.6133\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9566.0168\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9296.5849\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9077.2579\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9419.5259\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9695.9229\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9140.9185\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9294.2964\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9306.3435\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9313.2267\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 9316.0205\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9405.0177\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9201.5814\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9248.2118\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 9630.1448\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 9208.4585\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 9100.8273\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9274.5326\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9330.9917\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9474.3025\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 9221.4856\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 208us/step - loss: 9381.4293\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 9324.6460\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 5s 276us/step - loss: 9532.4667\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 9090.0404\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 207us/step - loss: 9246.3436\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 4s 268us/step - loss: 9396.2441\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 9244.8226\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 5s 276us/step - loss: 9184.0329\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 9291.3137\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9034.8773\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9074.1359\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9169.9233\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9023.1336\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9434.5283\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8905.2908\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9075.1833\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9302.1142\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9337.4763\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9327.9779\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 9184.0061\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8939.7079\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9437.2862\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9328.0669\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9016.2553\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9198.5794\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9013.3083\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8938.2617\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9322.1016\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9085.3394\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9236.7898\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8996.5071\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9209.8151\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9175.1280\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 4s 263us/step - loss: 9040.5200\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 8824.6853\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 8820.6223\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9126.2679\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9501.9599\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8875.7249\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8914.5750\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8994.1529\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9118.0656\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 4s 260us/step - loss: 8885.0562\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 4s 238us/step - loss: 8901.5003\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9105.4443\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8798.1240\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8816.1985\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9406.2614\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8852.6853\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 8915.3750\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8993.8708\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9149.1901\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 4s 261us/step - loss: 8853.7450\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 8685.2279\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9053.9715\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - ETA: 0s - loss: 8702.72 - 3s 190us/step - loss: 8711.0262\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8968.4796\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 8730.2886\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 5s 282us/step - loss: 9171.1681\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9246.6869\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8751.3513\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8799.8556\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 9082.9607\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8836.4314\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8980.1629\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8692.9204\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9189.4326\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8979.6197\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8887.2030\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8618.4795\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8877.8315\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8608.3684\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8803.0871\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8961.7363\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8884.8627\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9116.9701\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 8806.4589\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8728.8118\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8823.3384\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9150.3585\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8799.8289\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8904.8109\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8585.3404\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8881.8329\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8930.7388\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8552.2878\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8870.5308\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 4s 238us/step - loss: 8648.8402\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 8751.8759\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8576.0670\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8627.1489\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8846.4594\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8695.2564\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8677.9681\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8599.6645\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8591.0549\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8778.0686\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8459.1796\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8696.5798\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8645.4958\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8640.5222\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8645.1866\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8813.8313\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 4s 225us/step - loss: 8289.4748\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 8711.6744\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8743.6301\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8726.8454\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 8849.9644\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 5s 298us/step - loss: 8540.1629\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8651.9563\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8484.1670\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 8436.8875\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8411.9369\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8554.7508\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8630.7030\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 176us/step - loss: 8373.9922\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8664.8086\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8708.4475\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8381.4140\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8698.3089\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8523.8946\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8660.9989\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 5s 276us/step - loss: 8494.9827\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 8686.3981\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 8503.5224\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 8892.6870\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8398.0404\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 8708.6997\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8566.5320\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 8520.9485\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 8543.3860\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8480.7740\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8569.4839\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8203.9674\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8440.6627\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8425.7009\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 8438.3139\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8391.4233\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8403.0809\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8581.0155\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 5s 284us/step - loss: 8558.5337\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 8291.2650\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8302.3507\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8543.2785\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8362.9256\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8436.8971\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 5s 299us/step - loss: 8203.9777\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 4s 215us/step - loss: 8642.9771\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 8237.6022\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 8403.1792\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 8505.0795\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8308.9041\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8601.5831\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8464.4510\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8125.4808\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8407.4570\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8396.4063\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 8477.2704\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8075.2123\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8421.9956\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8169.6662\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8380.4439\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 8357.9753\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8296.1389\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8358.4277\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 5s 272us/step - loss: 8058.6918\n",
      "1844/1844 [==============================] - 0s 55us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 10307.9796\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 5s 285us/step - loss: 10671.2849\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 10346.8001\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9364.6607\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9735.9120\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9726.7289\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9717.6087\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9565.1027\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9449.9757\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9734.1604\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9708.7452\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9249.4660\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9632.4683\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9449.9198\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9370.6479\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9358.9595\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9423.5179\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9289.8697\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9527.3718\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9237.8970\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9430.3945\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 9330.6097\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9164.2164\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9528.8613\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9267.5096\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9271.5730\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9680.1585\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9268.0479\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9246.9611\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9257.5665\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 9065.4244\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9098.0687\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 9084.8054\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 9429.2001\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 4s 259us/step - loss: 9209.7037\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 9068.2013\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 9439.8817\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 4s 245us/step - loss: 9098.8547\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8987.7609\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8942.0969\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9091.0410\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 9285.6467\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8897.4516\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 9220.3405\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8990.9815\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 9048.7063\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8680.3500\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 9453.7054\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8933.8051\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9276.3898\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 9124.3944\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8699.2922\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8976.4408\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8784.4002\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8588.1116\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9133.4563\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 4s 221us/step - loss: 8990.5198\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 9032.2657\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8796.5529\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 9376.9006\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8792.1523\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 9013.7161\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8978.5855\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8960.4486\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8642.0684\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8763.0199\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8809.5428\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8985.3903\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8829.8419\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8792.7712\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 9000.7286\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8711.5595\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8770.5758\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8584.1187\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8622.3067\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8850.3344\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8927.0158\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8449.1509\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8585.5015\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8858.0942\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8759.6007\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8731.8955\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8959.3912\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8710.2440\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8788.0785\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 8999.3464\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 8377.6802\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8734.7624\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8979.6155\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8605.2081\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 8532.3788\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 4s 216us/step - loss: 8745.1755\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 8500.5240\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 9049.5717\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8250.5312\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8626.3774\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8518.3121\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8713.9127\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8799.6784\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8475.2495\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8779.2593\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8346.0208\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8715.8822\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8649.0356\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8475.1134\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8207.9157\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8810.8716\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 4s 249us/step - loss: 8693.3801\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 8506.7024\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8433.2293\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8405.9292\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8816.9336\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8530.7646\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8372.8442\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8384.4336\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8603.9934\n",
      "Epoch 117/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 186us/step - loss: 8778.9868\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8136.1263\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8612.6672\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8347.7768\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8088.8689\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8729.2254\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 8313.3380\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8280.6925\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8380.6162\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8515.9404\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8625.0280\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8384.0259\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 8352.3145\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 176us/step - loss: 8157.8510\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8276.2884\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 8643.4698\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8359.8298\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8548.0157\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8178.1502\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8213.1588\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8304.6381\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8261.6738\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8414.2407\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 8290.4665\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8460.1172\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 8059.7345\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8283.3643\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8479.4238\n",
      "Epoch 145/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8480.1676\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8330.4583\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 4s 242us/step - loss: 8337.6142\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 8215.7896\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7985.2378\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8299.1698\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7959.5663\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8232.1020\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8189.9833\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8134.0378\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8168.6403\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 4s 232us/step - loss: 8484.0021\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 7982.0154\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8359.2488\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7984.4072\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8019.0256\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8370.2746\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 8239.7308\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8070.5549\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8175.4321\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7987.9245\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8294.3633\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8359.0544\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8176.0393\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8320.9747\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8159.6589\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7987.9571\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7982.4225\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8008.4083\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8052.6704\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 8251.4770\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 5s 322us/step - loss: 7791.4845\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 8002.7401\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7992.0986\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8225.9364\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8065.4528\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 8362.4122\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 4s 229us/step - loss: 7989.5828\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7871.3508\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 8211.0524\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 8076.7123\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7895.9164\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8186.1936\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7944.7450\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7994.2444\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 8096.2044\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7885.3168\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8167.5236\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8292.8621\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 7886.9130\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 8504.1859\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7941.3998\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7734.2611\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7819.1292\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 8042.0195\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7854.0655\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7837.0774\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8037.2686\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7885.5410\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7785.7110\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7735.8828\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 7948.2174\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7801.1251\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7907.6290\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7915.2072\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8120.4052\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7646.0658\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8078.2100\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7671.2123\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7753.9915\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 8134.5887\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7907.3034\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7669.8990\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 7668.1014\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7880.4547\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7967.8133\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7656.5020\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7710.3788\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7882.7573\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7660.3820\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 7744.8723\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7869.5488\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7820.3328\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7803.4272\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7695.4020\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7823.7355\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7562.9805\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7802.8685\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7749.4895\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7804.8195\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7574.4088\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7568.9030\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7661.9889\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7838.9613\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7726.0144\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7778.5191\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 8015.0588\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7882.1873\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7703.6747\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7830.3635\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7572.6476\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7523.9355\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7805.3491\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7694.4768\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7766.8154\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 4s 247us/step - loss: 7638.0362\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 7566.9148\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 7452.0846\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7678.9849\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7649.8449\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7700.6725\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7668.1018\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7549.2133\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7934.3813\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7609.6073\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7711.7926\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7526.5984\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7599.1541\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7675.4902\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7605.5310\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7655.0889\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7584.6746\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7497.6618\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7442.9913\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7400.3113\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7704.1941\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7369.5619\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 7612.5196\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 4s 248us/step - loss: 7491.6980\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 7619.2891\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7606.4164\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 7473.9477\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7478.5775\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7338.0050\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7667.6783\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7574.4325\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7296.5493\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7482.9062\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 7526.1873\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7418.2100\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7584.3236\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7345.8762\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7949.5284\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7711.3743\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7480.8754\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7425.4325\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7520.5437\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 5s 272us/step - loss: 7456.3588\n",
      "Epoch 293/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 202us/step - loss: 7781.2177\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 7108.9905\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7496.4890\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7638.4409\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7296.3696\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 5s 294us/step - loss: 7351.0383\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 4s 221us/step - loss: 7529.3306\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 4s 251us/step - loss: 7245.9258\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 4s 249us/step - loss: 7474.4859\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 7419.4455\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7478.3027\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7259.6980\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7371.8024\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7342.6979\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7500.3515\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7266.8834\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7379.6271\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7202.9071\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7531.4897\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7359.0357\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 5s 296us/step - loss: 7292.4661\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 7296.7039\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7348.6956\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7318.6094\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7168.2268\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7474.9244\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7226.6224\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7437.1658\n",
      "Epoch 321/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7739.4462\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7439.8779\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7349.8597\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7339.2675\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 4s 221us/step - loss: 7427.6752\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 7208.0312\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 7199.1496\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7369.7692\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7548.4652\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7415.0636\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 4s 249us/step - loss: 7174.4771\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 4s 230us/step - loss: 7224.6048\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7277.5803\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7228.0146\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7299.4026\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7530.0692\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 7365.7141\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 4s 242us/step - loss: 7077.6433\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 7224.6927\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7267.0982\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7234.3086\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 7211.7116\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7047.3671\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7586.3937\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7164.3810\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7177.9452\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7075.0251\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7369.6022\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7717.1922\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 5s 318us/step - loss: 7231.4290\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 7109.9057\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7234.0872\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7001.3540\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7354.0241\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7097.2878\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7119.4118\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7284.3720\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7065.3189\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7114.0921\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7114.8930\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7396.3496\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7271.5930\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 7294.0324\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7110.7475\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6947.1861\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7089.3296\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7175.2628\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7216.2396\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7235.4118\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7186.9817\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6983.4269\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7365.5485\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 4s 238us/step - loss: 7170.5450\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 7038.7247\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7226.9363\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7122.6785\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7378.3388\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7060.7370\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6855.2941\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7239.9299\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7010.8780\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7168.5373\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7001.0984\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7063.2019\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7165.3253\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7963.6922\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7159.1283\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7612.7645\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6733.2826\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7196.3953\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6835.0513\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6980.2101\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7216.4222\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6888.7108\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 7192.9594\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 5s 283us/step - loss: 6992.0043\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 6926.5613\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 5s 296us/step - loss: 7064.1618\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 6962.8702\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7120.4322\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6903.9849\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6910.8429\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6807.7339\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6906.1583\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7320.0972\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 7015.3244\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 6960.2018\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 6966.9255\n",
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6967.9956\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6979.3269\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 7192.3399\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6903.2967\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6767.1309\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7243.4002\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6795.7055\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6958.0727\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6796.3891\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6790.8043\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6784.4964\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7060.3645\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7058.4474\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6986.2950\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7121.8725\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7001.7578\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6777.2356\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6971.2281\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6939.1888\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7025.7242\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6831.3521\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6937.4640\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6801.2642\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7215.4104\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6664.3481\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6626.1238\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 6767.0754\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7155.8761\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6776.5794\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6823.5737\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 6840.2458\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 4s 231us/step - loss: 6825.1439\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6833.2819\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 5s 273us/step - loss: 6904.9269\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 6833.5504\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7072.9787\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7084.7763\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6733.3319\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6748.2075\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 5s 282us/step - loss: 6698.1133\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 6748.1019\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6764.6886\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6815.6753\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6939.1841\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7027.7469\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6905.0702\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6792.5201\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6855.6264\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 4s 216us/step - loss: 6699.8113\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 5s 272us/step - loss: 6766.0292\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6567.3004\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6867.8770\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6687.1504\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6645.1289\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6788.2514\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6779.4336\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6583.5231\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6908.2012\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6895.0240\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7201.9917\n",
      "Epoch 469/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 191us/step - loss: 6837.0761\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 4s 270us/step - loss: 6829.1112\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 6776.0237\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6741.8079\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 5s 317us/step - loss: 6652.2756\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 4s 212us/step - loss: 6739.0547\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6727.9469\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6420.5069\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6639.0944\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 6785.7050\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 4s 240us/step - loss: 6782.9226\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6780.2380\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6639.2925\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6656.4581\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6594.8234\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 4s 211us/step - loss: 7008.9706\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 4s 229us/step - loss: 6610.9495\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6458.8498\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6531.1269\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6937.0961\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6704.5327\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6582.7980\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6743.7204\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6568.8544\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 6558.3237\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 4s 258us/step - loss: 6773.1110\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 6631.6609\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 6799.1019\n",
      "Epoch 497/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 6785.3399\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6620.9125\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6526.8509\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6648.2737\n",
      "1844/1844 [==============================] - 0s 43us/step\n",
      "Epoch 1/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 8632.7915\n",
      "Epoch 2/500\n",
      "16593/16593 [==============================] - 4s 263us/step - loss: 8292.1775\n",
      "Epoch 3/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 7955.0772\n",
      "Epoch 4/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 7986.5103\n",
      "Epoch 5/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 7712.8743\n",
      "Epoch 6/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 7768.7231\n",
      "Epoch 7/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7984.0864\n",
      "Epoch 8/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7432.1061\n",
      "Epoch 9/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7537.8854\n",
      "Epoch 10/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 8035.7764\n",
      "Epoch 11/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7330.5070\n",
      "Epoch 12/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7755.4939\n",
      "Epoch 13/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7524.8140\n",
      "Epoch 14/500\n",
      "16593/16593 [==============================] - 5s 279us/step - loss: 7532.3997\n",
      "Epoch 15/500\n",
      "16593/16593 [==============================] - 4s 236us/step - loss: 7349.1100\n",
      "Epoch 16/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7761.7844\n",
      "Epoch 17/500\n",
      "16593/16593 [==============================] - 4s 260us/step - loss: 7717.6421\n",
      "Epoch 18/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 7465.5839\n",
      "Epoch 19/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7550.6578\n",
      "Epoch 20/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 7252.6326\n",
      "Epoch 21/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7551.5897\n",
      "Epoch 22/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7500.9114\n",
      "Epoch 23/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7533.9585\n",
      "Epoch 24/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7439.5971\n",
      "Epoch 25/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7351.9587\n",
      "Epoch 26/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7281.8327\n",
      "Epoch 27/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7316.7064\n",
      "Epoch 28/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7250.2067\n",
      "Epoch 29/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 8047.1430\n",
      "Epoch 30/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7235.2663\n",
      "Epoch 31/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7656.5507\n",
      "Epoch 32/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7389.9077\n",
      "Epoch 33/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7429.8341\n",
      "Epoch 34/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6931.5432\n",
      "Epoch 35/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7299.7672\n",
      "Epoch 36/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7634.8283\n",
      "Epoch 37/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7164.6280\n",
      "Epoch 38/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7546.8041\n",
      "Epoch 39/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 7227.1493\n",
      "Epoch 40/500\n",
      "16593/16593 [==============================] - 4s 267us/step - loss: 7626.8191\n",
      "Epoch 41/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 7304.2106\n",
      "Epoch 42/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7140.6074\n",
      "Epoch 43/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7240.8523\n",
      "Epoch 44/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 7098.4102\n",
      "Epoch 45/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7236.2844\n",
      "Epoch 46/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7206.2164\n",
      "Epoch 47/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 7130.6812\n",
      "Epoch 48/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7243.1306\n",
      "Epoch 49/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7528.5517\n",
      "Epoch 50/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7025.7031\n",
      "Epoch 51/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 7247.1952\n",
      "Epoch 52/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 7071.0777\n",
      "Epoch 53/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7289.6383\n",
      "Epoch 54/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 7114.7705\n",
      "Epoch 55/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7700.1008\n",
      "Epoch 56/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6901.4123\n",
      "Epoch 57/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7362.0503\n",
      "Epoch 58/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7080.4729\n",
      "Epoch 59/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 6904.1127\n",
      "Epoch 60/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7086.3853\n",
      "Epoch 61/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 7245.2313\n",
      "Epoch 62/500\n",
      "16593/16593 [==============================] - 5s 287us/step - loss: 7190.8559\n",
      "Epoch 63/500\n",
      "16593/16593 [==============================] - 4s 245us/step - loss: 7281.3238\n",
      "Epoch 64/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 6836.7271\n",
      "Epoch 65/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 7059.8733\n",
      "Epoch 66/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7285.4611\n",
      "Epoch 67/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6933.6747\n",
      "Epoch 68/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 7196.2717\n",
      "Epoch 69/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6967.4898\n",
      "Epoch 70/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 7096.9002\n",
      "Epoch 71/500\n",
      "16593/16593 [==============================] - 5s 278us/step - loss: 6961.6049\n",
      "Epoch 72/500\n",
      "16593/16593 [==============================] - 4s 236us/step - loss: 7149.5985\n",
      "Epoch 73/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7489.3359\n",
      "Epoch 74/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 7088.5703\n",
      "Epoch 75/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6980.2213\n",
      "Epoch 76/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6906.2229\n",
      "Epoch 77/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6916.0607\n",
      "Epoch 78/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6993.8026\n",
      "Epoch 79/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6953.3777\n",
      "Epoch 80/500\n",
      "16593/16593 [==============================] - 5s 300us/step - loss: 6915.3022\n",
      "Epoch 81/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 6959.0522\n",
      "Epoch 82/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7073.9603\n",
      "Epoch 83/500\n",
      "16593/16593 [==============================] - 5s 299us/step - loss: 6900.5205\n",
      "Epoch 84/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6903.0349\n",
      "Epoch 85/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 7080.3513\n",
      "Epoch 86/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6930.8608\n",
      "Epoch 87/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 7187.6620\n",
      "Epoch 88/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6842.6290\n",
      "Epoch 89/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6973.4036\n",
      "Epoch 90/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6761.3691\n",
      "Epoch 91/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6688.0423\n",
      "Epoch 92/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6695.4786\n",
      "Epoch 93/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7093.0422\n",
      "Epoch 94/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6715.8651\n",
      "Epoch 95/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 7001.8868\n",
      "Epoch 96/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6854.1499\n",
      "Epoch 97/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6755.6280\n",
      "Epoch 98/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6855.8941\n",
      "Epoch 99/500\n",
      "16593/16593 [==============================] - 4s 260us/step - loss: 6930.2552\n",
      "Epoch 100/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 7102.2008\n",
      "Epoch 101/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 7098.3861\n",
      "Epoch 102/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6833.4091\n",
      "Epoch 103/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6773.1850\n",
      "Epoch 104/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6905.8055\n",
      "Epoch 105/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6748.2644\n",
      "Epoch 106/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 7043.6431\n",
      "Epoch 107/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6682.7215\n",
      "Epoch 108/500\n",
      "16593/16593 [==============================] - 5s 300us/step - loss: 6556.3620\n",
      "Epoch 109/500\n",
      "16593/16593 [==============================] - 3s 209us/step - loss: 8531.3838\n",
      "Epoch 110/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6721.3841\n",
      "Epoch 111/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6516.4765\n",
      "Epoch 112/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6704.8651\n",
      "Epoch 113/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6936.5728\n",
      "Epoch 114/500\n",
      "16593/16593 [==============================] - 4s 243us/step - loss: 7013.0269\n",
      "Epoch 115/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 6910.8475\n",
      "Epoch 116/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 6709.6604\n",
      "Epoch 117/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6649.8520\n",
      "Epoch 118/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6908.5457\n",
      "Epoch 119/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 7130.1901\n",
      "Epoch 120/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6565.4601\n",
      "Epoch 121/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6715.6160\n",
      "Epoch 122/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6862.5123\n",
      "Epoch 123/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6817.9026\n",
      "Epoch 124/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6685.3446\n",
      "Epoch 125/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6615.1434\n",
      "Epoch 126/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 6580.4039\n",
      "Epoch 127/500\n",
      "16593/16593 [==============================] - 4s 219us/step - loss: 6860.6855\n",
      "Epoch 128/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6686.3370\n",
      "Epoch 129/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6867.3090\n",
      "Epoch 130/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 7034.6161\n",
      "Epoch 131/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6621.0928\n",
      "Epoch 132/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 6766.7569\n",
      "Epoch 133/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 6505.9984\n",
      "Epoch 134/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6623.9846\n",
      "Epoch 135/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6466.6826\n",
      "Epoch 136/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6739.4723\n",
      "Epoch 137/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6571.0145\n",
      "Epoch 138/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 6553.1953\n",
      "Epoch 139/500\n",
      "16593/16593 [==============================] - 4s 224us/step - loss: 6780.9054\n",
      "Epoch 140/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6577.4013\n",
      "Epoch 141/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6623.5642\n",
      "Epoch 142/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6678.6737\n",
      "Epoch 143/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6725.0059\n",
      "Epoch 144/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6627.8367\n",
      "Epoch 145/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 180us/step - loss: 6735.3515\n",
      "Epoch 146/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6865.8648\n",
      "Epoch 147/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6426.9031\n",
      "Epoch 148/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6579.3393\n",
      "Epoch 149/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6419.5239\n",
      "Epoch 150/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6722.6034\n",
      "Epoch 151/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6792.2572\n",
      "Epoch 152/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6629.7035\n",
      "Epoch 153/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6706.7092\n",
      "Epoch 154/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 6491.2973\n",
      "Epoch 155/500\n",
      "16593/16593 [==============================] - 4s 218us/step - loss: 6776.5244\n",
      "Epoch 156/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6571.8200\n",
      "Epoch 157/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6715.0782\n",
      "Epoch 158/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6785.5934\n",
      "Epoch 159/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6489.2025\n",
      "Epoch 160/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 7203.4934\n",
      "Epoch 161/500\n",
      "16593/16593 [==============================] - 5s 303us/step - loss: 6369.6896\n",
      "Epoch 162/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6517.9013\n",
      "Epoch 163/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 6620.4464\n",
      "Epoch 164/500\n",
      "16593/16593 [==============================] - 5s 273us/step - loss: 6599.1515\n",
      "Epoch 165/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6411.2789\n",
      "Epoch 166/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 6833.3372\n",
      "Epoch 167/500\n",
      "16593/16593 [==============================] - 4s 224us/step - loss: 7098.1825\n",
      "Epoch 168/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6612.3157\n",
      "Epoch 169/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6265.6540\n",
      "Epoch 170/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6548.8035\n",
      "Epoch 171/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6507.1321\n",
      "Epoch 172/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6584.1892\n",
      "Epoch 173/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6693.6407\n",
      "Epoch 174/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6480.6271\n",
      "Epoch 175/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6531.9942\n",
      "Epoch 176/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6610.7295\n",
      "Epoch 177/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6238.5391\n",
      "Epoch 178/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 6558.3413\n",
      "Epoch 179/500\n",
      "16593/16593 [==============================] - 5s 272us/step - loss: 6631.9953\n",
      "Epoch 180/500\n",
      "16593/16593 [==============================] - 3s 210us/step - loss: 6404.7500\n",
      "Epoch 181/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6651.9289\n",
      "Epoch 182/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6613.8429\n",
      "Epoch 183/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6574.5498\n",
      "Epoch 184/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6263.1968\n",
      "Epoch 185/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6568.5942\n",
      "Epoch 186/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 6340.5152\n",
      "Epoch 187/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6632.5506\n",
      "Epoch 188/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6558.9814\n",
      "Epoch 189/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6349.4918\n",
      "Epoch 190/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6474.1310\n",
      "Epoch 191/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6731.4341\n",
      "Epoch 192/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6409.3606\n",
      "Epoch 193/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6236.8316\n",
      "Epoch 194/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6473.0038\n",
      "Epoch 195/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6530.7687\n",
      "Epoch 196/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6468.5432\n",
      "Epoch 197/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6480.5351\n",
      "Epoch 198/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6523.8201\n",
      "Epoch 199/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6340.8685\n",
      "Epoch 200/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6451.6756\n",
      "Epoch 201/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6435.2439\n",
      "Epoch 202/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6530.0857\n",
      "Epoch 203/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6347.6556\n",
      "Epoch 204/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6466.1091\n",
      "Epoch 205/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6388.0827\n",
      "Epoch 206/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6271.6007\n",
      "Epoch 207/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6421.1591\n",
      "Epoch 208/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6444.0162\n",
      "Epoch 209/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6256.1946\n",
      "Epoch 210/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6323.4915\n",
      "Epoch 211/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6521.5387\n",
      "Epoch 212/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6421.0401\n",
      "Epoch 213/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 6592.6413\n",
      "Epoch 214/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6339.9518\n",
      "Epoch 215/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6018.4572\n",
      "Epoch 216/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6403.3958\n",
      "Epoch 217/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6406.7362\n",
      "Epoch 218/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6688.2166\n",
      "Epoch 219/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6235.9033\n",
      "Epoch 220/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6235.8932\n",
      "Epoch 221/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6210.6271\n",
      "Epoch 222/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6241.7713\n",
      "Epoch 223/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6517.8042\n",
      "Epoch 224/500\n",
      "16593/16593 [==============================] - 4s 217us/step - loss: 6722.9314\n",
      "Epoch 225/500\n",
      "16593/16593 [==============================] - 4s 256us/step - loss: 6390.5372\n",
      "Epoch 226/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 6389.6517\n",
      "Epoch 227/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6162.6193\n",
      "Epoch 228/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6070.6346\n",
      "Epoch 229/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6619.3378\n",
      "Epoch 230/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6604.3935\n",
      "Epoch 231/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6224.8903\n",
      "Epoch 232/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 6167.5416\n",
      "Epoch 233/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6222.1655\n",
      "Epoch 234/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6167.1703\n",
      "Epoch 235/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6151.3903\n",
      "Epoch 236/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6301.5846\n",
      "Epoch 237/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6626.6599\n",
      "Epoch 238/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6037.0790\n",
      "Epoch 239/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6218.9100\n",
      "Epoch 240/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6240.9743\n",
      "Epoch 241/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6158.2465\n",
      "Epoch 242/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6071.4062\n",
      "Epoch 243/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 6416.4857\n",
      "Epoch 244/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6382.7985\n",
      "Epoch 245/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 6300.4252\n",
      "Epoch 246/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6134.6167\n",
      "Epoch 247/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6275.3153\n",
      "Epoch 248/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6335.8447\n",
      "Epoch 249/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6260.2840\n",
      "Epoch 250/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6527.4935\n",
      "Epoch 251/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6229.4314\n",
      "Epoch 252/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6370.0560\n",
      "Epoch 253/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6404.9890\n",
      "Epoch 254/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6103.4990\n",
      "Epoch 255/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6029.9636\n",
      "Epoch 256/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6353.3229\n",
      "Epoch 257/500\n",
      "16593/16593 [==============================] - 4s 267us/step - loss: 6003.9767\n",
      "Epoch 258/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 6248.8258\n",
      "Epoch 259/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6348.7101\n",
      "Epoch 260/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 6156.8405\n",
      "Epoch 261/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6125.2565\n",
      "Epoch 262/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6059.1996\n",
      "Epoch 263/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 6427.6015\n",
      "Epoch 264/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5964.0671\n",
      "Epoch 265/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5955.2562\n",
      "Epoch 266/500\n",
      "16593/16593 [==============================] - 4s 232us/step - loss: 5897.6128\n",
      "Epoch 267/500\n",
      "16593/16593 [==============================] - 3s 200us/step - loss: 6270.9716\n",
      "Epoch 268/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6142.6498\n",
      "Epoch 269/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6142.5672\n",
      "Epoch 270/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6104.7974\n",
      "Epoch 271/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6384.1128\n",
      "Epoch 272/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6298.0993\n",
      "Epoch 273/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5832.8191\n",
      "Epoch 274/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6342.8012\n",
      "Epoch 275/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 6084.8917\n",
      "Epoch 276/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6159.0038\n",
      "Epoch 277/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6290.8863\n",
      "Epoch 278/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6110.7367\n",
      "Epoch 279/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5964.9938\n",
      "Epoch 280/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6282.4218\n",
      "Epoch 281/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6300.8230\n",
      "Epoch 282/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5994.2944\n",
      "Epoch 283/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5880.8909\n",
      "Epoch 284/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6187.0854\n",
      "Epoch 285/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 5848.9334\n",
      "Epoch 286/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6190.6514\n",
      "Epoch 287/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6081.3594\n",
      "Epoch 288/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 6324.2113\n",
      "Epoch 289/500\n",
      "16593/16593 [==============================] - 5s 273us/step - loss: 6282.8101\n",
      "Epoch 290/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 5836.3781\n",
      "Epoch 291/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5933.7086\n",
      "Epoch 292/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6326.8568\n",
      "Epoch 293/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 6161.3183\n",
      "Epoch 294/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5789.6961\n",
      "Epoch 295/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6141.2295\n",
      "Epoch 296/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6332.1922\n",
      "Epoch 297/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6154.6276\n",
      "Epoch 298/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6060.4816\n",
      "Epoch 299/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 6172.9538\n",
      "Epoch 300/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6178.1088\n",
      "Epoch 301/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5898.5682\n",
      "Epoch 302/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6125.4868\n",
      "Epoch 303/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 6311.1516\n",
      "Epoch 304/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5650.3999\n",
      "Epoch 305/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6190.9002\n",
      "Epoch 306/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6079.2460\n",
      "Epoch 307/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6030.2875\n",
      "Epoch 308/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 6208.5641\n",
      "Epoch 309/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5984.6670\n",
      "Epoch 310/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 6251.2915\n",
      "Epoch 311/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6093.8614\n",
      "Epoch 312/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6154.2471\n",
      "Epoch 313/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5921.9342\n",
      "Epoch 314/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6053.7719\n",
      "Epoch 315/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5789.8128\n",
      "Epoch 316/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5951.5968\n",
      "Epoch 317/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5938.7290\n",
      "Epoch 318/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6015.4468\n",
      "Epoch 319/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 6170.2428\n",
      "Epoch 320/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5864.5146\n",
      "Epoch 321/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 188us/step - loss: 6012.6883\n",
      "Epoch 322/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5778.6714\n",
      "Epoch 323/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 6499.5167\n",
      "Epoch 324/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5813.0904\n",
      "Epoch 325/500\n",
      "16593/16593 [==============================] - 5s 311us/step - loss: 5986.3819\n",
      "Epoch 326/500\n",
      "16593/16593 [==============================] - 3s 199us/step - loss: 5843.3309\n",
      "Epoch 327/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6314.0692\n",
      "Epoch 328/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5943.3290\n",
      "Epoch 329/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6042.1167\n",
      "Epoch 330/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5901.3964\n",
      "Epoch 331/500\n",
      "16593/16593 [==============================] - 4s 230us/step - loss: 6168.7521\n",
      "Epoch 332/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5925.7568\n",
      "Epoch 333/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6009.9843\n",
      "Epoch 334/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5991.3702\n",
      "Epoch 335/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5873.7455\n",
      "Epoch 336/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5870.6580\n",
      "Epoch 337/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 6203.2283\n",
      "Epoch 338/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5969.2425\n",
      "Epoch 339/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 6145.8638\n",
      "Epoch 340/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5898.1284\n",
      "Epoch 341/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 6004.1507\n",
      "Epoch 342/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5965.1952\n",
      "Epoch 343/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5740.5579\n",
      "Epoch 344/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5942.1859\n",
      "Epoch 345/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5758.9568\n",
      "Epoch 346/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5949.5636\n",
      "Epoch 347/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5992.4513\n",
      "Epoch 348/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 5799.5293\n",
      "Epoch 349/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5959.9703\n",
      "Epoch 350/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 5781.5145\n",
      "Epoch 351/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 6031.4914\n",
      "Epoch 352/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5927.5240\n",
      "Epoch 353/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5956.6861\n",
      "Epoch 354/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5948.5595\n",
      "Epoch 355/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5989.0926\n",
      "Epoch 356/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5849.4516\n",
      "Epoch 357/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5821.6810\n",
      "Epoch 358/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5931.6381\n",
      "Epoch 359/500\n",
      "16593/16593 [==============================] - 3s 195us/step - loss: 5912.3403\n",
      "Epoch 360/500\n",
      "16593/16593 [==============================] - 4s 263us/step - loss: 5947.4094\n",
      "Epoch 361/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 5736.9721\n",
      "Epoch 362/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 6358.6157\n",
      "Epoch 363/500\n",
      "16593/16593 [==============================] - 5s 283us/step - loss: 5895.4052\n",
      "Epoch 364/500\n",
      "16593/16593 [==============================] - 3s 202us/step - loss: 5997.2290\n",
      "Epoch 365/500\n",
      "16593/16593 [==============================] - 4s 239us/step - loss: 5652.3642\n",
      "Epoch 366/500\n",
      "16593/16593 [==============================] - 3s 206us/step - loss: 5761.8485\n",
      "Epoch 367/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5778.7220\n",
      "Epoch 368/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5945.6728\n",
      "Epoch 369/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5793.2279\n",
      "Epoch 370/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5956.0936\n",
      "Epoch 371/500\n",
      "16593/16593 [==============================] - ETA: 0s - loss: 5831.75 - 3s 183us/step - loss: 5844.3693\n",
      "Epoch 372/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5922.1486\n",
      "Epoch 373/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5785.2550\n",
      "Epoch 374/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5665.8583\n",
      "Epoch 375/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5987.6463\n",
      "Epoch 376/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5826.8389\n",
      "Epoch 377/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5739.6746\n",
      "Epoch 378/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5517.4498\n",
      "Epoch 379/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5933.9479\n",
      "Epoch 380/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5998.1741\n",
      "Epoch 381/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5843.2331\n",
      "Epoch 382/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5800.1316\n",
      "Epoch 383/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5631.2840\n",
      "Epoch 384/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 6137.2343\n",
      "Epoch 385/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5602.2454\n",
      "Epoch 386/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5934.4344\n",
      "Epoch 387/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5843.1612\n",
      "Epoch 388/500\n",
      "16593/16593 [==============================] - 4s 234us/step - loss: 5675.8005\n",
      "Epoch 389/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5905.3281\n",
      "Epoch 390/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5647.6812\n",
      "Epoch 391/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5732.4859\n",
      "Epoch 392/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5765.3251\n",
      "Epoch 393/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5980.6906\n",
      "Epoch 394/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5708.7095\n",
      "Epoch 395/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5740.3383\n",
      "Epoch 396/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5750.8319\n",
      "Epoch 397/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 6024.7157\n",
      "Epoch 398/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5811.6581\n",
      "Epoch 399/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5695.1714\n",
      "Epoch 400/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5743.6053\n",
      "Epoch 401/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5734.5560\n",
      "Epoch 402/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5964.2284\n",
      "Epoch 403/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5543.3180\n",
      "Epoch 404/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5759.9925\n",
      "Epoch 405/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5789.2429\n",
      "Epoch 406/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5607.4756\n",
      "Epoch 407/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5702.9352\n",
      "Epoch 408/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5773.2818\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 409/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 5687.2559\n",
      "Epoch 410/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5931.4264\n",
      "Epoch 411/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5636.2239\n",
      "Epoch 412/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5883.1798\n",
      "Epoch 413/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 5636.0879\n",
      "Epoch 414/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5657.8214\n",
      "Epoch 415/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5610.9920\n",
      "Epoch 416/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5817.6349\n",
      "Epoch 417/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5677.8061\n",
      "Epoch 418/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5547.2003\n",
      "Epoch 419/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5799.0637\n",
      "Epoch 420/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5601.2373\n",
      "Epoch 421/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5876.3027\n",
      "Epoch 422/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5756.1490\n",
      "Epoch 423/500\n",
      "16593/16593 [==============================] - 3s 180us/step - loss: 5734.3303\n",
      "Epoch 424/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5544.3044\n",
      "Epoch 425/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5695.6280\n",
      "Epoch 426/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5601.1436\n",
      "Epoch 427/500\n",
      "16593/16593 [==============================] - 4s 235us/step - loss: 5768.5404\n",
      "Epoch 428/500\n",
      "16593/16593 [==============================] - 3s 196us/step - loss: 5539.0037\n",
      "Epoch 429/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5519.3348\n",
      "Epoch 430/500\n",
      "16593/16593 [==============================] - 4s 258us/step - loss: 6008.2346\n",
      "Epoch 431/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 5990.8045\n",
      "Epoch 432/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5411.3401\n",
      "Epoch 433/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5418.7381\n",
      "Epoch 434/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5773.9073\n",
      "Epoch 435/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5691.3097\n",
      "Epoch 436/500\n",
      "16593/16593 [==============================] - 4s 261us/step - loss: 5830.6584\n",
      "Epoch 437/500\n",
      "16593/16593 [==============================] - 3s 203us/step - loss: 5518.0799\n",
      "Epoch 438/500\n",
      "16593/16593 [==============================] - 3s 194us/step - loss: 5727.5615\n",
      "Epoch 439/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5521.7676\n",
      "Epoch 440/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5569.2580\n",
      "Epoch 441/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5701.3806\n",
      "Epoch 442/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 5764.4223\n",
      "Epoch 443/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5699.5142\n",
      "Epoch 444/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5423.4880\n",
      "Epoch 445/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5556.6292\n",
      "Epoch 446/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5603.6866\n",
      "Epoch 447/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 5745.4320\n",
      "Epoch 448/500\n",
      "16593/16593 [==============================] - 3s 189us/step - loss: 5616.7986\n",
      "Epoch 449/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5542.5374\n",
      "Epoch 450/500\n",
      "16593/16593 [==============================] - 3s 192us/step - loss: 5813.9286\n",
      "Epoch 451/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5725.8891\n",
      "Epoch 452/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5475.1979\n",
      "Epoch 453/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5440.8910\n",
      "Epoch 454/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5569.6801\n",
      "Epoch 455/500\n",
      "16593/16593 [==============================] - 5s 275us/step - loss: 5712.4486\n",
      "Epoch 456/500\n",
      "16593/16593 [==============================] - 3s 198us/step - loss: 5677.3620\n",
      "Epoch 457/500\n",
      "16593/16593 [==============================] - 3s 193us/step - loss: 5654.8836\n",
      "Epoch 458/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5426.5438\n",
      "Epoch 459/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5557.5430\n",
      "Epoch 460/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5646.8579\n",
      "Epoch 461/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5747.4875\n",
      "Epoch 462/500\n",
      "16593/16593 [==============================] - 3s 178us/step - loss: 5495.9382\n",
      "Epoch 463/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5412.4440\n",
      "Epoch 464/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5569.8834\n",
      "Epoch 465/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5774.4822\n",
      "Epoch 466/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5655.3608\n",
      "Epoch 467/500\n",
      "16593/16593 [==============================] - 3s 191us/step - loss: 5333.2030\n",
      "Epoch 468/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5635.1560\n",
      "Epoch 469/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5288.3662\n",
      "Epoch 470/500\n",
      "16593/16593 [==============================] - 3s 188us/step - loss: 5637.5014\n",
      "Epoch 471/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5599.2218\n",
      "Epoch 472/500\n",
      "16593/16593 [==============================] - 3s 183us/step - loss: 5691.9581\n",
      "Epoch 473/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5618.2809\n",
      "Epoch 474/500\n",
      "16593/16593 [==============================] - 4s 226us/step - loss: 5551.8138\n",
      "Epoch 475/500\n",
      "16593/16593 [==============================] - 4s 250us/step - loss: 5501.7894\n",
      "Epoch 476/500\n",
      "16593/16593 [==============================] - 3s 197us/step - loss: 5726.1161\n",
      "Epoch 477/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5548.8005\n",
      "Epoch 478/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5564.8027\n",
      "Epoch 479/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5356.4011\n",
      "Epoch 480/500\n",
      "16593/16593 [==============================] - 5s 288us/step - loss: 5407.5676\n",
      "Epoch 481/500\n",
      "16593/16593 [==============================] - 3s 205us/step - loss: 5897.2891\n",
      "Epoch 482/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5288.2029\n",
      "Epoch 483/500\n",
      "16593/16593 [==============================] - 5s 277us/step - loss: 5509.0692\n",
      "Epoch 484/500\n",
      "16593/16593 [==============================] - 3s 204us/step - loss: 5415.3117\n",
      "Epoch 485/500\n",
      "16593/16593 [==============================] - 3s 190us/step - loss: 5591.4545\n",
      "Epoch 486/500\n",
      "16593/16593 [==============================] - 3s 187us/step - loss: 5382.2548\n",
      "Epoch 487/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5625.5891\n",
      "Epoch 488/500\n",
      "16593/16593 [==============================] - 3s 186us/step - loss: 5445.4517\n",
      "Epoch 489/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5587.6040\n",
      "Epoch 490/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5769.7926\n",
      "Epoch 491/500\n",
      "16593/16593 [==============================] - 3s 181us/step - loss: 5511.5129\n",
      "Epoch 492/500\n",
      "16593/16593 [==============================] - 5s 281us/step - loss: 5253.6703\n",
      "Epoch 493/500\n",
      "16593/16593 [==============================] - 3s 201us/step - loss: 5418.2223\n",
      "Epoch 494/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5483.7569\n",
      "Epoch 495/500\n",
      "16593/16593 [==============================] - 3s 184us/step - loss: 5476.0614\n",
      "Epoch 496/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5630.6853\n",
      "Epoch 497/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16593/16593 [==============================] - 3s 190us/step - loss: 5922.2022\n",
      "Epoch 498/500\n",
      "16593/16593 [==============================] - 3s 182us/step - loss: 5490.3778\n",
      "Epoch 499/500\n",
      "16593/16593 [==============================] - 3s 185us/step - loss: 5476.6797\n",
      "Epoch 500/500\n",
      "16593/16593 [==============================] - 3s 179us/step - loss: 5453.7767\n",
      "1844/1844 [==============================] - 0s 42us/step\n",
      "Epoch 1/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 6980.8058\n",
      "Epoch 2/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 7034.2601\n",
      "Epoch 3/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 6500.4063\n",
      "Epoch 4/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 6640.8847\n",
      "Epoch 5/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 6289.7168\n",
      "Epoch 6/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 6589.9339\n",
      "Epoch 7/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 6571.3951\n",
      "Epoch 8/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6198.8034\n",
      "Epoch 9/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 6265.5388\n",
      "Epoch 10/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6695.7331\n",
      "Epoch 11/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 6387.5177\n",
      "Epoch 12/500\n",
      "16594/16594 [==============================] - 4s 221us/step - loss: 6314.1861\n",
      "Epoch 13/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 6106.7914\n",
      "Epoch 14/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6111.5134\n",
      "Epoch 15/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 6326.4801\n",
      "Epoch 16/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 6188.4870\n",
      "Epoch 17/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 6176.1480\n",
      "Epoch 18/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 6161.3162\n",
      "Epoch 19/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6249.7384\n",
      "Epoch 20/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 6237.9201\n",
      "Epoch 21/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6422.3457\n",
      "Epoch 22/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 6146.3207\n",
      "Epoch 23/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6093.5769\n",
      "Epoch 24/500\n",
      "16594/16594 [==============================] - 5s 310us/step - loss: 6064.0022\n",
      "Epoch 25/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 5934.7325\n",
      "Epoch 26/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6194.3546\n",
      "Epoch 27/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 6383.8777\n",
      "Epoch 28/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5956.4103\n",
      "Epoch 29/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 6018.1675\n",
      "Epoch 30/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6254.7139\n",
      "Epoch 31/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 6023.3349\n",
      "Epoch 32/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 5965.1484\n",
      "Epoch 33/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5960.8879\n",
      "Epoch 34/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5993.8651\n",
      "Epoch 35/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 6133.0806\n",
      "Epoch 36/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 6090.2750\n",
      "Epoch 37/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 6210.0753\n",
      "Epoch 38/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5990.9375\n",
      "Epoch 39/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5773.3789\n",
      "Epoch 40/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5968.6625\n",
      "Epoch 41/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5906.0825\n",
      "Epoch 42/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 6113.3044\n",
      "Epoch 43/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6058.5042\n",
      "Epoch 44/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 6164.0978\n",
      "Epoch 45/500\n",
      "16594/16594 [==============================] - 3s 209us/step - loss: 5855.9782\n",
      "Epoch 46/500\n",
      "16594/16594 [==============================] - 4s 269us/step - loss: 5970.2905\n",
      "Epoch 47/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 6120.9751\n",
      "Epoch 48/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5713.8474\n",
      "Epoch 49/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5800.2369\n",
      "Epoch 50/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 6041.0597\n",
      "Epoch 51/500\n",
      "16594/16594 [==============================] - 4s 247us/step - loss: 5654.4159\n",
      "Epoch 52/500\n",
      "16594/16594 [==============================] - 4s 247us/step - loss: 5889.7843\n",
      "Epoch 53/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 6078.1727\n",
      "Epoch 54/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5945.1220\n",
      "Epoch 55/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5732.0718\n",
      "Epoch 56/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5986.0796\n",
      "Epoch 57/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6002.5416\n",
      "Epoch 58/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 6280.7238\n",
      "Epoch 59/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6054.8670\n",
      "Epoch 60/500\n",
      "16594/16594 [==============================] - 3s 204us/step - loss: 5921.9458\n",
      "Epoch 61/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5603.7224\n",
      "Epoch 62/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5875.0978\n",
      "Epoch 63/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5894.5550\n",
      "Epoch 64/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5927.5280\n",
      "Epoch 65/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 6113.8152\n",
      "Epoch 66/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5864.2549\n",
      "Epoch 67/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5767.4236\n",
      "Epoch 68/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5689.0828\n",
      "Epoch 69/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5659.1142\n",
      "Epoch 70/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6222.9196\n",
      "Epoch 71/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5881.7301\n",
      "Epoch 72/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5874.3665\n",
      "Epoch 73/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5871.3515\n",
      "Epoch 74/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5511.7258\n",
      "Epoch 75/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 6115.2173\n",
      "Epoch 76/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5733.2194\n",
      "Epoch 77/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 5696.8828\n",
      "Epoch 78/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 5820.9534\n",
      "Epoch 79/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 6057.4591\n",
      "Epoch 80/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5629.8760\n",
      "Epoch 81/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5937.1231\n",
      "Epoch 82/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5702.2066\n",
      "Epoch 83/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5656.1164\n",
      "Epoch 84/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5810.7729\n",
      "Epoch 85/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 5680.2243\n",
      "Epoch 86/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 5797.4277\n",
      "Epoch 87/500\n",
      "16594/16594 [==============================] - 4s 262us/step - loss: 5893.0869\n",
      "Epoch 88/500\n",
      "16594/16594 [==============================] - 3s 205us/step - loss: 5983.9242\n",
      "Epoch 89/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5465.2932\n",
      "Epoch 90/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5828.3318\n",
      "Epoch 91/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5711.4891\n",
      "Epoch 92/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 6025.5189\n",
      "Epoch 93/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5521.3892\n",
      "Epoch 94/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5632.4465\n",
      "Epoch 95/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5823.5114\n",
      "Epoch 96/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5901.8545\n",
      "Epoch 97/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5623.3548\n",
      "Epoch 98/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 5672.7953\n",
      "Epoch 99/500\n",
      "16594/16594 [==============================] - 4s 236us/step - loss: 5670.6314\n",
      "Epoch 100/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 5510.0974\n",
      "Epoch 101/500\n",
      "16594/16594 [==============================] - 5s 278us/step - loss: 5720.8017\n",
      "Epoch 102/500\n",
      "16594/16594 [==============================] - 4s 255us/step - loss: 5630.4928\n",
      "Epoch 103/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5916.3905\n",
      "Epoch 104/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5781.6981\n",
      "Epoch 105/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 6095.7875\n",
      "Epoch 106/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5759.7499\n",
      "Epoch 107/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5634.6920\n",
      "Epoch 108/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5713.2417\n",
      "Epoch 109/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5640.8817\n",
      "Epoch 110/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5536.5699\n",
      "Epoch 111/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5573.2905\n",
      "Epoch 112/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 6030.5987\n",
      "Epoch 113/500\n",
      "16594/16594 [==============================] - 4s 224us/step - loss: 5635.1398\n",
      "Epoch 114/500\n",
      "16594/16594 [==============================] - 4s 247us/step - loss: 5692.9660\n",
      "Epoch 115/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5720.8496\n",
      "Epoch 116/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5645.8763\n",
      "Epoch 117/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5466.9753\n",
      "Epoch 118/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5766.9354\n",
      "Epoch 119/500\n",
      "16594/16594 [==============================] - 4s 265us/step - loss: 5676.9315\n",
      "Epoch 120/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 5791.2364\n",
      "Epoch 121/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5463.5566\n",
      "Epoch 122/500\n",
      "16594/16594 [==============================] - 5s 291us/step - loss: 5529.5492\n",
      "Epoch 123/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 5570.9103\n",
      "Epoch 124/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5786.2529\n",
      "Epoch 125/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5567.0655\n",
      "Epoch 126/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5576.5739\n",
      "Epoch 127/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5695.8606\n",
      "Epoch 128/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5553.7997\n",
      "Epoch 129/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5601.2709\n",
      "Epoch 130/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5769.3614\n",
      "Epoch 131/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5605.1561\n",
      "Epoch 132/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5633.1627\n",
      "Epoch 133/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5434.7506\n",
      "Epoch 134/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5404.3936\n",
      "Epoch 135/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5774.7883\n",
      "Epoch 136/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5550.0690\n",
      "Epoch 137/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5962.6528\n",
      "Epoch 138/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5435.3130\n",
      "Epoch 139/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5724.6757\n",
      "Epoch 140/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5425.8980\n",
      "Epoch 141/500\n",
      "16594/16594 [==============================] - 5s 289us/step - loss: 5516.2379\n",
      "Epoch 142/500\n",
      "16594/16594 [==============================] - 3s 206us/step - loss: 5470.9723\n",
      "Epoch 143/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5551.3505\n",
      "Epoch 144/500\n",
      "16594/16594 [==============================] - 4s 252us/step - loss: 5369.1049\n",
      "Epoch 145/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 5814.7301\n",
      "Epoch 146/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5466.2196\n",
      "Epoch 147/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5478.7773\n",
      "Epoch 148/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5414.4607\n",
      "Epoch 149/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5499.7596\n",
      "Epoch 150/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5417.8871\n",
      "Epoch 151/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5403.8035\n",
      "Epoch 152/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5521.9466\n",
      "Epoch 153/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5911.9725\n",
      "Epoch 154/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5257.6768\n",
      "Epoch 155/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5323.3757\n",
      "Epoch 156/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5582.7761\n",
      "Epoch 157/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5961.0165\n",
      "Epoch 158/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5456.8985\n",
      "Epoch 159/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5387.4580\n",
      "Epoch 160/500\n",
      "16594/16594 [==============================] - 5s 292us/step - loss: 5442.0736\n",
      "Epoch 161/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 5650.8966\n",
      "Epoch 162/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5464.7893\n",
      "Epoch 163/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5495.7931\n",
      "Epoch 164/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5332.1789\n",
      "Epoch 165/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5544.8004\n",
      "Epoch 166/500\n",
      "16594/16594 [==============================] - 5s 289us/step - loss: 5871.8190\n",
      "Epoch 167/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 5412.9934\n",
      "Epoch 168/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5198.0756\n",
      "Epoch 169/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5351.9570\n",
      "Epoch 170/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5476.1711\n",
      "Epoch 171/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5281.6392\n",
      "Epoch 172/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5676.6589\n",
      "Epoch 173/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 190us/step - loss: 5380.1824\n",
      "Epoch 174/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5475.6104\n",
      "Epoch 175/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5545.6166\n",
      "Epoch 176/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5382.4252\n",
      "Epoch 177/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5375.1831\n",
      "Epoch 178/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 5343.8321\n",
      "Epoch 179/500\n",
      "16594/16594 [==============================] - 4s 267us/step - loss: 5488.3497\n",
      "Epoch 180/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 5448.5277\n",
      "Epoch 181/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5412.5848\n",
      "Epoch 182/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5381.9087\n",
      "Epoch 183/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5344.3263\n",
      "Epoch 184/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5366.3270\n",
      "Epoch 185/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5436.9130\n",
      "Epoch 186/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5421.7340\n",
      "Epoch 187/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 5549.7044\n",
      "Epoch 188/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5445.8560\n",
      "Epoch 189/500\n",
      "16594/16594 [==============================] - ETA: 0s - loss: 5283.53 - 3s 188us/step - loss: 5293.1121\n",
      "Epoch 190/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5657.0064\n",
      "Epoch 191/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5339.7390\n",
      "Epoch 192/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5387.8538\n",
      "Epoch 193/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5317.9100\n",
      "Epoch 194/500\n",
      "16594/16594 [==============================] - 3s 210us/step - loss: 5244.9988\n",
      "Epoch 195/500\n",
      "16594/16594 [==============================] - 4s 258us/step - loss: 5476.0979\n",
      "Epoch 196/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 5464.9835\n",
      "Epoch 197/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5624.6593\n",
      "Epoch 198/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5275.6063\n",
      "Epoch 199/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5543.5270\n",
      "Epoch 200/500\n",
      "16594/16594 [==============================] - 4s 241us/step - loss: 5735.5161\n",
      "Epoch 201/500\n",
      "16594/16594 [==============================] - 4s 227us/step - loss: 5283.4127\n",
      "Epoch 202/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 5245.1224\n",
      "Epoch 203/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5168.5395\n",
      "Epoch 204/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5347.2351\n",
      "Epoch 205/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5620.2786\n",
      "Epoch 206/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5420.8744\n",
      "Epoch 207/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5303.8351\n",
      "Epoch 208/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5426.4058\n",
      "Epoch 209/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 5713.6363\n",
      "Epoch 210/500\n",
      "16594/16594 [==============================] - 4s 225us/step - loss: 5256.4907\n",
      "Epoch 211/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5195.0894\n",
      "Epoch 212/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5497.6821\n",
      "Epoch 213/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5363.1659\n",
      "Epoch 214/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5168.1493\n",
      "Epoch 215/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5419.8032\n",
      "Epoch 216/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5290.8200\n",
      "Epoch 217/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5360.8036\n",
      "Epoch 218/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5186.7766\n",
      "Epoch 219/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5223.7670\n",
      "Epoch 220/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5250.2369\n",
      "Epoch 221/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5409.4487\n",
      "Epoch 222/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5225.8972\n",
      "Epoch 223/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5387.2249\n",
      "Epoch 224/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5669.3680\n",
      "Epoch 225/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5257.6787\n",
      "Epoch 226/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5284.8080\n",
      "Epoch 227/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5234.2648\n",
      "Epoch 228/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5506.7357\n",
      "Epoch 229/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5594.0115\n",
      "Epoch 230/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5107.7741\n",
      "Epoch 231/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5433.2666\n",
      "Epoch 232/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5008.3126\n",
      "Epoch 233/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5310.0580\n",
      "Epoch 234/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5244.9594\n",
      "Epoch 235/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5358.7544\n",
      "Epoch 236/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5467.0172\n",
      "Epoch 237/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5533.8050\n",
      "Epoch 238/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5385.2807\n",
      "Epoch 239/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4987.3602\n",
      "Epoch 240/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 5237.3263\n",
      "Epoch 241/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5324.6303\n",
      "Epoch 242/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5241.5837\n",
      "Epoch 243/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5150.9879\n",
      "Epoch 244/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5249.4194\n",
      "Epoch 245/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5136.4941\n",
      "Epoch 246/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5121.1888\n",
      "Epoch 247/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5489.3396\n",
      "Epoch 248/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5892.8383\n",
      "Epoch 249/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5105.5707\n",
      "Epoch 250/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5041.3907\n",
      "Epoch 251/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5422.6357\n",
      "Epoch 252/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4999.3388\n",
      "Epoch 253/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5173.9225\n",
      "Epoch 254/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5239.8932\n",
      "Epoch 255/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5233.6309\n",
      "Epoch 256/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5211.7822\n",
      "Epoch 257/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5319.1882\n",
      "Epoch 258/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5123.9739\n",
      "Epoch 259/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5228.7462\n",
      "Epoch 260/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5225.3558\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 261/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5199.5878\n",
      "Epoch 262/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5085.9994\n",
      "Epoch 263/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5077.4075\n",
      "Epoch 264/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5076.5857\n",
      "Epoch 265/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5468.6842\n",
      "Epoch 266/500\n",
      "16594/16594 [==============================] - 5s 298us/step - loss: 5154.5750\n",
      "Epoch 267/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 5223.8234\n",
      "Epoch 268/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5010.6637\n",
      "Epoch 269/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5027.1031\n",
      "Epoch 270/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5287.6560\n",
      "Epoch 271/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5059.2117\n",
      "Epoch 272/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5401.2628\n",
      "Epoch 273/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5470.0003\n",
      "Epoch 274/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5069.8583\n",
      "Epoch 275/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5071.9670\n",
      "Epoch 276/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5233.5621\n",
      "Epoch 277/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5042.5867\n",
      "Epoch 278/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4984.2560\n",
      "Epoch 279/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 5095.4279\n",
      "Epoch 280/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5131.7765\n",
      "Epoch 281/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5244.8127\n",
      "Epoch 282/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5064.5494\n",
      "Epoch 283/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5232.1070\n",
      "Epoch 284/500\n",
      "16594/16594 [==============================] - 3s 208us/step - loss: 5088.0154\n",
      "Epoch 285/500\n",
      "16594/16594 [==============================] - 4s 257us/step - loss: 5087.7554\n",
      "Epoch 286/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4946.3895\n",
      "Epoch 287/500\n",
      "16594/16594 [==============================] - 4s 258us/step - loss: 5308.1655\n",
      "Epoch 288/500\n",
      "16594/16594 [==============================] - 3s 205us/step - loss: 5137.7584\n",
      "Epoch 289/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4989.3397\n",
      "Epoch 290/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5062.7507\n",
      "Epoch 291/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5093.0464\n",
      "Epoch 292/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5175.5732\n",
      "Epoch 293/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5357.7359\n",
      "Epoch 294/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5101.1254\n",
      "Epoch 295/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4969.8524\n",
      "Epoch 296/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5192.8440\n",
      "Epoch 297/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5165.2976\n",
      "Epoch 298/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5080.7371\n",
      "Epoch 299/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5262.2839\n",
      "Epoch 300/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5072.9344\n",
      "Epoch 301/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4968.8989\n",
      "Epoch 302/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5298.7096\n",
      "Epoch 303/500\n",
      "16594/16594 [==============================] - 4s 225us/step - loss: 4998.5622\n",
      "Epoch 304/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4948.3163\n",
      "Epoch 305/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5085.2698\n",
      "Epoch 306/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5028.9561\n",
      "Epoch 307/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5096.4150\n",
      "Epoch 308/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5054.0729\n",
      "Epoch 309/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5025.3630\n",
      "Epoch 310/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5469.7750\n",
      "Epoch 311/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5017.8104\n",
      "Epoch 312/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5174.5770\n",
      "Epoch 313/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4952.2018\n",
      "Epoch 314/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4981.6500\n",
      "Epoch 315/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4971.3636\n",
      "Epoch 316/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4960.3791\n",
      "Epoch 317/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5063.5552\n",
      "Epoch 318/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5308.9554\n",
      "Epoch 319/500\n",
      "16594/16594 [==============================] - 3s 211us/step - loss: 4990.5915\n",
      "Epoch 320/500\n",
      "16594/16594 [==============================] - 4s 243us/step - loss: 4885.7578\n",
      "Epoch 321/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 5128.2993\n",
      "Epoch 322/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4919.7859\n",
      "Epoch 323/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5201.0316\n",
      "Epoch 324/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 5487.9425\n",
      "Epoch 325/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4870.3279\n",
      "Epoch 326/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5262.0904\n",
      "Epoch 327/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4836.2487\n",
      "Epoch 328/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5195.7371\n",
      "Epoch 329/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5167.5078\n",
      "Epoch 330/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4721.6197\n",
      "Epoch 331/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 5109.9215\n",
      "Epoch 332/500\n",
      "16594/16594 [==============================] - 4s 267us/step - loss: 5294.8937\n",
      "Epoch 333/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 5008.2922\n",
      "Epoch 334/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5085.2364\n",
      "Epoch 335/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4903.6061\n",
      "Epoch 336/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4950.0708\n",
      "Epoch 337/500\n",
      "16594/16594 [==============================] - 3s 209us/step - loss: 4833.4284\n",
      "Epoch 338/500\n",
      "16594/16594 [==============================] - 4s 225us/step - loss: 5057.4583\n",
      "Epoch 339/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5002.5927\n",
      "Epoch 340/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5084.7205\n",
      "Epoch 341/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4859.4218\n",
      "Epoch 342/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4975.7072\n",
      "Epoch 343/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4995.5362\n",
      "Epoch 344/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4892.2990\n",
      "Epoch 345/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4809.3371\n",
      "Epoch 346/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5148.4878\n",
      "Epoch 347/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5117.7554\n",
      "Epoch 348/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4969.1726\n",
      "Epoch 349/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 190us/step - loss: 4904.1916\n",
      "Epoch 350/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5426.3014\n",
      "Epoch 351/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4913.4238\n",
      "Epoch 352/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4862.6613\n",
      "Epoch 353/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4970.7165\n",
      "Epoch 354/500\n",
      "16594/16594 [==============================] - 4s 221us/step - loss: 5071.2644\n",
      "Epoch 355/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4773.2090\n",
      "Epoch 356/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4999.0735\n",
      "Epoch 357/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5132.6685\n",
      "Epoch 358/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5081.9531\n",
      "Epoch 359/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4806.1220\n",
      "Epoch 360/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4941.5280\n",
      "Epoch 361/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4919.9784\n",
      "Epoch 362/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4759.1706\n",
      "Epoch 363/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5068.1471\n",
      "Epoch 364/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5049.8493\n",
      "Epoch 365/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4859.1521\n",
      "Epoch 366/500\n",
      "16594/16594 [==============================] - 4s 232us/step - loss: 4766.0461\n",
      "Epoch 367/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 5012.0794\n",
      "Epoch 368/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5081.2291\n",
      "Epoch 369/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5008.3557\n",
      "Epoch 370/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4881.7831\n",
      "Epoch 371/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5010.5234\n",
      "Epoch 372/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4859.9469\n",
      "Epoch 373/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4993.1222\n",
      "Epoch 374/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4835.4008\n",
      "Epoch 375/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4904.5316\n",
      "Epoch 376/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5002.6836\n",
      "Epoch 377/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4797.6986\n",
      "Epoch 378/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4916.0339\n",
      "Epoch 379/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4712.0557\n",
      "Epoch 380/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5164.0085\n",
      "Epoch 381/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4725.2526\n",
      "Epoch 382/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4869.1192\n",
      "Epoch 383/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4786.1815\n",
      "Epoch 384/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4819.6341\n",
      "Epoch 385/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4997.0548\n",
      "Epoch 386/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4935.7895\n",
      "Epoch 387/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4881.2160\n",
      "Epoch 388/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4933.2201\n",
      "Epoch 389/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4892.7683\n",
      "Epoch 390/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5051.3281\n",
      "Epoch 391/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4744.4202\n",
      "Epoch 392/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4799.2195\n",
      "Epoch 393/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 5041.9679\n",
      "Epoch 394/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4724.6332\n",
      "Epoch 395/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4933.0833\n",
      "Epoch 396/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4651.8560\n",
      "Epoch 397/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4656.1148\n",
      "Epoch 398/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5073.4873\n",
      "Epoch 399/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4921.1706\n",
      "Epoch 400/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4899.6515\n",
      "Epoch 401/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4770.9966\n",
      "Epoch 402/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5030.2894\n",
      "Epoch 403/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4970.0662\n",
      "Epoch 404/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4674.9224\n",
      "Epoch 405/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4877.7973\n",
      "Epoch 406/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4818.3856\n",
      "Epoch 407/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4824.1836\n",
      "Epoch 408/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4728.3684\n",
      "Epoch 409/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5030.6179\n",
      "Epoch 410/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4908.5973\n",
      "Epoch 411/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4782.2963\n",
      "Epoch 412/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4769.9401\n",
      "Epoch 413/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4903.6157\n",
      "Epoch 414/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4722.0333\n",
      "Epoch 415/500\n",
      "16594/16594 [==============================] - 4s 251us/step - loss: 4871.7305\n",
      "Epoch 416/500\n",
      "16594/16594 [==============================] - 4s 261us/step - loss: 4804.2095\n",
      "Epoch 417/500\n",
      "16594/16594 [==============================] - 3s 207us/step - loss: 4738.7169\n",
      "Epoch 418/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4787.8084\n",
      "Epoch 419/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4981.0052\n",
      "Epoch 420/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4643.6329\n",
      "Epoch 421/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4891.9378\n",
      "Epoch 422/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5051.5952\n",
      "Epoch 423/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4792.2856\n",
      "Epoch 424/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4945.6522\n",
      "Epoch 425/500\n",
      "16594/16594 [==============================] - 4s 260us/step - loss: 4631.8927\n",
      "Epoch 426/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4880.8247\n",
      "Epoch 427/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4762.9590\n",
      "Epoch 428/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4793.8958\n",
      "Epoch 429/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4823.1803\n",
      "Epoch 430/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5145.2478\n",
      "Epoch 431/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4601.1423\n",
      "Epoch 432/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4581.3452\n",
      "Epoch 433/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4899.8296\n",
      "Epoch 434/500\n",
      "16594/16594 [==============================] - 5s 299us/step - loss: 4805.9534\n",
      "Epoch 435/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4657.6849\n",
      "Epoch 436/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4827.7604\n",
      "Epoch 437/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4780.6484\n",
      "Epoch 438/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4837.2595\n",
      "Epoch 439/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4848.2268\n",
      "Epoch 440/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4900.4109\n",
      "Epoch 441/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4885.3234\n",
      "Epoch 442/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4640.4846\n",
      "Epoch 443/500\n",
      "16594/16594 [==============================] - 5s 274us/step - loss: 4770.7841\n",
      "Epoch 444/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4790.0797\n",
      "Epoch 445/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4677.1098\n",
      "Epoch 446/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4713.8622\n",
      "Epoch 447/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4930.7310\n",
      "Epoch 448/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4822.5369\n",
      "Epoch 449/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5163.2648\n",
      "Epoch 450/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4569.9495\n",
      "Epoch 451/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4508.7677\n",
      "Epoch 452/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 4785.9821\n",
      "Epoch 453/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 4797.6388\n",
      "Epoch 454/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4578.7950\n",
      "Epoch 455/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4553.4708\n",
      "Epoch 456/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4765.0109\n",
      "Epoch 457/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4926.7116\n",
      "Epoch 458/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4575.8245\n",
      "Epoch 459/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4548.6369\n",
      "Epoch 460/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4951.6257\n",
      "Epoch 461/500\n",
      "16594/16594 [==============================] - 4s 247us/step - loss: 4795.0991\n",
      "Epoch 462/500\n",
      "16594/16594 [==============================] - 4s 240us/step - loss: 4774.4588\n",
      "Epoch 463/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4864.8727\n",
      "Epoch 464/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4789.3957\n",
      "Epoch 465/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4935.4958\n",
      "Epoch 466/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4761.5996\n",
      "Epoch 467/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4729.2918\n",
      "Epoch 468/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4654.5097\n",
      "Epoch 469/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 4901.5844\n",
      "Epoch 470/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4747.9159\n",
      "Epoch 471/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4742.6457\n",
      "Epoch 472/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4649.5934\n",
      "Epoch 473/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4795.3159\n",
      "Epoch 474/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4850.7725\n",
      "Epoch 475/500\n",
      "16594/16594 [==============================] - 3s 177us/step - loss: 4668.0571\n",
      "Epoch 476/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4766.4109\n",
      "Epoch 477/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4731.8384\n",
      "Epoch 478/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4558.3792\n",
      "Epoch 479/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4984.2255\n",
      "Epoch 480/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4946.1432\n",
      "Epoch 481/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4607.7021\n",
      "Epoch 482/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4536.7889\n",
      "Epoch 483/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4917.5372\n",
      "Epoch 484/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4618.2779\n",
      "Epoch 485/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4546.1107\n",
      "Epoch 486/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4402.6112\n",
      "Epoch 487/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4744.3979\n",
      "Epoch 488/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5358.0225\n",
      "Epoch 489/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4530.7753\n",
      "Epoch 490/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4622.7409\n",
      "Epoch 491/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4764.7367\n",
      "Epoch 492/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4565.4128\n",
      "Epoch 493/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4838.2341\n",
      "Epoch 494/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4711.1115\n",
      "Epoch 495/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4600.1753\n",
      "Epoch 496/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4606.5182\n",
      "Epoch 497/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4461.9693\n",
      "Epoch 498/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4792.4895\n",
      "Epoch 499/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4790.2014\n",
      "Epoch 500/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4591.7533\n",
      "1843/1843 [==============================] - 0s 40us/step\n",
      "Epoch 1/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 6155.4458\n",
      "Epoch 2/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 5725.3063\n",
      "Epoch 3/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5507.3062\n",
      "Epoch 4/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5448.1263\n",
      "Epoch 5/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5291.5861\n",
      "Epoch 6/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5609.0920\n",
      "Epoch 7/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 5232.5790\n",
      "Epoch 8/500\n",
      "16594/16594 [==============================] - 4s 263us/step - loss: 5255.8164\n",
      "Epoch 9/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 5506.6853\n",
      "Epoch 10/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5369.7924\n",
      "Epoch 11/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5168.0130\n",
      "Epoch 12/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5178.2025\n",
      "Epoch 13/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 5219.8189\n",
      "Epoch 14/500\n",
      "16594/16594 [==============================] - 4s 218us/step - loss: 5154.6057\n",
      "Epoch 15/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5075.4265\n",
      "Epoch 16/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5340.3312\n",
      "Epoch 17/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5212.1431\n",
      "Epoch 18/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5445.8636\n",
      "Epoch 19/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5158.3659\n",
      "Epoch 20/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5074.2370\n",
      "Epoch 21/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5229.3431\n",
      "Epoch 22/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5093.3422\n",
      "Epoch 23/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5095.0794\n",
      "Epoch 24/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5182.1342\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5050.8075\n",
      "Epoch 26/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5529.3443\n",
      "Epoch 27/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5438.8181\n",
      "Epoch 28/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4818.6672\n",
      "Epoch 29/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5080.9039\n",
      "Epoch 30/500\n",
      "16594/16594 [==============================] - 5s 299us/step - loss: 5180.3201\n",
      "Epoch 31/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 4854.7769\n",
      "Epoch 32/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5828.4987\n",
      "Epoch 33/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 4687.2159\n",
      "Epoch 34/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5015.2536\n",
      "Epoch 35/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5193.0079\n",
      "Epoch 36/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5139.5878\n",
      "Epoch 37/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 5105.8135\n",
      "Epoch 38/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4887.8918\n",
      "Epoch 39/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5336.2761\n",
      "Epoch 40/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4927.3719\n",
      "Epoch 41/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4952.5170\n",
      "Epoch 42/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5167.4606\n",
      "Epoch 43/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5034.2741\n",
      "Epoch 44/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5050.8317\n",
      "Epoch 45/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5114.4382\n",
      "Epoch 46/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5017.1688\n",
      "Epoch 47/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4890.2755\n",
      "Epoch 48/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4935.1373\n",
      "Epoch 49/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5155.3572\n",
      "Epoch 50/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4934.5497\n",
      "Epoch 51/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4807.4227\n",
      "Epoch 52/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 5499.2953\n",
      "Epoch 53/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4741.3778\n",
      "Epoch 54/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4828.9938\n",
      "Epoch 55/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4779.8120\n",
      "Epoch 56/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5256.4493\n",
      "Epoch 57/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4718.5422\n",
      "Epoch 58/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4737.1248\n",
      "Epoch 59/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4981.5277\n",
      "Epoch 60/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4764.8397\n",
      "Epoch 61/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4929.5977\n",
      "Epoch 62/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4995.4130\n",
      "Epoch 63/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4978.3660\n",
      "Epoch 64/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4951.4531\n",
      "Epoch 65/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4933.8014\n",
      "Epoch 66/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4836.4836\n",
      "Epoch 67/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4875.9418\n",
      "Epoch 68/500\n",
      "16594/16594 [==============================] - ETA: 0s - loss: 4808.18 - 3s 186us/step - loss: 4810.0096\n",
      "Epoch 69/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5029.7263\n",
      "Epoch 70/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4636.5276\n",
      "Epoch 71/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4961.5737\n",
      "Epoch 72/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 5058.9257\n",
      "Epoch 73/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5031.6993\n",
      "Epoch 74/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4785.6451\n",
      "Epoch 75/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4712.4800\n",
      "Epoch 76/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 5031.0844\n",
      "Epoch 77/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4879.0275\n",
      "Epoch 78/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4810.0677\n",
      "Epoch 79/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4807.2138\n",
      "Epoch 80/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4810.2436\n",
      "Epoch 81/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4913.2534\n",
      "Epoch 82/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5013.7752\n",
      "Epoch 83/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4820.7365\n",
      "Epoch 84/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4952.8178\n",
      "Epoch 85/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4668.1017\n",
      "Epoch 86/500\n",
      "16594/16594 [==============================] - 4s 240us/step - loss: 4725.9442\n",
      "Epoch 87/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4664.4712\n",
      "Epoch 88/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4824.3288\n",
      "Epoch 89/500\n",
      "16594/16594 [==============================] - 5s 307us/step - loss: 4746.7948\n",
      "Epoch 90/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4934.3328\n",
      "Epoch 91/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4795.9926\n",
      "Epoch 92/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 5053.0061\n",
      "Epoch 93/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4823.8865\n",
      "Epoch 94/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4619.4551\n",
      "Epoch 95/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4684.5870\n",
      "Epoch 96/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4890.8542\n",
      "Epoch 97/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4879.9094\n",
      "Epoch 98/500\n",
      "16594/16594 [==============================] - 4s 248us/step - loss: 4764.8746\n",
      "Epoch 99/500\n",
      "16594/16594 [==============================] - 3s 205us/step - loss: 4578.3467\n",
      "Epoch 100/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4563.0094\n",
      "Epoch 101/500\n",
      "16594/16594 [==============================] - 5s 275us/step - loss: 4920.0733\n",
      "Epoch 102/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4841.6279\n",
      "Epoch 103/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4768.5356\n",
      "Epoch 104/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4805.4554\n",
      "Epoch 105/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4648.3643\n",
      "Epoch 106/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4655.7942\n",
      "Epoch 107/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5067.5989\n",
      "Epoch 108/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4732.9823\n",
      "Epoch 109/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4656.2933\n",
      "Epoch 110/500\n",
      "16594/16594 [==============================] - 5s 276us/step - loss: 4648.1631\n",
      "Epoch 111/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 4620.5305\n",
      "Epoch 112/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4815.2534\n",
      "Epoch 113/500\n",
      "16594/16594 [==============================] - 5s 286us/step - loss: 4929.4283\n",
      "Epoch 114/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4630.9466\n",
      "Epoch 115/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4701.4606\n",
      "Epoch 116/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4781.3453\n",
      "Epoch 117/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4581.7874\n",
      "Epoch 118/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4589.2959\n",
      "Epoch 119/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4676.8271\n",
      "Epoch 120/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 5083.0908\n",
      "Epoch 121/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4681.8239\n",
      "Epoch 122/500\n",
      "16594/16594 [==============================] - 4s 213us/step - loss: 4841.7591\n",
      "Epoch 123/500\n",
      "16594/16594 [==============================] - 4s 224us/step - loss: 4567.5380\n",
      "Epoch 124/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4687.0457\n",
      "Epoch 125/500\n",
      "16594/16594 [==============================] - 4s 245us/step - loss: 4795.7234\n",
      "Epoch 126/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4582.3013\n",
      "Epoch 127/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4466.6003\n",
      "Epoch 128/500\n",
      "16594/16594 [==============================] - 4s 228us/step - loss: 4797.6752\n",
      "Epoch 129/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4869.3940\n",
      "Epoch 130/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4583.6728\n",
      "Epoch 131/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4934.7051\n",
      "Epoch 132/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 5005.5315\n",
      "Epoch 133/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4762.2570\n",
      "Epoch 134/500\n",
      "16594/16594 [==============================] - 5s 280us/step - loss: 4670.5350\n",
      "Epoch 135/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 4454.1036\n",
      "Epoch 136/500\n",
      "16594/16594 [==============================] - 5s 298us/step - loss: 5008.0933\n",
      "Epoch 137/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4394.6948\n",
      "Epoch 138/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4765.0014\n",
      "Epoch 139/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4966.8963\n",
      "Epoch 140/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4586.3192\n",
      "Epoch 141/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4461.2098\n",
      "Epoch 142/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4577.9091\n",
      "Epoch 143/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4739.7342\n",
      "Epoch 144/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4786.2308\n",
      "Epoch 145/500\n",
      "16594/16594 [==============================] - 5s 292us/step - loss: 4604.4554\n",
      "Epoch 146/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4891.5191\n",
      "Epoch 147/500\n",
      "16594/16594 [==============================] - 4s 216us/step - loss: 4796.8020\n",
      "Epoch 148/500\n",
      "16594/16594 [==============================] - 5s 273us/step - loss: 4702.6989\n",
      "Epoch 149/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4410.0292\n",
      "Epoch 150/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4631.7546\n",
      "Epoch 151/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4657.8646\n",
      "Epoch 152/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4715.1582\n",
      "Epoch 153/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4599.0790\n",
      "Epoch 154/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4570.9207\n",
      "Epoch 155/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4543.0715\n",
      "Epoch 156/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4571.1242\n",
      "Epoch 157/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4768.4952\n",
      "Epoch 158/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4466.1720\n",
      "Epoch 159/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4773.9915\n",
      "Epoch 160/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4540.9875\n",
      "Epoch 161/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4524.0809\n",
      "Epoch 162/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4455.4480\n",
      "Epoch 163/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4648.0511\n",
      "Epoch 164/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4629.2745\n",
      "Epoch 165/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4637.6330\n",
      "Epoch 166/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4687.4361\n",
      "Epoch 167/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4593.4383\n",
      "Epoch 168/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4625.0154\n",
      "Epoch 169/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4467.6745\n",
      "Epoch 170/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4580.2603\n",
      "Epoch 171/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4614.2509\n",
      "Epoch 172/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4619.5700\n",
      "Epoch 173/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4655.5603\n",
      "Epoch 174/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4458.3558\n",
      "Epoch 175/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4747.4110\n",
      "Epoch 176/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4525.6152\n",
      "Epoch 177/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4607.3017\n",
      "Epoch 178/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4639.7560\n",
      "Epoch 179/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4602.7533\n",
      "Epoch 180/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4740.5857\n",
      "Epoch 181/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4651.1425\n",
      "Epoch 182/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4479.9065\n",
      "Epoch 183/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4410.5061\n",
      "Epoch 184/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4511.0151\n",
      "Epoch 185/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4604.1313\n",
      "Epoch 186/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4828.2040\n",
      "Epoch 187/500\n",
      "16594/16594 [==============================] - 3s 204us/step - loss: 4485.5995\n",
      "Epoch 188/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 5142.7007\n",
      "Epoch 189/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4759.5710\n",
      "Epoch 190/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4887.1444\n",
      "Epoch 191/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4637.3435\n",
      "Epoch 192/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4327.4383\n",
      "Epoch 193/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4510.6522\n",
      "Epoch 194/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4333.2967\n",
      "Epoch 195/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4516.3410\n",
      "Epoch 196/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4639.3495\n",
      "Epoch 197/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4457.8159\n",
      "Epoch 198/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4450.5077\n",
      "Epoch 199/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4674.9337\n",
      "Epoch 200/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4574.2778\n",
      "Epoch 201/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 185us/step - loss: 4425.5004\n",
      "Epoch 202/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4455.4646\n",
      "Epoch 203/500\n",
      "16594/16594 [==============================] - 3s 209us/step - loss: 4584.2350\n",
      "Epoch 204/500\n",
      "16594/16594 [==============================] - 4s 220us/step - loss: 4422.3066\n",
      "Epoch 205/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4669.3963\n",
      "Epoch 206/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4448.9466\n",
      "Epoch 207/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4581.8705\n",
      "Epoch 208/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4656.0448\n",
      "Epoch 209/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4462.7333\n",
      "Epoch 210/500\n",
      "16594/16594 [==============================] - 5s 294us/step - loss: 4519.8321\n",
      "Epoch 211/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4594.9111\n",
      "Epoch 212/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4430.3218\n",
      "Epoch 213/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4391.8989\n",
      "Epoch 214/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4607.4399\n",
      "Epoch 215/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4548.4970\n",
      "Epoch 216/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4496.8726\n",
      "Epoch 217/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4336.8034\n",
      "Epoch 218/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4627.0484\n",
      "Epoch 219/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 4434.9413\n",
      "Epoch 220/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4406.9265\n",
      "Epoch 221/500\n",
      "16594/16594 [==============================] - 3s 206us/step - loss: 4563.5258\n",
      "Epoch 222/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 4528.3814\n",
      "Epoch 223/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4329.2510\n",
      "Epoch 224/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4723.9753\n",
      "Epoch 225/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4229.6103\n",
      "Epoch 226/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4437.6275\n",
      "Epoch 227/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4382.4987\n",
      "Epoch 228/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4594.7673\n",
      "Epoch 229/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4387.7580\n",
      "Epoch 230/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4360.6527\n",
      "Epoch 231/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4641.5065\n",
      "Epoch 232/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4394.0809\n",
      "Epoch 233/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4305.2842\n",
      "Epoch 234/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4572.6316\n",
      "Epoch 235/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4479.6476\n",
      "Epoch 236/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4406.9030\n",
      "Epoch 237/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4398.9176\n",
      "Epoch 238/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4492.1509\n",
      "Epoch 239/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4754.1717\n",
      "Epoch 240/500\n",
      "16594/16594 [==============================] - 3s 206us/step - loss: 4238.8784\n",
      "Epoch 241/500\n",
      "16594/16594 [==============================] - 4s 251us/step - loss: 4304.6926\n",
      "Epoch 242/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4535.7569\n",
      "Epoch 243/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4258.3125\n",
      "Epoch 244/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4261.2656\n",
      "Epoch 245/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4676.0952\n",
      "Epoch 246/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4638.2536\n",
      "Epoch 247/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4437.3698\n",
      "Epoch 248/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4312.7910\n",
      "Epoch 249/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4540.9023\n",
      "Epoch 250/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4608.0025\n",
      "Epoch 251/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4555.8532\n",
      "Epoch 252/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4393.7769\n",
      "Epoch 253/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4331.1626\n",
      "Epoch 254/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4576.4310\n",
      "Epoch 255/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4364.5587\n",
      "Epoch 256/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4501.0963\n",
      "Epoch 257/500\n",
      "16594/16594 [==============================] - 4s 254us/step - loss: 4360.2964\n",
      "Epoch 258/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4425.9164\n",
      "Epoch 259/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4254.2976\n",
      "Epoch 260/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4482.7802\n",
      "Epoch 261/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4322.0252\n",
      "Epoch 262/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4280.5678\n",
      "Epoch 263/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4663.1089\n",
      "Epoch 264/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4357.6630\n",
      "Epoch 265/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4291.1984\n",
      "Epoch 266/500\n",
      "16594/16594 [==============================] - 5s 286us/step - loss: 4577.7762\n",
      "Epoch 267/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4415.1541\n",
      "Epoch 268/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4316.3393\n",
      "Epoch 269/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4327.9699\n",
      "Epoch 270/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4407.6047\n",
      "Epoch 271/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4317.6568\n",
      "Epoch 272/500\n",
      "16594/16594 [==============================] - 5s 297us/step - loss: 4242.0562\n",
      "Epoch 273/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4735.9912\n",
      "Epoch 274/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4199.4003\n",
      "Epoch 275/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4351.1244\n",
      "Epoch 276/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4210.6954\n",
      "Epoch 277/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4595.5162\n",
      "Epoch 278/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4608.7712\n",
      "Epoch 279/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4471.5227\n",
      "Epoch 280/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4308.7542\n",
      "Epoch 281/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4279.2075\n",
      "Epoch 282/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4116.3290\n",
      "Epoch 283/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4624.2678\n",
      "Epoch 284/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4364.4096\n",
      "Epoch 285/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4477.3751\n",
      "Epoch 286/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4268.9978\n",
      "Epoch 287/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4198.6344\n",
      "Epoch 288/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4369.9523\n",
      "Epoch 289/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4324.7575\n",
      "Epoch 290/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4481.8226\n",
      "Epoch 291/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4417.2558\n",
      "Epoch 292/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4159.1246\n",
      "Epoch 293/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4491.9974\n",
      "Epoch 294/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4374.3196\n",
      "Epoch 295/500\n",
      "16594/16594 [==============================] - 3s 178us/step - loss: 4112.2051\n",
      "Epoch 296/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4267.3004\n",
      "Epoch 297/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4489.7603\n",
      "Epoch 298/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4333.0743\n",
      "Epoch 299/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 5006.7822\n",
      "Epoch 300/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4307.2976\n",
      "Epoch 301/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4140.6371\n",
      "Epoch 302/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4263.1708\n",
      "Epoch 303/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4373.3114\n",
      "Epoch 304/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4477.4421\n",
      "Epoch 305/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4276.6028\n",
      "Epoch 306/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4528.6804\n",
      "Epoch 307/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4222.0453\n",
      "Epoch 308/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4202.9459\n",
      "Epoch 309/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4229.7613\n",
      "Epoch 310/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4211.8286\n",
      "Epoch 311/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4325.8547\n",
      "Epoch 312/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4240.0478\n",
      "Epoch 313/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4452.8930\n",
      "Epoch 314/500\n",
      "16594/16594 [==============================] - 3s 204us/step - loss: 4262.1781\n",
      "Epoch 315/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 4121.0613\n",
      "Epoch 316/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4448.0635\n",
      "Epoch 317/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4399.3984\n",
      "Epoch 318/500\n",
      "16594/16594 [==============================] - 4s 220us/step - loss: 4477.7993\n",
      "Epoch 319/500\n",
      "16594/16594 [==============================] - 4s 213us/step - loss: 4222.2252\n",
      "Epoch 320/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3955.8275\n",
      "Epoch 321/500\n",
      "16594/16594 [==============================] - 5s 275us/step - loss: 4535.0163\n",
      "Epoch 322/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 5391.8506\n",
      "Epoch 323/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4111.9284\n",
      "Epoch 324/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3967.3248\n",
      "Epoch 325/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4638.5324\n",
      "Epoch 326/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4160.3573\n",
      "Epoch 327/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4146.0539\n",
      "Epoch 328/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4265.7697\n",
      "Epoch 329/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4252.8245\n",
      "Epoch 330/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4503.2093\n",
      "Epoch 331/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4759.8446\n",
      "Epoch 332/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4148.7124\n",
      "Epoch 333/500\n",
      "16594/16594 [==============================] - 4s 268us/step - loss: 4157.5846\n",
      "Epoch 334/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4229.1894\n",
      "Epoch 335/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4280.5861\n",
      "Epoch 336/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4353.0044\n",
      "Epoch 337/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4045.8951\n",
      "Epoch 338/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4094.4056\n",
      "Epoch 339/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4399.3517\n",
      "Epoch 340/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4249.5882\n",
      "Epoch 341/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4243.1415\n",
      "Epoch 342/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4363.5352\n",
      "Epoch 343/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4182.8807\n",
      "Epoch 344/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4691.5493\n",
      "Epoch 345/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4015.2442\n",
      "Epoch 346/500\n",
      "16594/16594 [==============================] - 4s 227us/step - loss: 4475.6397\n",
      "Epoch 347/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4137.6425\n",
      "Epoch 348/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4255.2249\n",
      "Epoch 349/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4114.6472\n",
      "Epoch 350/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4343.2726\n",
      "Epoch 351/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4194.9893\n",
      "Epoch 352/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4148.1234\n",
      "Epoch 353/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4347.4834\n",
      "Epoch 354/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4164.1681\n",
      "Epoch 355/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4313.4477\n",
      "Epoch 356/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4630.2249\n",
      "Epoch 357/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4036.7653\n",
      "Epoch 358/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4118.7519\n",
      "Epoch 359/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4165.6269\n",
      "Epoch 360/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4136.0813\n",
      "Epoch 361/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4362.3569\n",
      "Epoch 362/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4306.8464\n",
      "Epoch 363/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4122.1357\n",
      "Epoch 364/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4275.1669\n",
      "Epoch 365/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4107.9316\n",
      "Epoch 366/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4104.5381\n",
      "Epoch 367/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4214.6461\n",
      "Epoch 368/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3934.2990\n",
      "Epoch 369/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4406.6596\n",
      "Epoch 370/500\n",
      "16594/16594 [==============================] - 3s 178us/step - loss: 4270.4674\n",
      "Epoch 371/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4379.1053\n",
      "Epoch 372/500\n",
      "16594/16594 [==============================] - 5s 281us/step - loss: 4307.0722\n",
      "Epoch 373/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4062.5254\n",
      "Epoch 374/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4204.3673\n",
      "Epoch 375/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4519.9357\n",
      "Epoch 376/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4254.0420\n",
      "Epoch 377/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 183us/step - loss: 4004.0553\n",
      "Epoch 378/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4237.9006\n",
      "Epoch 379/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4369.8085\n",
      "Epoch 380/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4039.0872\n",
      "Epoch 381/500\n",
      "16594/16594 [==============================] - 4s 215us/step - loss: 4108.6442\n",
      "Epoch 382/500\n",
      "16594/16594 [==============================] - 4s 257us/step - loss: 4038.4077\n",
      "Epoch 383/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4095.1002\n",
      "Epoch 384/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4372.7173\n",
      "Epoch 385/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4017.7590\n",
      "Epoch 386/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4166.5724\n",
      "Epoch 387/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4379.1452\n",
      "Epoch 388/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4322.9175\n",
      "Epoch 389/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4233.3861\n",
      "Epoch 390/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4075.7258\n",
      "Epoch 391/500\n",
      "16594/16594 [==============================] - 4s 271us/step - loss: 4102.6654\n",
      "Epoch 392/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4103.4716\n",
      "Epoch 393/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4013.5212\n",
      "Epoch 394/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4065.0908\n",
      "Epoch 395/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4202.9197\n",
      "Epoch 396/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4065.8874\n",
      "Epoch 397/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4195.5977\n",
      "Epoch 398/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4112.4646\n",
      "Epoch 399/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4152.9265\n",
      "Epoch 400/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4251.8734\n",
      "Epoch 401/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4026.5413\n",
      "Epoch 402/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4366.3331\n",
      "Epoch 403/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4005.2646\n",
      "Epoch 404/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4231.5024\n",
      "Epoch 405/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4291.0770\n",
      "Epoch 406/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4155.8475\n",
      "Epoch 407/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4300.8315\n",
      "Epoch 408/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4013.6310\n",
      "Epoch 409/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4031.6994\n",
      "Epoch 410/500\n",
      "16594/16594 [==============================] - 5s 295us/step - loss: 4032.4172\n",
      "Epoch 411/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4189.7565\n",
      "Epoch 412/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4130.9843\n",
      "Epoch 413/500\n",
      "16594/16594 [==============================] - 4s 239us/step - loss: 4256.4813\n",
      "Epoch 414/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4120.7974\n",
      "Epoch 415/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 4090.4473\n",
      "Epoch 416/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4041.7178\n",
      "Epoch 417/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4182.4749\n",
      "Epoch 418/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4132.8390\n",
      "Epoch 419/500\n",
      "16594/16594 [==============================] - 4s 257us/step - loss: 4320.6619\n",
      "Epoch 420/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 3995.9866\n",
      "Epoch 421/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4035.9716\n",
      "Epoch 422/500\n",
      "16594/16594 [==============================] - 3s 178us/step - loss: 4014.8277\n",
      "Epoch 423/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4153.7577\n",
      "Epoch 424/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4012.0887\n",
      "Epoch 425/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4192.8300\n",
      "Epoch 426/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4109.8784\n",
      "Epoch 427/500\n",
      "16594/16594 [==============================] - 3s 177us/step - loss: 4156.7871\n",
      "Epoch 428/500\n",
      "16594/16594 [==============================] - 4s 243us/step - loss: 4102.2731\n",
      "Epoch 429/500\n",
      "16594/16594 [==============================] - 4s 222us/step - loss: 4090.9575\n",
      "Epoch 430/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 4025.5164\n",
      "Epoch 431/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4113.2864\n",
      "Epoch 432/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3869.4352\n",
      "Epoch 433/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4022.8326\n",
      "Epoch 434/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4439.5519\n",
      "Epoch 435/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3963.2822\n",
      "Epoch 436/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4213.6803\n",
      "Epoch 437/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4143.6285\n",
      "Epoch 438/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4124.7964\n",
      "Epoch 439/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4076.6786\n",
      "Epoch 440/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4322.2055\n",
      "Epoch 441/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4033.9529\n",
      "Epoch 442/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4125.2151\n",
      "Epoch 443/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3826.2669\n",
      "Epoch 444/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4368.9833\n",
      "Epoch 445/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4189.9062\n",
      "Epoch 446/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3915.4542\n",
      "Epoch 447/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4084.8339\n",
      "Epoch 448/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3963.3043\n",
      "Epoch 449/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4355.2840\n",
      "Epoch 450/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4047.1825\n",
      "Epoch 451/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3948.0346\n",
      "Epoch 452/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4029.7459\n",
      "Epoch 453/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4151.5442\n",
      "Epoch 454/500\n",
      "16594/16594 [==============================] - 4s 263us/step - loss: 4070.6055\n",
      "Epoch 455/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4046.7821\n",
      "Epoch 456/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4118.7066\n",
      "Epoch 457/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3860.0014\n",
      "Epoch 458/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4035.4648\n",
      "Epoch 459/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4158.8754\n",
      "Epoch 460/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4762.2113\n",
      "Epoch 461/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3818.6507\n",
      "Epoch 462/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3989.9552\n",
      "Epoch 463/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3927.0693\n",
      "Epoch 464/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4126.7693\n",
      "Epoch 465/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4276.7890\n",
      "Epoch 466/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4104.4454\n",
      "Epoch 467/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4101.8413\n",
      "Epoch 468/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4050.6223\n",
      "Epoch 469/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4091.4312\n",
      "Epoch 470/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 3891.7622\n",
      "Epoch 471/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4295.6618\n",
      "Epoch 472/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3893.6083\n",
      "Epoch 473/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4022.3406\n",
      "Epoch 474/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3975.3452\n",
      "Epoch 475/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3996.2969\n",
      "Epoch 476/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4184.9239\n",
      "Epoch 477/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3916.1347\n",
      "Epoch 478/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4059.9809\n",
      "Epoch 479/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3933.5232\n",
      "Epoch 480/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3986.1295\n",
      "Epoch 481/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3904.5322\n",
      "Epoch 482/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4044.4866\n",
      "Epoch 483/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4068.7959\n",
      "Epoch 484/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4304.1036\n",
      "Epoch 485/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3829.2719\n",
      "Epoch 486/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3961.7469\n",
      "Epoch 487/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4155.0468\n",
      "Epoch 488/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4113.1281\n",
      "Epoch 489/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4125.3085\n",
      "Epoch 490/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3996.7467\n",
      "Epoch 491/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3961.1696\n",
      "Epoch 492/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3814.7967\n",
      "Epoch 493/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4570.8654\n",
      "Epoch 494/500\n",
      "16594/16594 [==============================] - 5s 302us/step - loss: 3815.3233\n",
      "Epoch 495/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3846.9966\n",
      "Epoch 496/500\n",
      "16594/16594 [==============================] - 5s 275us/step - loss: 3929.1787\n",
      "Epoch 497/500\n",
      "16594/16594 [==============================] - 4s 213us/step - loss: 4305.0581\n",
      "Epoch 498/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4115.4484\n",
      "Epoch 499/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3763.1004\n",
      "Epoch 500/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4098.1542\n",
      "1843/1843 [==============================] - 0s 44us/step\n",
      "Epoch 1/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 5572.4589\n",
      "Epoch 2/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 5248.7292\n",
      "Epoch 3/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4850.5799\n",
      "Epoch 4/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4641.1367\n",
      "Epoch 5/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 4513.7191\n",
      "Epoch 6/500\n",
      "16594/16594 [==============================] - 3s 210us/step - loss: 4744.8626\n",
      "Epoch 7/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4636.5912\n",
      "Epoch 8/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4555.9124\n",
      "Epoch 9/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4545.5728\n",
      "Epoch 10/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4478.9927\n",
      "Epoch 11/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4705.2607\n",
      "Epoch 12/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4470.7939\n",
      "Epoch 13/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4471.8431\n",
      "Epoch 14/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4583.3859\n",
      "Epoch 15/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4409.8123\n",
      "Epoch 16/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4606.5597\n",
      "Epoch 17/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4579.0539\n",
      "Epoch 18/500\n",
      "16594/16594 [==============================] - 4s 268us/step - loss: 4298.1979\n",
      "Epoch 19/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4305.8598\n",
      "Epoch 20/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4392.8076\n",
      "Epoch 21/500\n",
      "16594/16594 [==============================] - 5s 278us/step - loss: 4642.5210\n",
      "Epoch 22/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4681.3363\n",
      "Epoch 23/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4267.5199\n",
      "Epoch 24/500\n",
      "16594/16594 [==============================] - 4s 265us/step - loss: 4453.2748\n",
      "Epoch 25/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4357.2135\n",
      "Epoch 26/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4761.9537\n",
      "Epoch 27/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4440.7295\n",
      "Epoch 28/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4307.6719\n",
      "Epoch 29/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4559.6406\n",
      "Epoch 30/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4358.2633\n",
      "Epoch 31/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4379.2545\n",
      "Epoch 32/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4332.4518\n",
      "Epoch 33/500\n",
      "16594/16594 [==============================] - 4s 252us/step - loss: 4439.5212\n",
      "Epoch 34/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 4325.2078\n",
      "Epoch 35/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4473.8612\n",
      "Epoch 36/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4421.6781\n",
      "Epoch 37/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4392.1711\n",
      "Epoch 38/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4351.3029\n",
      "Epoch 39/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4749.9606\n",
      "Epoch 40/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4255.8983\n",
      "Epoch 41/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4229.0599\n",
      "Epoch 42/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4216.9542\n",
      "Epoch 43/500\n",
      "16594/16594 [==============================] - 5s 280us/step - loss: 4171.6755\n",
      "Epoch 44/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4583.6073\n",
      "Epoch 45/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4221.1952\n",
      "Epoch 46/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4195.1156\n",
      "Epoch 47/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4210.9641\n",
      "Epoch 48/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4424.4109\n",
      "Epoch 49/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4300.4969\n",
      "Epoch 50/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4148.1737\n",
      "Epoch 51/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4175.4678\n",
      "Epoch 52/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4577.1895\n",
      "Epoch 53/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 184us/step - loss: 4177.9196\n",
      "Epoch 54/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4234.7789\n",
      "Epoch 55/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4287.5523\n",
      "Epoch 56/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4259.3629\n",
      "Epoch 57/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4409.7530\n",
      "Epoch 58/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4332.8485\n",
      "Epoch 59/500\n",
      "16594/16594 [==============================] - 5s 282us/step - loss: 4271.5048\n",
      "Epoch 60/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4192.7761\n",
      "Epoch 61/500\n",
      "16594/16594 [==============================] - 5s 280us/step - loss: 4367.1224\n",
      "Epoch 62/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4607.9782\n",
      "Epoch 63/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4121.9176\n",
      "Epoch 64/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4217.2599\n",
      "Epoch 65/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4283.7849\n",
      "Epoch 66/500\n",
      "16594/16594 [==============================] - 3s 178us/step - loss: 4152.0761\n",
      "Epoch 67/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4246.0222\n",
      "Epoch 68/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4284.5206\n",
      "Epoch 69/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4106.3078\n",
      "Epoch 70/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4248.6527\n",
      "Epoch 71/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4291.9528\n",
      "Epoch 72/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4420.4629\n",
      "Epoch 73/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4324.6492\n",
      "Epoch 74/500\n",
      "16594/16594 [==============================] - 5s 312us/step - loss: 4424.7922\n",
      "Epoch 75/500\n",
      "16594/16594 [==============================] - 3s 205us/step - loss: 4186.2662\n",
      "Epoch 76/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4305.9416\n",
      "Epoch 77/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4160.2442\n",
      "Epoch 78/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4071.5204\n",
      "Epoch 79/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4112.6170\n",
      "Epoch 80/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4372.0262\n",
      "Epoch 81/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4352.3885\n",
      "Epoch 82/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4186.2063\n",
      "Epoch 83/500\n",
      "16594/16594 [==============================] - 5s 293us/step - loss: 4145.7513\n",
      "Epoch 84/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 4635.4944\n",
      "Epoch 85/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4086.7034\n",
      "Epoch 86/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4240.6814\n",
      "Epoch 87/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4093.5449\n",
      "Epoch 88/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4103.9020\n",
      "Epoch 89/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3956.9396\n",
      "Epoch 90/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4382.5094\n",
      "Epoch 91/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4134.9508\n",
      "Epoch 92/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4148.5660\n",
      "Epoch 93/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4358.1480\n",
      "Epoch 94/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4206.4527\n",
      "Epoch 95/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4354.9519\n",
      "Epoch 96/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4231.1901\n",
      "Epoch 97/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4219.3814\n",
      "Epoch 98/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4267.7598\n",
      "Epoch 99/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4119.1479\n",
      "Epoch 100/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4017.0948\n",
      "Epoch 101/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4097.6977\n",
      "Epoch 102/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4111.9690\n",
      "Epoch 103/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4226.7956\n",
      "Epoch 104/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4112.4893\n",
      "Epoch 105/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4163.8722\n",
      "Epoch 106/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3992.9917\n",
      "Epoch 107/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4253.5798\n",
      "Epoch 108/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 4091.2037\n",
      "Epoch 109/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4130.5700\n",
      "Epoch 110/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4609.7354\n",
      "Epoch 111/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4250.4810\n",
      "Epoch 112/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3952.7682\n",
      "Epoch 113/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4252.9317\n",
      "Epoch 114/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4420.9645\n",
      "Epoch 115/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3818.3483\n",
      "Epoch 116/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4468.8954\n",
      "Epoch 117/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4022.7996\n",
      "Epoch 118/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4275.5053\n",
      "Epoch 119/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4106.5032\n",
      "Epoch 120/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4015.5220\n",
      "Epoch 121/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4416.3403\n",
      "Epoch 122/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4140.5079\n",
      "Epoch 123/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4409.3681\n",
      "Epoch 124/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4222.1362\n",
      "Epoch 125/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4189.7198\n",
      "Epoch 126/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3896.6096\n",
      "Epoch 127/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4212.7336\n",
      "Epoch 128/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4305.3459\n",
      "Epoch 129/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4135.4118\n",
      "Epoch 130/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4080.8658\n",
      "Epoch 131/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4282.5668\n",
      "Epoch 132/500\n",
      "16594/16594 [==============================] - 4s 257us/step - loss: 3892.7190\n",
      "Epoch 133/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 4085.3378\n",
      "Epoch 134/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 4197.1235\n",
      "Epoch 135/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3752.5420\n",
      "Epoch 136/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4100.8975\n",
      "Epoch 137/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4318.0076\n",
      "Epoch 138/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4024.4815\n",
      "Epoch 139/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4110.7492\n",
      "Epoch 140/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4038.9144\n",
      "Epoch 141/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4396.3829\n",
      "Epoch 142/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4267.8032\n",
      "Epoch 143/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3838.7288\n",
      "Epoch 144/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4173.3992\n",
      "Epoch 145/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4144.7277\n",
      "Epoch 146/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 4103.5083\n",
      "Epoch 147/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4033.4327\n",
      "Epoch 148/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 4035.2346\n",
      "Epoch 149/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4113.6053\n",
      "Epoch 150/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4047.3244\n",
      "Epoch 151/500\n",
      "16594/16594 [==============================] - 3s 210us/step - loss: 4154.1036\n",
      "Epoch 152/500\n",
      "16594/16594 [==============================] - 4s 253us/step - loss: 4108.8923\n",
      "Epoch 153/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3984.7086\n",
      "Epoch 154/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4105.6308\n",
      "Epoch 155/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3871.7376\n",
      "Epoch 156/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3929.0335\n",
      "Epoch 157/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3961.6200\n",
      "Epoch 158/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4030.0170\n",
      "Epoch 159/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4197.9400\n",
      "Epoch 160/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4113.5727\n",
      "Epoch 161/500\n",
      "16594/16594 [==============================] - 5s 298us/step - loss: 3847.2035\n",
      "Epoch 162/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 4281.0875\n",
      "Epoch 163/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3913.8430\n",
      "Epoch 164/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4318.5255\n",
      "Epoch 165/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3886.1585\n",
      "Epoch 166/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4239.9803\n",
      "Epoch 167/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3945.4163\n",
      "Epoch 168/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3799.2998\n",
      "Epoch 169/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 4144.5683\n",
      "Epoch 170/500\n",
      "16594/16594 [==============================] - 4s 269us/step - loss: 4123.5353\n",
      "Epoch 171/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 4183.2156\n",
      "Epoch 172/500\n",
      "16594/16594 [==============================] - 3s 204us/step - loss: 3767.9893\n",
      "Epoch 173/500\n",
      "16594/16594 [==============================] - 4s 261us/step - loss: 4205.6188\n",
      "Epoch 174/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 3978.7103\n",
      "Epoch 175/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4160.0251\n",
      "Epoch 176/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4042.6590\n",
      "Epoch 177/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3920.5135\n",
      "Epoch 178/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4176.5214\n",
      "Epoch 179/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3940.7350\n",
      "Epoch 180/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3969.0493\n",
      "Epoch 181/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4003.2272\n",
      "Epoch 182/500\n",
      "16594/16594 [==============================] - 5s 308us/step - loss: 3898.6570\n",
      "Epoch 183/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 4058.0647\n",
      "Epoch 184/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3920.5257\n",
      "Epoch 185/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4096.3989\n",
      "Epoch 186/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4156.9780\n",
      "Epoch 187/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3766.1157\n",
      "Epoch 188/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3922.8584\n",
      "Epoch 189/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4010.6844\n",
      "Epoch 190/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3960.8578\n",
      "Epoch 191/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3975.7073\n",
      "Epoch 192/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 4197.5485\n",
      "Epoch 193/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4157.3336\n",
      "Epoch 194/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4021.8939\n",
      "Epoch 195/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4047.1250\n",
      "Epoch 196/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3787.2553\n",
      "Epoch 197/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4120.5683\n",
      "Epoch 198/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3893.0219\n",
      "Epoch 199/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3833.2200\n",
      "Epoch 200/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3957.7765\n",
      "Epoch 201/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4101.6842\n",
      "Epoch 202/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3936.3699\n",
      "Epoch 203/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4149.2506\n",
      "Epoch 204/500\n",
      "16594/16594 [==============================] - 4s 228us/step - loss: 4057.1273\n",
      "Epoch 205/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3936.5376\n",
      "Epoch 206/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4038.5744\n",
      "Epoch 207/500\n",
      "16594/16594 [==============================] - ETA: 0s - loss: 4021.14 - 3s 189us/step - loss: 4022.0128\n",
      "Epoch 208/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3732.7875\n",
      "Epoch 209/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3761.2927\n",
      "Epoch 210/500\n",
      "16594/16594 [==============================] - 4s 230us/step - loss: 4038.6570\n",
      "Epoch 211/500\n",
      "16594/16594 [==============================] - 5s 283us/step - loss: 4105.1883\n",
      "Epoch 212/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3981.0708\n",
      "Epoch 213/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3829.0522\n",
      "Epoch 214/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4054.6763\n",
      "Epoch 215/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3996.0736\n",
      "Epoch 216/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4255.0215\n",
      "Epoch 217/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3786.8063\n",
      "Epoch 218/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3963.6302\n",
      "Epoch 219/500\n",
      "16594/16594 [==============================] - 3s 206us/step - loss: 3995.3048\n",
      "Epoch 220/500\n",
      "16594/16594 [==============================] - 4s 230us/step - loss: 4132.2840\n",
      "Epoch 221/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 3886.6414\n",
      "Epoch 222/500\n",
      "16594/16594 [==============================] - 4s 245us/step - loss: 3677.9357\n",
      "Epoch 223/500\n",
      "16594/16594 [==============================] - 4s 222us/step - loss: 4035.1361\n",
      "Epoch 224/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 3944.0074\n",
      "Epoch 225/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4023.3466\n",
      "Epoch 226/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4008.5617\n",
      "Epoch 227/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3788.6244\n",
      "Epoch 228/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 4251.2210\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 229/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3774.8219\n",
      "Epoch 230/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 3844.2877\n",
      "Epoch 231/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4181.1413\n",
      "Epoch 232/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3627.2318\n",
      "Epoch 233/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4051.2025\n",
      "Epoch 234/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3889.2029\n",
      "Epoch 235/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 4097.0145\n",
      "Epoch 236/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3795.4397\n",
      "Epoch 237/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3917.0574\n",
      "Epoch 238/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3945.2614\n",
      "Epoch 239/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3836.6939\n",
      "Epoch 240/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3876.0412\n",
      "Epoch 241/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4041.6308\n",
      "Epoch 242/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3935.7460\n",
      "Epoch 243/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3634.0728\n",
      "Epoch 244/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4181.0671\n",
      "Epoch 245/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4113.4134\n",
      "Epoch 246/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3937.4616\n",
      "Epoch 247/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4108.1145\n",
      "Epoch 248/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3681.6130\n",
      "Epoch 249/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3945.0742\n",
      "Epoch 250/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3793.0162\n",
      "Epoch 251/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3985.5422\n",
      "Epoch 252/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3963.2893\n",
      "Epoch 253/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3980.3847\n",
      "Epoch 254/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3945.7047\n",
      "Epoch 255/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3809.7151\n",
      "Epoch 256/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3843.6722\n",
      "Epoch 257/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3872.0900\n",
      "Epoch 258/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3896.8301\n",
      "Epoch 259/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3987.8905\n",
      "Epoch 260/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3877.3666\n",
      "Epoch 261/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3761.9918\n",
      "Epoch 262/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3971.3406\n",
      "Epoch 263/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3866.8951\n",
      "Epoch 264/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3854.4401\n",
      "Epoch 265/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3941.3652\n",
      "Epoch 266/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3840.1879\n",
      "Epoch 267/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3748.1237\n",
      "Epoch 268/500\n",
      "16594/16594 [==============================] - 4s 241us/step - loss: 3950.9705\n",
      "Epoch 269/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3847.0187\n",
      "Epoch 270/500\n",
      "16594/16594 [==============================] - 4s 212us/step - loss: 3940.2032\n",
      "Epoch 271/500\n",
      "16594/16594 [==============================] - 4s 240us/step - loss: 3660.8420\n",
      "Epoch 272/500\n",
      "16594/16594 [==============================] - 3s 204us/step - loss: 4040.7803\n",
      "Epoch 273/500\n",
      "16594/16594 [==============================] - 5s 275us/step - loss: 4080.6862\n",
      "Epoch 274/500\n",
      "16594/16594 [==============================] - 3s 199us/step - loss: 3664.2836\n",
      "Epoch 275/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3933.3945\n",
      "Epoch 276/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3673.0423\n",
      "Epoch 277/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3777.0214\n",
      "Epoch 278/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3914.0092\n",
      "Epoch 279/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3888.3582\n",
      "Epoch 280/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3803.6512\n",
      "Epoch 281/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3834.3350\n",
      "Epoch 282/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3836.3891\n",
      "Epoch 283/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3818.4201\n",
      "Epoch 284/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3873.4646\n",
      "Epoch 285/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3908.7039\n",
      "Epoch 286/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3641.7362\n",
      "Epoch 287/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4030.3620\n",
      "Epoch 288/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3961.4582\n",
      "Epoch 289/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3786.6379\n",
      "Epoch 290/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3536.5380\n",
      "Epoch 291/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3846.4907\n",
      "Epoch 292/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3887.1145\n",
      "Epoch 293/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3714.8130\n",
      "Epoch 294/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3868.9827\n",
      "Epoch 295/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3727.0314\n",
      "Epoch 296/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3806.9946\n",
      "Epoch 297/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3738.3507\n",
      "Epoch 298/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3877.5382\n",
      "Epoch 299/500\n",
      "16594/16594 [==============================] - 5s 285us/step - loss: 4680.5790\n",
      "Epoch 300/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 3668.2180\n",
      "Epoch 301/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3561.2733\n",
      "Epoch 302/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3757.8595\n",
      "Epoch 303/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3663.1501\n",
      "Epoch 304/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3867.0343\n",
      "Epoch 305/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3650.0133\n",
      "Epoch 306/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3770.4712\n",
      "Epoch 307/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 4047.1023\n",
      "Epoch 308/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3844.5168\n",
      "Epoch 309/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3928.6473\n",
      "Epoch 310/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3954.7888\n",
      "Epoch 311/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3847.1148\n",
      "Epoch 312/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3698.2596\n",
      "Epoch 313/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3708.6344\n",
      "Epoch 314/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3766.8889\n",
      "Epoch 315/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3877.7122\n",
      "Epoch 316/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4095.3026\n",
      "Epoch 317/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 183us/step - loss: 3647.3305\n",
      "Epoch 318/500\n",
      "16594/16594 [==============================] - 3s 203us/step - loss: 3707.9801\n",
      "Epoch 319/500\n",
      "16594/16594 [==============================] - 4s 222us/step - loss: 3899.4905\n",
      "Epoch 320/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3840.6343\n",
      "Epoch 321/500\n",
      "16594/16594 [==============================] - 4s 224us/step - loss: 3717.6714\n",
      "Epoch 322/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 3609.1160\n",
      "Epoch 323/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3764.5600\n",
      "Epoch 324/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 4033.3121\n",
      "Epoch 325/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 4014.5381\n",
      "Epoch 326/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3684.6062\n",
      "Epoch 327/500\n",
      "16594/16594 [==============================] - 4s 213us/step - loss: 3767.8598\n",
      "Epoch 328/500\n",
      "16594/16594 [==============================] - 4s 214us/step - loss: 3857.7407\n",
      "Epoch 329/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3937.9517\n",
      "Epoch 330/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3857.9894\n",
      "Epoch 331/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3845.0014\n",
      "Epoch 332/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3660.1290\n",
      "Epoch 333/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4007.9854\n",
      "Epoch 334/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3523.4239\n",
      "Epoch 335/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3791.4634\n",
      "Epoch 336/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3699.6782\n",
      "Epoch 337/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3544.8299\n",
      "Epoch 338/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3826.1640\n",
      "Epoch 339/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3927.5417\n",
      "Epoch 340/500\n",
      "16594/16594 [==============================] - 4s 221us/step - loss: 3824.6271\n",
      "Epoch 341/500\n",
      "16594/16594 [==============================] - 4s 253us/step - loss: 3750.4454\n",
      "Epoch 342/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3644.3719\n",
      "Epoch 343/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3676.9116\n",
      "Epoch 344/500\n",
      "16594/16594 [==============================] - 3s 177us/step - loss: 3987.1038\n",
      "Epoch 345/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3708.3188\n",
      "Epoch 346/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3698.8130\n",
      "Epoch 347/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3659.3458\n",
      "Epoch 348/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3972.1663\n",
      "Epoch 349/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 4054.5490\n",
      "Epoch 350/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3713.8339\n",
      "Epoch 351/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3547.1622\n",
      "Epoch 352/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3845.9452\n",
      "Epoch 353/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3904.3290\n",
      "Epoch 354/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3585.0491\n",
      "Epoch 355/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3692.2111\n",
      "Epoch 356/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3989.2127\n",
      "Epoch 357/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3724.8936\n",
      "Epoch 358/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3619.3518\n",
      "Epoch 359/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4024.5456\n",
      "Epoch 360/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 3909.3306\n",
      "Epoch 361/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3797.5135\n",
      "Epoch 362/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3483.8092\n",
      "Epoch 363/500\n",
      "16594/16594 [==============================] - 5s 298us/step - loss: 3624.3209\n",
      "Epoch 364/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3866.5254\n",
      "Epoch 365/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3825.4600\n",
      "Epoch 366/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3677.2621\n",
      "Epoch 367/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3669.6087\n",
      "Epoch 368/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3775.6506\n",
      "Epoch 369/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3664.6447\n",
      "Epoch 370/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3731.5655\n",
      "Epoch 371/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3830.3659\n",
      "Epoch 372/500\n",
      "16594/16594 [==============================] - 4s 269us/step - loss: 3647.6629\n",
      "Epoch 373/500\n",
      "16594/16594 [==============================] - 3s 201us/step - loss: 3747.2503\n",
      "Epoch 374/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3565.7179\n",
      "Epoch 375/500\n",
      "16594/16594 [==============================] - 5s 276us/step - loss: 3797.5885\n",
      "Epoch 376/500\n",
      "16594/16594 [==============================] - 3s 202us/step - loss: 4065.6661\n",
      "Epoch 377/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3648.9925\n",
      "Epoch 378/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3740.4332\n",
      "Epoch 379/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3788.2360\n",
      "Epoch 380/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3692.4138\n",
      "Epoch 381/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3517.0178\n",
      "Epoch 382/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 4022.1720\n",
      "Epoch 383/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3717.2575\n",
      "Epoch 384/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3851.6499\n",
      "Epoch 385/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3963.5190\n",
      "Epoch 386/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3572.5422\n",
      "Epoch 387/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3579.0100\n",
      "Epoch 388/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 4038.0838\n",
      "Epoch 389/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3483.3234\n",
      "Epoch 390/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 4287.3113\n",
      "Epoch 391/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3539.8256\n",
      "Epoch 392/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3567.2950\n",
      "Epoch 393/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3838.1272\n",
      "Epoch 394/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3744.7043\n",
      "Epoch 395/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3530.7664\n",
      "Epoch 396/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3967.7008\n",
      "Epoch 397/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3710.0235\n",
      "Epoch 398/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3751.5543\n",
      "Epoch 399/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3407.0906\n",
      "Epoch 400/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3597.8654\n",
      "Epoch 401/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 4062.6993\n",
      "Epoch 402/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3782.3866\n",
      "Epoch 403/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3519.9781\n",
      "Epoch 404/500\n",
      "16594/16594 [==============================] - 5s 287us/step - loss: 3640.2967\n",
      "Epoch 405/500\n",
      "16594/16594 [==============================] - 3s 193us/step - loss: 3703.8180\n",
      "Epoch 406/500\n",
      "16594/16594 [==============================] - 3s 205us/step - loss: 3845.0898\n",
      "Epoch 407/500\n",
      "16594/16594 [==============================] - 4s 271us/step - loss: 3883.5382\n",
      "Epoch 408/500\n",
      "16594/16594 [==============================] - 3s 195us/step - loss: 3488.6376\n",
      "Epoch 409/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3747.9646\n",
      "Epoch 410/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3689.6498\n",
      "Epoch 411/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3660.7879\n",
      "Epoch 412/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3556.3223\n",
      "Epoch 413/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3659.2824\n",
      "Epoch 414/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3711.6040\n",
      "Epoch 415/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3625.7776\n",
      "Epoch 416/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3773.6079\n",
      "Epoch 417/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3640.8785\n",
      "Epoch 418/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3732.8591\n",
      "Epoch 419/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3529.6285\n",
      "Epoch 420/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3951.9658\n",
      "Epoch 421/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3613.9806\n",
      "Epoch 422/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3569.2097\n",
      "Epoch 423/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3703.8335\n",
      "Epoch 424/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3629.5086\n",
      "Epoch 425/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3582.1561\n",
      "Epoch 426/500\n",
      "16594/16594 [==============================] - 4s 241us/step - loss: 3740.6761\n",
      "Epoch 427/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3797.2261\n",
      "Epoch 428/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3708.5818\n",
      "Epoch 429/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3756.1707\n",
      "Epoch 430/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3692.5399\n",
      "Epoch 431/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3634.9225\n",
      "Epoch 432/500\n",
      "16594/16594 [==============================] - 4s 271us/step - loss: 3705.0796\n",
      "Epoch 433/500\n",
      "16594/16594 [==============================] - 3s 206us/step - loss: 3503.9897\n",
      "Epoch 434/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3821.2056\n",
      "Epoch 435/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3888.9564\n",
      "Epoch 436/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3665.6708\n",
      "Epoch 437/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3555.7150\n",
      "Epoch 438/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3528.6030\n",
      "Epoch 439/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3653.2470\n",
      "Epoch 440/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3811.8813\n",
      "Epoch 441/500\n",
      "16594/16594 [==============================] - 5s 331us/step - loss: 3563.4152\n",
      "Epoch 442/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3533.3506\n",
      "Epoch 443/500\n",
      "16594/16594 [==============================] - 3s 196us/step - loss: 3767.0101\n",
      "Epoch 444/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3703.8974\n",
      "Epoch 445/500\n",
      "16594/16594 [==============================] - 3s 190us/step - loss: 3550.0561\n",
      "Epoch 446/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3524.3731\n",
      "Epoch 447/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3795.2873\n",
      "Epoch 448/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3849.5212\n",
      "Epoch 449/500\n",
      "16594/16594 [==============================] - 3s 194us/step - loss: 3601.9701\n",
      "Epoch 450/500\n",
      "16594/16594 [==============================] - 5s 300us/step - loss: 3705.6491\n",
      "Epoch 451/500\n",
      "16594/16594 [==============================] - 3s 197us/step - loss: 3516.8408\n",
      "Epoch 452/500\n",
      "16594/16594 [==============================] - 4s 221us/step - loss: 3533.0784\n",
      "Epoch 453/500\n",
      "16594/16594 [==============================] - 4s 270us/step - loss: 3537.3440\n",
      "Epoch 454/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3781.5850\n",
      "Epoch 455/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3743.5120\n",
      "Epoch 456/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3641.0505\n",
      "Epoch 457/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3760.6183\n",
      "Epoch 458/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3749.0842\n",
      "Epoch 459/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3487.2502\n",
      "Epoch 460/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3563.1971\n",
      "Epoch 461/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3551.4077\n",
      "Epoch 462/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3688.2158\n",
      "Epoch 463/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3874.7892\n",
      "Epoch 464/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3558.1861\n",
      "Epoch 465/500\n",
      "16594/16594 [==============================] - 4s 233us/step - loss: 3424.1635\n",
      "Epoch 466/500\n",
      "16594/16594 [==============================] - 3s 200us/step - loss: 3812.5226\n",
      "Epoch 467/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3572.3271\n",
      "Epoch 468/500\n",
      "16594/16594 [==============================] - 4s 236us/step - loss: 3651.2402\n",
      "Epoch 469/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3522.6541\n",
      "Epoch 470/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3676.6516\n",
      "Epoch 471/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3417.9249\n",
      "Epoch 472/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3814.7845\n",
      "Epoch 473/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3673.5905\n",
      "Epoch 474/500\n",
      "16594/16594 [==============================] - 4s 256us/step - loss: 3759.5836\n",
      "Epoch 475/500\n",
      "16594/16594 [==============================] - 3s 198us/step - loss: 3393.4304\n",
      "Epoch 476/500\n",
      "16594/16594 [==============================] - 3s 189us/step - loss: 3742.1571\n",
      "Epoch 477/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3524.4293\n",
      "Epoch 478/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3694.0372\n",
      "Epoch 479/500\n",
      "16594/16594 [==============================] - 3s 191us/step - loss: 3571.7741\n",
      "Epoch 480/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3517.6816\n",
      "Epoch 481/500\n",
      "16594/16594 [==============================] - 3s 179us/step - loss: 3810.3267\n",
      "Epoch 482/500\n",
      "16594/16594 [==============================] - 3s 181us/step - loss: 3425.9957\n",
      "Epoch 483/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3574.6520\n",
      "Epoch 484/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3739.0470\n",
      "Epoch 485/500\n",
      "16594/16594 [==============================] - 3s 185us/step - loss: 3474.0047\n",
      "Epoch 486/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3304.5031\n",
      "Epoch 487/500\n",
      "16594/16594 [==============================] - 3s 182us/step - loss: 3970.4484\n",
      "Epoch 488/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3803.8199\n",
      "Epoch 489/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3371.7415\n",
      "Epoch 490/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3616.0874\n",
      "Epoch 491/500\n",
      "16594/16594 [==============================] - 3s 183us/step - loss: 3691.4635\n",
      "Epoch 492/500\n",
      "16594/16594 [==============================] - 3s 186us/step - loss: 3469.8054\n",
      "Epoch 493/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16594/16594 [==============================] - 3s 207us/step - loss: 3522.0884\n",
      "Epoch 494/500\n",
      "16594/16594 [==============================] - 4s 264us/step - loss: 3692.1428\n",
      "Epoch 495/500\n",
      "16594/16594 [==============================] - 3s 192us/step - loss: 3892.2794\n",
      "Epoch 496/500\n",
      "16594/16594 [==============================] - 3s 180us/step - loss: 3281.9155\n",
      "Epoch 497/500\n",
      "16594/16594 [==============================] - 3s 188us/step - loss: 3555.3014\n",
      "Epoch 498/500\n",
      "16594/16594 [==============================] - 3s 184us/step - loss: 3574.9736\n",
      "Epoch 499/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3803.6076\n",
      "Epoch 500/500\n",
      "16594/16594 [==============================] - 3s 187us/step - loss: 3681.3791\n",
      "1843/1843 [==============================] - 0s 40us/step\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=10, shuffle=True, random_state=123)\n",
    "cvresults = []\n",
    "for train, test in kfold.split(X, y):\n",
    "    model.fit(X[train], y[train], epochs=500, batch_size=30, verbose=1)\n",
    "    results = model.evaluate(X[test], y[test], verbose=1)\n",
    "    cvresults.append(results)\n",
    "model.save(\"model_4_improved.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[52967.155758541216,\n",
       " 39487.5281487256,\n",
       " 37311.016870593819,\n",
       " 30657.2527284436,\n",
       " 26142.261900928686,\n",
       " 20825.912113611714,\n",
       " 15803.284963479528,\n",
       " 13087.993551402265,\n",
       " 11148.745220248491,\n",
       " 9834.8587304051653]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cvresults"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1844/1844 [==============================] - 0s 48us/step\n"
     ]
    }
   ],
   "source": [
    "results = model.evaluate(x_test, y_test, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3001.1819318990647"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE:  40.3979618461\n",
      "MSE:  3001.18191403\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "y_pred = model.predict(x_test)\n",
    "print(\"MAE: \", mean_absolute_error(y_test, y_pred))\n",
    "print(\"MSE: \", mean_squared_error(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAFkCAYAAADv13iSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngEmjFkIIIQqYNGohhBCigP0FaHxuurl/uEoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x19b848ae588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x19b849a7f28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = [i for i in range(len(y_test))]\n",
    "plt.scatter(x, y_test, label='test', alpha=1)\n",
    "plt.scatter(x, y_pred, label='test predictions', color='red', alpha=0.2)\n",
    "plt.legend(bbox_to_anchor=(0., 1.02, 1., .102), loc=3,\n",
    "           ncol=2, mode=\"expand\", borderaxespad=0.4)\n",
    "plt.figure(figsize=(16, 16), dpi=80)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}