20-autoencoder.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "os.environ[\"KERAS_BACKEND\"] = \"tensorflow\"\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Dense, Activation, BatchNormalization, Dropout, Input\n",
    "from IPython.display import SVG\n",
    "from keras.utils.vis_utils import model_to_dot\n",
    "import keras\n",
    "from datetime import datetime\n",
    "from sklearn import preprocessing\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from keras.utils import np_utils\n",
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "from keras.utils import plot_model\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.utils import shuffle\n",
    "import pickle\n",
    "from keras.utils import to_categorical"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load processed dataset\n",
    "The data we load here are already shuffled and normalized."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train: (966, 26)\n",
      "y_train: (966, 2)\n",
      "x_test: (242, 26)\n",
      "y_test: (242, 2)\n"
     ]
    }
   ],
   "source": [
    "data_directory = 'dataset/3. normalized/'\n",
    "x_train = pickle.load(open(data_directory + \"x_train.p\", \"rb\"))\n",
    "y_train = pickle.load(open(data_directory + \"y_train.p\", \"rb\"))\n",
    "x_test = pickle.load(open(data_directory + \"x_test.p\", \"rb\"))\n",
    "y_test = pickle.load(open(data_directory + \"y_test.p\", \"rb\"))\n",
    "print('x_train: ' + str(x_train.shape))\n",
    "print('y_train: ' + str(y_train.shape))\n",
    "print('x_test: ' + str(x_test.shape))\n",
    "print('y_test: ' + str(y_test.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.226135</td>\n",
       "      <td>0.165461</td>\n",
       "      <td>0.218373</td>\n",
       "      <td>0.136944</td>\n",
       "      <td>0.218364</td>\n",
       "      <td>0.326976</td>\n",
       "      <td>0.467346</td>\n",
       "      <td>0.250757</td>\n",
       "      <td>0.120083</td>\n",
       "      <td>0.321016</td>\n",
       "      <td>...</td>\n",
       "      <td>0.100836</td>\n",
       "      <td>0.338785</td>\n",
       "      <td>0.407917</td>\n",
       "      <td>0.155705</td>\n",
       "      <td>0.151108</td>\n",
       "      <td>0.181358</td>\n",
       "      <td>0.077172</td>\n",
       "      <td>0.484244</td>\n",
       "      <td>0.416667</td>\n",
       "      <td>0.675275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.079400</td>\n",
       "      <td>0.089118</td>\n",
       "      <td>0.061998</td>\n",
       "      <td>0.030564</td>\n",
       "      <td>0.062079</td>\n",
       "      <td>0.089922</td>\n",
       "      <td>0.191393</td>\n",
       "      <td>0.043280</td>\n",
       "      <td>0.024083</td>\n",
       "      <td>0.080073</td>\n",
       "      <td>...</td>\n",
       "      <td>0.071548</td>\n",
       "      <td>0.018426</td>\n",
       "      <td>0.228516</td>\n",
       "      <td>0.074497</td>\n",
       "      <td>0.071189</td>\n",
       "      <td>0.362160</td>\n",
       "      <td>0.183293</td>\n",
       "      <td>0.296468</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.077304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.103458</td>\n",
       "      <td>0.103848</td>\n",
       "      <td>0.100653</td>\n",
       "      <td>0.065950</td>\n",
       "      <td>0.100690</td>\n",
       "      <td>0.158392</td>\n",
       "      <td>0.283503</td>\n",
       "      <td>0.082511</td>\n",
       "      <td>0.046162</td>\n",
       "      <td>0.196872</td>\n",
       "      <td>...</td>\n",
       "      <td>0.033205</td>\n",
       "      <td>0.257567</td>\n",
       "      <td>0.234512</td>\n",
       "      <td>0.075839</td>\n",
       "      <td>0.075218</td>\n",
       "      <td>0.310095</td>\n",
       "      <td>0.049176</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.146945</td>\n",
       "      <td>0.220966</td>\n",
       "      <td>0.091491</td>\n",
       "      <td>0.064243</td>\n",
       "      <td>0.091571</td>\n",
       "      <td>0.334515</td>\n",
       "      <td>0.488109</td>\n",
       "      <td>0.191163</td>\n",
       "      <td>0.105050</td>\n",
       "      <td>0.330666</td>\n",
       "      <td>...</td>\n",
       "      <td>0.032132</td>\n",
       "      <td>0.115608</td>\n",
       "      <td>0.119201</td>\n",
       "      <td>0.020805</td>\n",
       "      <td>0.020148</td>\n",
       "      <td>0.553460</td>\n",
       "      <td>0.109024</td>\n",
       "      <td>0.291681</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.130953</td>\n",
       "      <td>0.136021</td>\n",
       "      <td>0.060241</td>\n",
       "      <td>0.029970</td>\n",
       "      <td>0.060280</td>\n",
       "      <td>0.179025</td>\n",
       "      <td>0.251038</td>\n",
       "      <td>0.107355</td>\n",
       "      <td>0.057036</td>\n",
       "      <td>0.181617</td>\n",
       "      <td>...</td>\n",
       "      <td>0.076168</td>\n",
       "      <td>0.191104</td>\n",
       "      <td>0.285241</td>\n",
       "      <td>0.037584</td>\n",
       "      <td>0.035594</td>\n",
       "      <td>0.328655</td>\n",
       "      <td>0.115067</td>\n",
       "      <td>0.408358</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.482370</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2         3         4         5         6   \\\n",
       "0  0.226135  0.165461  0.218373  0.136944  0.218364  0.326976  0.467346   \n",
       "1  0.079400  0.089118  0.061998  0.030564  0.062079  0.089922  0.191393   \n",
       "2  0.103458  0.103848  0.100653  0.065950  0.100690  0.158392  0.283503   \n",
       "3  0.146945  0.220966  0.091491  0.064243  0.091571  0.334515  0.488109   \n",
       "4  0.130953  0.136021  0.060241  0.029970  0.060280  0.179025  0.251038   \n",
       "\n",
       "         7         8         9     ...           16        17        18  \\\n",
       "0  0.250757  0.120083  0.321016    ...     0.100836  0.338785  0.407917   \n",
       "1  0.043280  0.024083  0.080073    ...     0.071548  0.018426  0.228516   \n",
       "2  0.082511  0.046162  0.196872    ...     0.033205  0.257567  0.234512   \n",
       "3  0.191163  0.105050  0.330666    ...     0.032132  0.115608  0.119201   \n",
       "4  0.107355  0.057036  0.181617    ...     0.076168  0.191104  0.285241   \n",
       "\n",
       "         19        20        21        22        23        24        25  \n",
       "0  0.155705  0.151108  0.181358  0.077172  0.484244  0.416667  0.675275  \n",
       "1  0.074497  0.071189  0.362160  0.183293  0.296468  0.083333  0.077304  \n",
       "2  0.075839  0.075218  0.310095  0.049176  0.000000  0.000000  0.000000  \n",
       "3  0.020805  0.020148  0.553460  0.109024  0.291681  0.000000  0.000000  \n",
       "4  0.037584  0.035594  0.328655  0.115067  0.408358  0.083333  0.482370  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(x_train).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.092726</td>\n",
       "      <td>0.130350</td>\n",
       "      <td>0.048820</td>\n",
       "      <td>0.025074</td>\n",
       "      <td>0.048818</td>\n",
       "      <td>0.206552</td>\n",
       "      <td>0.278973</td>\n",
       "      <td>0.123197</td>\n",
       "      <td>0.069596</td>\n",
       "      <td>0.157321</td>\n",
       "      <td>...</td>\n",
       "      <td>0.030981</td>\n",
       "      <td>0.126401</td>\n",
       "      <td>0.125133</td>\n",
       "      <td>0.033557</td>\n",
       "      <td>0.032908</td>\n",
       "      <td>0.496924</td>\n",
       "      <td>0.083581</td>\n",
       "      <td>0.357097</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.017255</td>\n",
       "      <td>0.026758</td>\n",
       "      <td>0.015688</td>\n",
       "      <td>0.010905</td>\n",
       "      <td>0.015687</td>\n",
       "      <td>0.108347</td>\n",
       "      <td>0.153643</td>\n",
       "      <td>0.081450</td>\n",
       "      <td>0.033064</td>\n",
       "      <td>0.096120</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003376</td>\n",
       "      <td>0.186343</td>\n",
       "      <td>0.123078</td>\n",
       "      <td>0.209396</td>\n",
       "      <td>0.208865</td>\n",
       "      <td>0.466729</td>\n",
       "      <td>0.007642</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.178509</td>\n",
       "      <td>0.286687</td>\n",
       "      <td>0.142696</td>\n",
       "      <td>0.088427</td>\n",
       "      <td>0.142732</td>\n",
       "      <td>0.279412</td>\n",
       "      <td>0.436769</td>\n",
       "      <td>0.188018</td>\n",
       "      <td>0.079033</td>\n",
       "      <td>0.236513</td>\n",
       "      <td>...</td>\n",
       "      <td>0.027032</td>\n",
       "      <td>0.099181</td>\n",
       "      <td>0.093156</td>\n",
       "      <td>0.045638</td>\n",
       "      <td>0.042982</td>\n",
       "      <td>0.606600</td>\n",
       "      <td>0.101034</td>\n",
       "      <td>0.203158</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.266331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.156274</td>\n",
       "      <td>0.177721</td>\n",
       "      <td>0.077686</td>\n",
       "      <td>0.058234</td>\n",
       "      <td>0.077682</td>\n",
       "      <td>0.160971</td>\n",
       "      <td>0.343903</td>\n",
       "      <td>0.068792</td>\n",
       "      <td>0.033838</td>\n",
       "      <td>0.099012</td>\n",
       "      <td>...</td>\n",
       "      <td>0.083206</td>\n",
       "      <td>0.102144</td>\n",
       "      <td>0.217839</td>\n",
       "      <td>0.032215</td>\n",
       "      <td>0.030893</td>\n",
       "      <td>0.377669</td>\n",
       "      <td>0.122733</td>\n",
       "      <td>0.614431</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.240163</td>\n",
       "      <td>0.460682</td>\n",
       "      <td>0.150351</td>\n",
       "      <td>0.120549</td>\n",
       "      <td>0.150387</td>\n",
       "      <td>0.305724</td>\n",
       "      <td>0.386183</td>\n",
       "      <td>0.190613</td>\n",
       "      <td>0.076670</td>\n",
       "      <td>0.253248</td>\n",
       "      <td>...</td>\n",
       "      <td>0.025370</td>\n",
       "      <td>0.029456</td>\n",
       "      <td>0.043714</td>\n",
       "      <td>0.011409</td>\n",
       "      <td>0.010745</td>\n",
       "      <td>0.765075</td>\n",
       "      <td>0.141341</td>\n",
       "      <td>0.567164</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2         3         4         5         6   \\\n",
       "0  0.092726  0.130350  0.048820  0.025074  0.048818  0.206552  0.278973   \n",
       "1  0.017255  0.026758  0.015688  0.010905  0.015687  0.108347  0.153643   \n",
       "2  0.178509  0.286687  0.142696  0.088427  0.142732  0.279412  0.436769   \n",
       "3  0.156274  0.177721  0.077686  0.058234  0.077682  0.160971  0.343903   \n",
       "4  0.240163  0.460682  0.150351  0.120549  0.150387  0.305724  0.386183   \n",
       "\n",
       "         7         8         9     ...           16        17        18  \\\n",
       "0  0.123197  0.069596  0.157321    ...     0.030981  0.126401  0.125133   \n",
       "1  0.081450  0.033064  0.096120    ...     0.003376  0.186343  0.123078   \n",
       "2  0.188018  0.079033  0.236513    ...     0.027032  0.099181  0.093156   \n",
       "3  0.068792  0.033838  0.099012    ...     0.083206  0.102144  0.217839   \n",
       "4  0.190613  0.076670  0.253248    ...     0.025370  0.029456  0.043714   \n",
       "\n",
       "         19        20        21        22        23        24        25  \n",
       "0  0.033557  0.032908  0.496924  0.083581  0.357097  0.000000  0.000000  \n",
       "1  0.209396  0.208865  0.466729  0.007642  0.000000  0.000000  0.000000  \n",
       "2  0.045638  0.042982  0.606600  0.101034  0.203158  0.166667  0.266331  \n",
       "3  0.032215  0.030893  0.377669  0.122733  0.614431  0.000000  0.000000  \n",
       "4  0.011409  0.010745  0.765075  0.141341  0.567164  0.000000  0.000000  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(x_test).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tunable parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs = 5000\n",
    "batch_size = 20\n",
    "learning_rate = 0.0001\n",
    "decay = 0.0001"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Autoencoder #1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "date = datetime.now().strftime(\"%Y-%m-%d-%H-%M-%S\")\n",
    "root_directory = 'results/results-autoencoder-01-02/'\n",
    "model_directory = root_directory + \"model/\"\n",
    "tensorboard_directory = root_directory + \"tensorboard/\"\n",
    "def create_model(layers, dropout):\n",
    "    global date\n",
    "    x = Input(shape=(x_train.shape[1],))\n",
    "\n",
    "    y = Dense(units = layers['layer_1'], activation='relu')(x)\n",
    "    y = Dropout(dropout)(y)\n",
    "\n",
    "    if (layers['layer_2'] > 0):\n",
    "        y = Dense(units = layers['layer_2'], activation='relu')(y)\n",
    "        y = Dropout(dropout)(y)\n",
    "    \n",
    "    y = Dense(units = x_train.shape[1], activation='sigmoid')(y)\n",
    "    model = Model(x, y)\n",
    "    \n",
    "    # Create directory\n",
    "    directory = model_directory + date + '/'\n",
    "    if not os.path.exists(directory):\n",
    "        os.makedirs(directory)\n",
    "    \n",
    "    # Write model hyper-parameters\n",
    "    df = pd.DataFrame({\n",
    "        'Layer 1': layers['layer_1'],\n",
    "        'Layer 2': layers['layer_2'],\n",
    "        'Dropout': dropout,\n",
    "        'Learning Rate': learning_rate,\n",
    "        'Decay': decay,\n",
    "        'Epochs': epochs,\n",
    "        'Batch size': batch_size\n",
    "    }, index=[0])\n",
    "    df.to_csv(model_directory + date + '/params.csv')\n",
    "    \n",
    "    # Write model summary\n",
    "    file2 = open(directory + \"summary.txt\", \"a\")\n",
    "    model.summary(print_fn=lambda line: file2.write(line + '\\n'))\n",
    "    file2.close()\n",
    "\n",
    "    # Write model diagram\n",
    "    plot_model(model, to_file=directory + 'model.png', show_shapes=True, show_layer_names=False)\n",
    "    \n",
    "    # Compile the model\n",
    "    model.compile(optimizer=keras.optimizers.Adam(lr=learning_rate, decay=decay), loss=keras.losses.categorical_crossentropy, metrics=['accuracy'])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepare callbacks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_callbacks():\n",
    "    date = datetime.now().strftime(\"%Y-%m-%d-%H-%M-%S\")\n",
    "    \n",
    "    directory = model_directory + date + '/'\n",
    "    if not os.path.exists(directory):\n",
    "        os.makedirs(directory)\n",
    "    \n",
    "    tensorboard_callback = keras.callbacks.TensorBoard(log_dir=tensorboard_directory + date + '/')\n",
    "    csv_logger = keras.callbacks.CSVLogger(model_directory + date + '/epochs.csv')\n",
    "    model_checkpoint = keras.callbacks.ModelCheckpoint(model_directory + date + '/weights.hdf5', save_best_only=True)\n",
    "    callbacks = [tensorboard_callback, csv_logger, model_checkpoint]\n",
    "    return callbacks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Grid search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 966 samples, validate on 242 samples\n",
      "Epoch 1/5000\n",
      "966/966 [==============================] - 0s - loss: 18.1016 - acc: 0.0383 - val_loss: 17.6237 - val_acc: 0.0165\n",
      "Epoch 2/5000\n",
      "966/966 [==============================] - 0s - loss: 18.0833 - acc: 0.0476 - val_loss: 17.6063 - val_acc: 0.0207\n",
      "Epoch 3/5000\n",
      "966/966 [==============================] - 0s - loss: 18.0608 - acc: 0.0487 - val_loss: 17.5894 - val_acc: 0.0248\n",
      "Epoch 4/5000\n",
      "966/966 [==============================] - 0s - loss: 18.0478 - acc: 0.0611 - val_loss: 17.5729 - val_acc: 0.0289\n",
      "Epoch 5/5000\n",
      "966/966 [==============================] - 0s - loss: 18.0269 - acc: 0.0745 - val_loss: 17.5560 - val_acc: 0.0413\n",
      "Epoch 6/5000\n",
      "966/966 [==============================] - 0s - loss: 18.0120 - acc: 0.0725 - val_loss: 17.5378 - val_acc: 0.0413\n",
      "Epoch 7/5000\n",
      "966/966 [==============================] - 0s - loss: 17.9939 - acc: 0.0839 - val_loss: 17.5187 - val_acc: 0.0372\n",
      "Epoch 8/5000\n",
      "966/966 [==============================] - 0s - loss: 17.9759 - acc: 0.1014 - val_loss: 17.4986 - val_acc: 0.0372\n",
      "Epoch 9/5000\n",
      "966/966 [==============================] - 0s - loss: 17.9562 - acc: 0.1418 - val_loss: 17.4770 - val_acc: 0.0537\n",
      "Epoch 10/5000\n",
      "966/966 [==============================] - 0s - loss: 17.9324 - acc: 0.1294 - val_loss: 17.4544 - val_acc: 0.0579\n",
      "Epoch 11/5000\n",
      "966/966 [==============================] - 0s - loss: 17.9108 - acc: 0.1522 - val_loss: 17.4309 - val_acc: 0.0661\n",
      "Epoch 12/5000\n",
      "966/966 [==============================] - 0s - loss: 17.8936 - acc: 0.1584 - val_loss: 17.4073 - val_acc: 0.0826\n",
      "Epoch 13/5000\n",
      "966/966 [==============================] - 0s - loss: 17.8673 - acc: 0.1573 - val_loss: 17.3825 - val_acc: 0.0950\n",
      "Epoch 14/5000\n",
      "966/966 [==============================] - 0s - loss: 17.8519 - acc: 0.1749 - val_loss: 17.3575 - val_acc: 0.1074\n",
      "Epoch 15/5000\n",
      "966/966 [==============================] - 0s - loss: 17.8241 - acc: 0.1863 - val_loss: 17.3324 - val_acc: 0.1240\n",
      "Epoch 16/5000\n",
      "966/966 [==============================] - 0s - loss: 17.8083 - acc: 0.1894 - val_loss: 17.3074 - val_acc: 0.1488\n",
      "Epoch 17/5000\n",
      "966/966 [==============================] - 0s - loss: 17.7868 - acc: 0.2008 - val_loss: 17.2834 - val_acc: 0.1529\n",
      "Epoch 18/5000\n",
      "966/966 [==============================] - 0s - loss: 17.7653 - acc: 0.2133 - val_loss: 17.2589 - val_acc: 0.1653\n",
      "Epoch 19/5000\n",
      "966/966 [==============================] - 0s - loss: 17.7487 - acc: 0.2246 - val_loss: 17.2350 - val_acc: 0.1777\n",
      "Epoch 20/5000\n",
      "966/966 [==============================] - 0s - loss: 17.7188 - acc: 0.2567 - val_loss: 17.2108 - val_acc: 0.1983\n",
      "Epoch 21/5000\n",
      "966/966 [==============================] - 0s - loss: 17.7094 - acc: 0.2692 - val_loss: 17.1875 - val_acc: 0.2314\n",
      "Epoch 22/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6906 - acc: 0.2629 - val_loss: 17.1653 - val_acc: 0.2438\n",
      "Epoch 23/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6752 - acc: 0.2712 - val_loss: 17.1433 - val_acc: 0.2686\n",
      "Epoch 24/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6581 - acc: 0.2992 - val_loss: 17.1220 - val_acc: 0.2934\n",
      "Epoch 25/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6325 - acc: 0.3199 - val_loss: 17.1013 - val_acc: 0.3099\n",
      "Epoch 26/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6244 - acc: 0.3106 - val_loss: 17.0816 - val_acc: 0.3388\n",
      "Epoch 27/5000\n",
      "966/966 [==============================] - 0s - loss: 17.6066 - acc: 0.3313 - val_loss: 17.0627 - val_acc: 0.3719\n",
      "Epoch 28/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5877 - acc: 0.3344 - val_loss: 17.0440 - val_acc: 0.3843\n",
      "Epoch 29/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5788 - acc: 0.3095 - val_loss: 17.0259 - val_acc: 0.4050\n",
      "Epoch 30/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5493 - acc: 0.3178 - val_loss: 17.0078 - val_acc: 0.4298\n",
      "Epoch 31/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5529 - acc: 0.3261 - val_loss: 16.9902 - val_acc: 0.4711\n",
      "Epoch 32/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5324 - acc: 0.3458 - val_loss: 16.9737 - val_acc: 0.5000\n",
      "Epoch 33/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5240 - acc: 0.3385 - val_loss: 16.9577 - val_acc: 0.5372\n",
      "Epoch 34/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5105 - acc: 0.3489 - val_loss: 16.9419 - val_acc: 0.5785\n",
      "Epoch 35/5000\n",
      "966/966 [==============================] - 0s - loss: 17.5038 - acc: 0.3841 - val_loss: 16.9275 - val_acc: 0.6198\n",
      "Epoch 36/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4855 - acc: 0.3654 - val_loss: 16.9141 - val_acc: 0.6446\n",
      "Epoch 37/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4731 - acc: 0.3737 - val_loss: 16.9002 - val_acc: 0.6570\n",
      "Epoch 38/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4528 - acc: 0.4317 - val_loss: 16.8874 - val_acc: 0.6612\n",
      "Epoch 39/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4497 - acc: 0.4224 - val_loss: 16.8743 - val_acc: 0.6612\n",
      "Epoch 40/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4426 - acc: 0.4068 - val_loss: 16.8621 - val_acc: 0.6612\n",
      "Epoch 41/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4428 - acc: 0.3954 - val_loss: 16.8498 - val_acc: 0.6612\n",
      "Epoch 42/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4298 - acc: 0.4089 - val_loss: 16.8380 - val_acc: 0.6612\n",
      "Epoch 43/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4260 - acc: 0.3965 - val_loss: 16.8269 - val_acc: 0.6653\n",
      "Epoch 44/5000\n",
      "966/966 [==============================] - 0s - loss: 17.4038 - acc: 0.3882 - val_loss: 16.8162 - val_acc: 0.6653\n",
      "Epoch 45/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3949 - acc: 0.4234 - val_loss: 16.8051 - val_acc: 0.6653\n",
      "Epoch 46/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3874 - acc: 0.4296 - val_loss: 16.7946 - val_acc: 0.6653\n",
      "Epoch 47/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3809 - acc: 0.4120 - val_loss: 16.7839 - val_acc: 0.6653\n",
      "Epoch 48/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3620 - acc: 0.4275 - val_loss: 16.7737 - val_acc: 0.6653\n",
      "Epoch 49/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3674 - acc: 0.4182 - val_loss: 16.7636 - val_acc: 0.6653\n",
      "Epoch 50/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3589 - acc: 0.4286 - val_loss: 16.7545 - val_acc: 0.6653\n",
      "Epoch 51/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3507 - acc: 0.4348 - val_loss: 16.7449 - val_acc: 0.6653\n",
      "Epoch 52/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3382 - acc: 0.4348 - val_loss: 16.7354 - val_acc: 0.6653\n",
      "Epoch 53/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3315 - acc: 0.4193 - val_loss: 16.7261 - val_acc: 0.6653\n",
      "Epoch 54/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3301 - acc: 0.4689 - val_loss: 16.7181 - val_acc: 0.6653\n",
      "Epoch 55/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3149 - acc: 0.4503 - val_loss: 16.7097 - val_acc: 0.6653\n",
      "Epoch 56/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3178 - acc: 0.4389 - val_loss: 16.7005 - val_acc: 0.6653\n",
      "Epoch 57/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3259 - acc: 0.4358 - val_loss: 16.6928 - val_acc: 0.6653\n",
      "Epoch 58/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2983 - acc: 0.4234 - val_loss: 16.6848 - val_acc: 0.6653\n",
      "Epoch 59/5000\n",
      "966/966 [==============================] - 0s - loss: 17.3086 - acc: 0.4193 - val_loss: 16.6771 - val_acc: 0.6653\n",
      "Epoch 60/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2832 - acc: 0.4482 - val_loss: 16.6690 - val_acc: 0.6653\n",
      "Epoch 61/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 17.1583 - acc: 0.45 - 0s - loss: 17.2781 - acc: 0.4472 - val_loss: 16.6613 - val_acc: 0.6653\n",
      "Epoch 62/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2621 - acc: 0.4410 - val_loss: 16.6539 - val_acc: 0.6653\n",
      "Epoch 63/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2649 - acc: 0.4410 - val_loss: 16.6469 - val_acc: 0.6653\n",
      "Epoch 64/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 17.2645 - acc: 0.4389 - val_loss: 16.6395 - val_acc: 0.6653\n",
      "Epoch 65/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2499 - acc: 0.4648 - val_loss: 16.6321 - val_acc: 0.6653\n",
      "Epoch 66/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2475 - acc: 0.4586 - val_loss: 16.6253 - val_acc: 0.6653\n",
      "Epoch 67/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2533 - acc: 0.4431 - val_loss: 16.6189 - val_acc: 0.6653\n",
      "Epoch 68/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2399 - acc: 0.4555 - val_loss: 16.6119 - val_acc: 0.6653\n",
      "Epoch 69/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2205 - acc: 0.4462 - val_loss: 16.6048 - val_acc: 0.6653\n",
      "Epoch 70/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2106 - acc: 0.4317 - val_loss: 16.5979 - val_acc: 0.6653\n",
      "Epoch 71/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2103 - acc: 0.4576 - val_loss: 16.5917 - val_acc: 0.6653\n",
      "Epoch 72/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2007 - acc: 0.4783 - val_loss: 16.5845 - val_acc: 0.6653\n",
      "Epoch 73/5000\n",
      "966/966 [==============================] - 0s - loss: 17.2175 - acc: 0.4627 - val_loss: 16.5790 - val_acc: 0.6653\n",
      "Epoch 74/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1991 - acc: 0.4617 - val_loss: 16.5733 - val_acc: 0.6653\n",
      "Epoch 75/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1981 - acc: 0.4596 - val_loss: 16.5678 - val_acc: 0.6653\n",
      "Epoch 76/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1832 - acc: 0.4607 - val_loss: 16.5618 - val_acc: 0.6653\n",
      "Epoch 77/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1725 - acc: 0.4824 - val_loss: 16.5557 - val_acc: 0.6653\n",
      "Epoch 78/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1809 - acc: 0.4658 - val_loss: 16.5495 - val_acc: 0.6653\n",
      "Epoch 79/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1691 - acc: 0.4793 - val_loss: 16.5438 - val_acc: 0.6653\n",
      "Epoch 80/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1623 - acc: 0.4720 - val_loss: 16.5383 - val_acc: 0.6653\n",
      "Epoch 81/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1621 - acc: 0.4545 - val_loss: 16.5322 - val_acc: 0.6653\n",
      "Epoch 82/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1671 - acc: 0.4741 - val_loss: 16.5264 - val_acc: 0.6653\n",
      "Epoch 83/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1455 - acc: 0.5031 - val_loss: 16.5211 - val_acc: 0.6653\n",
      "Epoch 84/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1488 - acc: 0.4762 - val_loss: 16.5156 - val_acc: 0.6653\n",
      "Epoch 85/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1556 - acc: 0.4969 - val_loss: 16.5098 - val_acc: 0.6653\n",
      "Epoch 86/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1237 - acc: 0.4907 - val_loss: 16.5047 - val_acc: 0.6653\n",
      "Epoch 87/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1408 - acc: 0.4938 - val_loss: 16.4995 - val_acc: 0.6653\n",
      "Epoch 88/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1275 - acc: 0.4990 - val_loss: 16.4943 - val_acc: 0.6653\n",
      "Epoch 89/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1221 - acc: 0.5083 - val_loss: 16.4890 - val_acc: 0.6653\n",
      "Epoch 90/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1201 - acc: 0.5114 - val_loss: 16.4843 - val_acc: 0.6653\n",
      "Epoch 91/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1147 - acc: 0.5083 - val_loss: 16.4800 - val_acc: 0.6653\n",
      "Epoch 92/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1272 - acc: 0.5031 - val_loss: 16.4753 - val_acc: 0.6653\n",
      "Epoch 93/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1076 - acc: 0.4803 - val_loss: 16.4709 - val_acc: 0.6653\n",
      "Epoch 94/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1070 - acc: 0.5010 - val_loss: 16.4660 - val_acc: 0.6653\n",
      "Epoch 95/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1074 - acc: 0.4917 - val_loss: 16.4609 - val_acc: 0.6653\n",
      "Epoch 96/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1026 - acc: 0.5228 - val_loss: 16.4565 - val_acc: 0.6653\n",
      "Epoch 97/5000\n",
      "966/966 [==============================] - 0s - loss: 17.1018 - acc: 0.5176 - val_loss: 16.4524 - val_acc: 0.6653\n",
      "Epoch 98/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0950 - acc: 0.5031 - val_loss: 16.4482 - val_acc: 0.6653\n",
      "Epoch 99/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0938 - acc: 0.4886 - val_loss: 16.4435 - val_acc: 0.6653\n",
      "Epoch 100/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0689 - acc: 0.5217 - val_loss: 16.4390 - val_acc: 0.6653\n",
      "Epoch 101/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0839 - acc: 0.5104 - val_loss: 16.4347 - val_acc: 0.6653\n",
      "Epoch 102/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0748 - acc: 0.5186 - val_loss: 16.4304 - val_acc: 0.6653\n",
      "Epoch 103/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0779 - acc: 0.5342 - val_loss: 16.4265 - val_acc: 0.6653\n",
      "Epoch 104/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0716 - acc: 0.5259 - val_loss: 16.4220 - val_acc: 0.6653\n",
      "Epoch 105/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0723 - acc: 0.5300 - val_loss: 16.4176 - val_acc: 0.6653\n",
      "Epoch 106/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0671 - acc: 0.5259 - val_loss: 16.4135 - val_acc: 0.6653\n",
      "Epoch 107/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0692 - acc: 0.5280 - val_loss: 16.4094 - val_acc: 0.6653\n",
      "Epoch 108/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0527 - acc: 0.5238 - val_loss: 16.4055 - val_acc: 0.6653\n",
      "Epoch 109/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0676 - acc: 0.5352 - val_loss: 16.4018 - val_acc: 0.6653\n",
      "Epoch 110/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0590 - acc: 0.5166 - val_loss: 16.3983 - val_acc: 0.6653\n",
      "Epoch 111/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0418 - acc: 0.5445 - val_loss: 16.3943 - val_acc: 0.6653\n",
      "Epoch 112/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0427 - acc: 0.5383 - val_loss: 16.3905 - val_acc: 0.6653\n",
      "Epoch 113/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0499 - acc: 0.5383 - val_loss: 16.3865 - val_acc: 0.6653\n",
      "Epoch 114/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0432 - acc: 0.5393 - val_loss: 16.3832 - val_acc: 0.6653\n",
      "Epoch 115/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0345 - acc: 0.5331 - val_loss: 16.3794 - val_acc: 0.6653\n",
      "Epoch 116/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0343 - acc: 0.5466 - val_loss: 16.3757 - val_acc: 0.6653\n",
      "Epoch 117/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0349 - acc: 0.5621 - val_loss: 16.3718 - val_acc: 0.6653\n",
      "Epoch 118/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0254 - acc: 0.5507 - val_loss: 16.3680 - val_acc: 0.6653\n",
      "Epoch 119/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0221 - acc: 0.5549 - val_loss: 16.3642 - val_acc: 0.6653\n",
      "Epoch 120/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0279 - acc: 0.5342 - val_loss: 16.3610 - val_acc: 0.6653\n",
      "Epoch 121/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0146 - acc: 0.5538 - val_loss: 16.3575 - val_acc: 0.6653\n",
      "Epoch 122/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0220 - acc: 0.5445 - val_loss: 16.3540 - val_acc: 0.6653\n",
      "Epoch 123/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0294 - acc: 0.5497 - val_loss: 16.3501 - val_acc: 0.6653\n",
      "Epoch 124/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0110 - acc: 0.5683 - val_loss: 16.3469 - val_acc: 0.6653\n",
      "Epoch 125/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0022 - acc: 0.5559 - val_loss: 16.3436 - val_acc: 0.6653\n",
      "Epoch 126/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0059 - acc: 0.5714 - val_loss: 16.3396 - val_acc: 0.6653\n",
      "Epoch 127/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 17.0095 - acc: 0.5559 - val_loss: 16.3363 - val_acc: 0.6653\n",
      "Epoch 128/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9926 - acc: 0.5518 - val_loss: 16.3325 - val_acc: 0.6653\n",
      "Epoch 129/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9908 - acc: 0.5600 - val_loss: 16.3290 - val_acc: 0.6653\n",
      "Epoch 130/5000\n",
      "966/966 [==============================] - 0s - loss: 17.0009 - acc: 0.5797 - val_loss: 16.3254 - val_acc: 0.6653\n",
      "Epoch 131/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9951 - acc: 0.5600 - val_loss: 16.3222 - val_acc: 0.6653\n",
      "Epoch 132/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9806 - acc: 0.5735 - val_loss: 16.3183 - val_acc: 0.6653\n",
      "Epoch 133/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9751 - acc: 0.5673 - val_loss: 16.3147 - val_acc: 0.6653\n",
      "Epoch 134/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9740 - acc: 0.5787 - val_loss: 16.3116 - val_acc: 0.6653\n",
      "Epoch 135/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9891 - acc: 0.5745 - val_loss: 16.3083 - val_acc: 0.6653\n",
      "Epoch 136/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9756 - acc: 0.5942 - val_loss: 16.3049 - val_acc: 0.6653\n",
      "Epoch 137/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9779 - acc: 0.5859 - val_loss: 16.3016 - val_acc: 0.6653\n",
      "Epoch 138/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9758 - acc: 0.5714 - val_loss: 16.2984 - val_acc: 0.6653\n",
      "Epoch 139/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9675 - acc: 0.5942 - val_loss: 16.2952 - val_acc: 0.6653\n",
      "Epoch 140/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9571 - acc: 0.5797 - val_loss: 16.2921 - val_acc: 0.6653\n",
      "Epoch 141/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9815 - acc: 0.5911 - val_loss: 16.2883 - val_acc: 0.6653\n",
      "Epoch 142/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9543 - acc: 0.6025 - val_loss: 16.2856 - val_acc: 0.6653\n",
      "Epoch 143/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9514 - acc: 0.5859 - val_loss: 16.2831 - val_acc: 0.6653\n",
      "Epoch 144/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9353 - acc: 0.5973 - val_loss: 16.2806 - val_acc: 0.6653\n",
      "Epoch 145/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9353 - acc: 0.5849 - val_loss: 16.2777 - val_acc: 0.6653\n",
      "Epoch 146/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9306 - acc: 0.5766 - val_loss: 16.2742 - val_acc: 0.6653\n",
      "Epoch 147/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9481 - acc: 0.6035 - val_loss: 16.2714 - val_acc: 0.6653\n",
      "Epoch 148/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9461 - acc: 0.5963 - val_loss: 16.2686 - val_acc: 0.6653\n",
      "Epoch 149/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9442 - acc: 0.6056 - val_loss: 16.2660 - val_acc: 0.6653\n",
      "Epoch 150/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9289 - acc: 0.6004 - val_loss: 16.2634 - val_acc: 0.6653\n",
      "Epoch 151/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9296 - acc: 0.5963 - val_loss: 16.2606 - val_acc: 0.6653.\n",
      "Epoch 152/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9189 - acc: 0.5932 - val_loss: 16.2580 - val_acc: 0.6653\n",
      "Epoch 153/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9456 - acc: 0.6025 - val_loss: 16.2549 - val_acc: 0.6653\n",
      "Epoch 154/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9196 - acc: 0.5735 - val_loss: 16.2520 - val_acc: 0.6653\n",
      "Epoch 155/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9289 - acc: 0.5973 - val_loss: 16.2493 - val_acc: 0.6653\n",
      "Epoch 156/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9193 - acc: 0.6056 - val_loss: 16.2467 - val_acc: 0.6653\n",
      "Epoch 157/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9240 - acc: 0.6025 - val_loss: 16.2448 - val_acc: 0.6653\n",
      "Epoch 158/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9295 - acc: 0.5932 - val_loss: 16.2426 - val_acc: 0.6653\n",
      "Epoch 159/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9140 - acc: 0.6046 - val_loss: 16.2400 - val_acc: 0.6653\n",
      "Epoch 160/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9240 - acc: 0.5921 - val_loss: 16.2379 - val_acc: 0.6653\n",
      "Epoch 161/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9266 - acc: 0.6035 - val_loss: 16.2353 - val_acc: 0.6653\n",
      "Epoch 162/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9138 - acc: 0.6190 - val_loss: 16.2333 - val_acc: 0.6653\n",
      "Epoch 163/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9056 - acc: 0.6128 - val_loss: 16.2306 - val_acc: 0.6653\n",
      "Epoch 164/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9059 - acc: 0.6211 - val_loss: 16.2279 - val_acc: 0.6653\n",
      "Epoch 165/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9085 - acc: 0.6139 - val_loss: 16.2252 - val_acc: 0.6653\n",
      "Epoch 166/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9136 - acc: 0.6108 - val_loss: 16.2224 - val_acc: 0.6653\n",
      "Epoch 167/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8945 - acc: 0.6066 - val_loss: 16.2203 - val_acc: 0.6653\n",
      "Epoch 168/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9074 - acc: 0.6159 - val_loss: 16.2182 - val_acc: 0.6653\n",
      "Epoch 169/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8936 - acc: 0.6108 - val_loss: 16.2154 - val_acc: 0.6653\n",
      "Epoch 170/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9178 - acc: 0.6066 - val_loss: 16.2127 - val_acc: 0.6653\n",
      "Epoch 171/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9009 - acc: 0.6201 - val_loss: 16.2110 - val_acc: 0.6653\n",
      "Epoch 172/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9017 - acc: 0.6190 - val_loss: 16.2088 - val_acc: 0.6653\n",
      "Epoch 173/5000\n",
      "966/966 [==============================] - 0s - loss: 16.9069 - acc: 0.6108 - val_loss: 16.2060 - val_acc: 0.6653\n",
      "Epoch 174/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8881 - acc: 0.6201 - val_loss: 16.2039 - val_acc: 0.6653\n",
      "Epoch 175/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8950 - acc: 0.6284 - val_loss: 16.2019 - val_acc: 0.6653\n",
      "Epoch 176/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8727 - acc: 0.6294 - val_loss: 16.1998 - val_acc: 0.6653\n",
      "Epoch 177/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8834 - acc: 0.6211 - val_loss: 16.1982 - val_acc: 0.6653\n",
      "Epoch 178/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8854 - acc: 0.6304 - val_loss: 16.1963 - val_acc: 0.6653\n",
      "Epoch 179/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8764 - acc: 0.6232 - val_loss: 16.1936 - val_acc: 0.6653\n",
      "Epoch 180/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8706 - acc: 0.6273 - val_loss: 16.1917 - val_acc: 0.6653\n",
      "Epoch 181/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8668 - acc: 0.6180 - val_loss: 16.1895 - val_acc: 0.6653\n",
      "Epoch 182/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8799 - acc: 0.6273 - val_loss: 16.1875 - val_acc: 0.6653\n",
      "Epoch 183/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8712 - acc: 0.6222 - val_loss: 16.1857 - val_acc: 0.6653\n",
      "Epoch 184/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8682 - acc: 0.6263 - val_loss: 16.1836 - val_acc: 0.6653\n",
      "Epoch 185/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8777 - acc: 0.6304 - val_loss: 16.1815 - val_acc: 0.6653\n",
      "Epoch 186/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8517 - acc: 0.6356 - val_loss: 16.1798 - val_acc: 0.6653\n",
      "Epoch 187/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8632 - acc: 0.6377 - val_loss: 16.1779 - val_acc: 0.6653\n",
      "Epoch 188/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8388 - acc: 0.6315 - val_loss: 16.1758 - val_acc: 0.6653\n",
      "Epoch 189/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8665 - acc: 0.6398 - val_loss: 16.1740 - val_acc: 0.6653\n",
      "Epoch 190/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.8587 - acc: 0.6304 - val_loss: 16.1724 - val_acc: 0.6653\n",
      "Epoch 191/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8613 - acc: 0.6284 - val_loss: 16.1704 - val_acc: 0.6653\n",
      "Epoch 192/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8623 - acc: 0.6325 - val_loss: 16.1684 - val_acc: 0.6653\n",
      "Epoch 193/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8755 - acc: 0.6346 - val_loss: 16.1668 - val_acc: 0.6653\n",
      "Epoch 194/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8596 - acc: 0.6449 - val_loss: 16.1655 - val_acc: 0.6653\n",
      "Epoch 195/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8389 - acc: 0.6398 - val_loss: 16.1635 - val_acc: 0.6653\n",
      "Epoch 196/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8569 - acc: 0.6356 - val_loss: 16.1615 - val_acc: 0.6653\n",
      "Epoch 197/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8535 - acc: 0.6356 - val_loss: 16.1598 - val_acc: 0.6653\n",
      "Epoch 198/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8429 - acc: 0.6449 - val_loss: 16.1576 - val_acc: 0.6653\n",
      "Epoch 199/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8561 - acc: 0.6418 - val_loss: 16.1557 - val_acc: 0.6653\n",
      "Epoch 200/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8455 - acc: 0.6439 - val_loss: 16.1542 - val_acc: 0.6653\n",
      "Epoch 201/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8223 - acc: 0.6304 - val_loss: 16.1521 - val_acc: 0.6653\n",
      "Epoch 202/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8319 - acc: 0.6449 - val_loss: 16.1503 - val_acc: 0.6653\n",
      "Epoch 203/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8539 - acc: 0.6480 - val_loss: 16.1481 - val_acc: 0.6653\n",
      "Epoch 204/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8457 - acc: 0.6439 - val_loss: 16.1463 - val_acc: 0.6653\n",
      "Epoch 205/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8446 - acc: 0.6418 - val_loss: 16.1448 - val_acc: 0.6653\n",
      "Epoch 206/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8404 - acc: 0.6491 - val_loss: 16.1432 - val_acc: 0.6653\n",
      "Epoch 207/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8265 - acc: 0.6408 - val_loss: 16.1414 - val_acc: 0.6653\n",
      "Epoch 208/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8317 - acc: 0.6501 - val_loss: 16.1394 - val_acc: 0.6653\n",
      "Epoch 209/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8414 - acc: 0.6449 - val_loss: 16.1378 - val_acc: 0.6653\n",
      "Epoch 210/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8371 - acc: 0.6522 - val_loss: 16.1368 - val_acc: 0.6653\n",
      "Epoch 211/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8489 - acc: 0.6418 - val_loss: 16.1351 - val_acc: 0.6653\n",
      "Epoch 212/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8314 - acc: 0.6542 - val_loss: 16.1329 - val_acc: 0.6653\n",
      "Epoch 213/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.9445 - acc: 0.63 - 0s - loss: 16.8232 - acc: 0.6563 - val_loss: 16.1311 - val_acc: 0.6653\n",
      "Epoch 214/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8296 - acc: 0.6584 - val_loss: 16.1300 - val_acc: 0.6653\n",
      "Epoch 215/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8054 - acc: 0.6553 - val_loss: 16.1285 - val_acc: 0.6653\n",
      "Epoch 216/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8196 - acc: 0.6522 - val_loss: 16.1269 - val_acc: 0.6653\n",
      "Epoch 217/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8123 - acc: 0.6511 - val_loss: 16.1253 - val_acc: 0.6653\n",
      "Epoch 218/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8390 - acc: 0.6542 - val_loss: 16.1242 - val_acc: 0.6653\n",
      "Epoch 219/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8257 - acc: 0.6460 - val_loss: 16.1225 - val_acc: 0.6653\n",
      "Epoch 220/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8331 - acc: 0.6532 - val_loss: 16.1209 - val_acc: 0.6653\n",
      "Epoch 221/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8120 - acc: 0.6625 - val_loss: 16.1195 - val_acc: 0.6653\n",
      "Epoch 222/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8114 - acc: 0.6522 - val_loss: 16.1182 - val_acc: 0.6653\n",
      "Epoch 223/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8071 - acc: 0.6511 - val_loss: 16.1164 - val_acc: 0.6653\n",
      "Epoch 224/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8021 - acc: 0.6522 - val_loss: 16.1152 - val_acc: 0.6653\n",
      "Epoch 225/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8207 - acc: 0.6532 - val_loss: 16.1134 - val_acc: 0.6653\n",
      "Epoch 226/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8115 - acc: 0.6532 - val_loss: 16.1116 - val_acc: 0.6653\n",
      "Epoch 227/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8029 - acc: 0.6542 - val_loss: 16.1100 - val_acc: 0.6653\n",
      "Epoch 228/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8239 - acc: 0.6532 - val_loss: 16.1083 - val_acc: 0.6653\n",
      "Epoch 229/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8062 - acc: 0.6511 - val_loss: 16.1071 - val_acc: 0.6653\n",
      "Epoch 230/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8002 - acc: 0.6553 - val_loss: 16.1057 - val_acc: 0.6653\n",
      "Epoch 231/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8169 - acc: 0.6532 - val_loss: 16.1045 - val_acc: 0.6653\n",
      "Epoch 232/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8111 - acc: 0.6449 - val_loss: 16.1032 - val_acc: 0.6653\n",
      "Epoch 233/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7986 - acc: 0.6573 - val_loss: 16.1016 - val_acc: 0.6653\n",
      "Epoch 234/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8010 - acc: 0.6573 - val_loss: 16.1000 - val_acc: 0.6653\n",
      "Epoch 235/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7888 - acc: 0.6563 - val_loss: 16.0989 - val_acc: 0.6653\n",
      "Epoch 236/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7934 - acc: 0.6615 - val_loss: 16.0979 - val_acc: 0.6653\n",
      "Epoch 237/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7822 - acc: 0.6584 - val_loss: 16.0966 - val_acc: 0.6653\n",
      "Epoch 238/5000\n",
      "966/966 [==============================] - 0s - loss: 16.8120 - acc: 0.6594 - val_loss: 16.0950 - val_acc: 0.6653\n",
      "Epoch 239/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7945 - acc: 0.6584 - val_loss: 16.0932 - val_acc: 0.6653\n",
      "Epoch 240/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7952 - acc: 0.6563 - val_loss: 16.0925 - val_acc: 0.6653\n",
      "Epoch 241/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7902 - acc: 0.6656 - val_loss: 16.0916 - val_acc: 0.6653\n",
      "Epoch 242/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7919 - acc: 0.6573 - val_loss: 16.0904 - val_acc: 0.6653\n",
      "Epoch 243/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7853 - acc: 0.6625 - val_loss: 16.0890 - val_acc: 0.6653\n",
      "Epoch 244/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7903 - acc: 0.6625 - val_loss: 16.0876 - val_acc: 0.6653\n",
      "Epoch 245/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7905 - acc: 0.6656 - val_loss: 16.0867 - val_acc: 0.6653\n",
      "Epoch 246/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7956 - acc: 0.6656 - val_loss: 16.0855 - val_acc: 0.6653\n",
      "Epoch 247/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7777 - acc: 0.6563 - val_loss: 16.0845 - val_acc: 0.6653\n",
      "Epoch 248/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7762 - acc: 0.6646 - val_loss: 16.0828 - val_acc: 0.6653\n",
      "Epoch 249/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7657 - acc: 0.6625 - val_loss: 16.0812 - val_acc: 0.6653\n",
      "Epoch 250/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7739 - acc: 0.6563 - val_loss: 16.0800 - val_acc: 0.6653\n",
      "Epoch 251/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7623 - acc: 0.6656 - val_loss: 16.0791 - val_acc: 0.6653\n",
      "Epoch 252/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7706 - acc: 0.6625 - val_loss: 16.0780 - val_acc: 0.6653\n",
      "Epoch 253/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.7961 - acc: 0.6625 - val_loss: 16.0766 - val_acc: 0.6653\n",
      "Epoch 254/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7895 - acc: 0.6698 - val_loss: 16.0753 - val_acc: 0.6653\n",
      "Epoch 255/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7698 - acc: 0.6615 - val_loss: 16.0740 - val_acc: 0.6653\n",
      "Epoch 256/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7808 - acc: 0.6687 - val_loss: 16.0726 - val_acc: 0.6653\n",
      "Epoch 257/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7820 - acc: 0.6636 - val_loss: 16.0715 - val_acc: 0.6653\n",
      "Epoch 258/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7826 - acc: 0.6615 - val_loss: 16.0705 - val_acc: 0.6653\n",
      "Epoch 259/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7806 - acc: 0.6615 - val_loss: 16.0694 - val_acc: 0.6653\n",
      "Epoch 260/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7901 - acc: 0.6636 - val_loss: 16.0679 - val_acc: 0.6653\n",
      "Epoch 261/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7756 - acc: 0.6584 - val_loss: 16.0672 - val_acc: 0.6653\n",
      "Epoch 262/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7606 - acc: 0.6625 - val_loss: 16.0658 - val_acc: 0.6653\n",
      "Epoch 263/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7820 - acc: 0.6656 - val_loss: 16.0645 - val_acc: 0.6653\n",
      "Epoch 264/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7793 - acc: 0.6636 - val_loss: 16.0633 - val_acc: 0.6653\n",
      "Epoch 265/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7628 - acc: 0.6625 - val_loss: 16.0620 - val_acc: 0.6653\n",
      "Epoch 266/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7652 - acc: 0.6656 - val_loss: 16.0608 - val_acc: 0.6653\n",
      "Epoch 267/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7560 - acc: 0.6646 - val_loss: 16.0595 - val_acc: 0.6653\n",
      "Epoch 268/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7646 - acc: 0.6667 - val_loss: 16.0585 - val_acc: 0.6653\n",
      "Epoch 269/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7670 - acc: 0.6605 - val_loss: 16.0577 - val_acc: 0.6653\n",
      "Epoch 270/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7677 - acc: 0.6646 - val_loss: 16.0565 - val_acc: 0.6653\n",
      "Epoch 271/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7615 - acc: 0.6646 - val_loss: 16.0552 - val_acc: 0.6653\n",
      "Epoch 272/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7528 - acc: 0.6677 - val_loss: 16.0539 - val_acc: 0.6653\n",
      "Epoch 273/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7632 - acc: 0.6667 - val_loss: 16.0532 - val_acc: 0.6653\n",
      "Epoch 274/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7615 - acc: 0.6615 - val_loss: 16.0523 - val_acc: 0.6653\n",
      "Epoch 275/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7623 - acc: 0.6667 - val_loss: 16.0515 - val_acc: 0.6653\n",
      "Epoch 276/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7687 - acc: 0.6677 - val_loss: 16.0503 - val_acc: 0.6653\n",
      "Epoch 277/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7748 - acc: 0.6698 - val_loss: 16.0494 - val_acc: 0.6653\n",
      "Epoch 278/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7402 - acc: 0.6636 - val_loss: 16.0485 - val_acc: 0.6653\n",
      "Epoch 279/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7572 - acc: 0.6667 - val_loss: 16.0474 - val_acc: 0.6653\n",
      "Epoch 280/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7594 - acc: 0.6646 - val_loss: 16.0462 - val_acc: 0.6653\n",
      "Epoch 281/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7681 - acc: 0.6656 - val_loss: 16.0449 - val_acc: 0.6653\n",
      "Epoch 282/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7537 - acc: 0.6636 - val_loss: 16.0438 - val_acc: 0.6653\n",
      "Epoch 283/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7507 - acc: 0.6594 - val_loss: 16.0428 - val_acc: 0.6653\n",
      "Epoch 284/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7584 - acc: 0.6656 - val_loss: 16.0417 - val_acc: 0.6653\n",
      "Epoch 285/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7607 - acc: 0.6615 - val_loss: 16.0408 - val_acc: 0.6653\n",
      "Epoch 286/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7680 - acc: 0.6636 - val_loss: 16.0400 - val_acc: 0.6653\n",
      "Epoch 287/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7355 - acc: 0.6625 - val_loss: 16.0389 - val_acc: 0.6653\n",
      "Epoch 288/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7382 - acc: 0.6656 - val_loss: 16.0381 - val_acc: 0.6653\n",
      "Epoch 289/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7543 - acc: 0.6667 - val_loss: 16.0372 - val_acc: 0.6653\n",
      "Epoch 290/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7453 - acc: 0.6625 - val_loss: 16.0361 - val_acc: 0.6653\n",
      "Epoch 291/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7334 - acc: 0.6646 - val_loss: 16.0350 - val_acc: 0.6653\n",
      "Epoch 292/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7300 - acc: 0.6656 - val_loss: 16.0341 - val_acc: 0.6653\n",
      "Epoch 293/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7574 - acc: 0.6656 - val_loss: 16.0331 - val_acc: 0.6653\n",
      "Epoch 294/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7405 - acc: 0.6625 - val_loss: 16.0320 - val_acc: 0.6653\n",
      "Epoch 295/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7662 - acc: 0.6656 - val_loss: 16.0312 - val_acc: 0.6653\n",
      "Epoch 296/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7508 - acc: 0.6708 - val_loss: 16.0305 - val_acc: 0.6653\n",
      "Epoch 297/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7408 - acc: 0.6646 - val_loss: 16.0298 - val_acc: 0.6653\n",
      "Epoch 298/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7478 - acc: 0.6636 - val_loss: 16.0286 - val_acc: 0.6653\n",
      "Epoch 299/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7506 - acc: 0.6656 - val_loss: 16.0279 - val_acc: 0.6653\n",
      "Epoch 300/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7492 - acc: 0.6656 - val_loss: 16.0273 - val_acc: 0.6653\n",
      "Epoch 301/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7461 - acc: 0.6636 - val_loss: 16.0270 - val_acc: 0.6653\n",
      "Epoch 302/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7415 - acc: 0.6615 - val_loss: 16.0261 - val_acc: 0.6653\n",
      "Epoch 303/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7371 - acc: 0.6636 - val_loss: 16.0251 - val_acc: 0.6653\n",
      "Epoch 304/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7535 - acc: 0.6677 - val_loss: 16.0245 - val_acc: 0.6653\n",
      "Epoch 305/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7363 - acc: 0.6636 - val_loss: 16.0240 - val_acc: 0.6653\n",
      "Epoch 306/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7476 - acc: 0.6656 - val_loss: 16.0233 - val_acc: 0.6653\n",
      "Epoch 307/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7387 - acc: 0.6636 - val_loss: 16.0223 - val_acc: 0.6653\n",
      "Epoch 308/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7372 - acc: 0.6636 - val_loss: 16.0215 - val_acc: 0.6653\n",
      "Epoch 309/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7268 - acc: 0.6646 - val_loss: 16.0211 - val_acc: 0.6653\n",
      "Epoch 310/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7440 - acc: 0.6667 - val_loss: 16.0204 - val_acc: 0.6653\n",
      "Epoch 311/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7344 - acc: 0.6656 - val_loss: 16.0191 - val_acc: 0.6653\n",
      "Epoch 312/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7497 - acc: 0.6677 - val_loss: 16.0182 - val_acc: 0.6653\n",
      "Epoch 313/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7272 - acc: 0.6615 - val_loss: 16.0175 - val_acc: 0.6653\n",
      "Epoch 314/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7254 - acc: 0.6656 - val_loss: 16.0168 - val_acc: 0.6653\n",
      "Epoch 315/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7462 - acc: 0.6625 - val_loss: 16.0156 - val_acc: 0.6653\n",
      "Epoch 316/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.7348 - acc: 0.6656 - val_loss: 16.0152 - val_acc: 0.6653\n",
      "Epoch 317/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7499 - acc: 0.6677 - val_loss: 16.0146 - val_acc: 0.6653\n",
      "Epoch 318/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7291 - acc: 0.6636 - val_loss: 16.0141 - val_acc: 0.6653\n",
      "Epoch 319/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7271 - acc: 0.6646 - val_loss: 16.0134 - val_acc: 0.6653\n",
      "Epoch 320/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7335 - acc: 0.6667 - val_loss: 16.0125 - val_acc: 0.6653\n",
      "Epoch 321/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7460 - acc: 0.6677 - val_loss: 16.0120 - val_acc: 0.6653\n",
      "Epoch 322/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7263 - acc: 0.6667 - val_loss: 16.0112 - val_acc: 0.6653\n",
      "Epoch 323/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7356 - acc: 0.6687 - val_loss: 16.0108 - val_acc: 0.6653\n",
      "Epoch 324/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7383 - acc: 0.6636 - val_loss: 16.0099 - val_acc: 0.6653\n",
      "Epoch 325/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7468 - acc: 0.6594 - val_loss: 16.0089 - val_acc: 0.6653\n",
      "Epoch 326/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7331 - acc: 0.6615 - val_loss: 16.0085 - val_acc: 0.6653\n",
      "Epoch 327/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7146 - acc: 0.6656 - val_loss: 16.0077 - val_acc: 0.6653\n",
      "Epoch 328/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7306 - acc: 0.6636 - val_loss: 16.0069 - val_acc: 0.6653\n",
      "Epoch 329/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7163 - acc: 0.6605 - val_loss: 16.0060 - val_acc: 0.6653\n",
      "Epoch 330/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7244 - acc: 0.6615 - val_loss: 16.0048 - val_acc: 0.6653\n",
      "Epoch 331/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7352 - acc: 0.6656 - val_loss: 16.0044 - val_acc: 0.6653\n",
      "Epoch 332/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7309 - acc: 0.6625 - val_loss: 16.0037 - val_acc: 0.6653\n",
      "Epoch 333/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7409 - acc: 0.6656 - val_loss: 16.0030 - val_acc: 0.6653\n",
      "Epoch 334/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7248 - acc: 0.6656 - val_loss: 16.0026 - val_acc: 0.6653\n",
      "Epoch 335/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7312 - acc: 0.6646 - val_loss: 16.0016 - val_acc: 0.6653\n",
      "Epoch 336/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7258 - acc: 0.6605 - val_loss: 16.0009 - val_acc: 0.6653\n",
      "Epoch 337/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7244 - acc: 0.6656 - val_loss: 16.0001 - val_acc: 0.6653\n",
      "Epoch 338/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7203 - acc: 0.6667 - val_loss: 15.9997 - val_acc: 0.6653\n",
      "Epoch 339/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7228 - acc: 0.6625 - val_loss: 15.9992 - val_acc: 0.6653\n",
      "Epoch 340/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6981 - acc: 0.6636 - val_loss: 15.9987 - val_acc: 0.6653\n",
      "Epoch 341/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7307 - acc: 0.6625 - val_loss: 15.9983 - val_acc: 0.6653\n",
      "Epoch 342/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7409 - acc: 0.6656 - val_loss: 15.9979 - val_acc: 0.6653\n",
      "Epoch 343/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7037 - acc: 0.6636 - val_loss: 15.9976 - val_acc: 0.6653\n",
      "Epoch 344/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7198 - acc: 0.6667 - val_loss: 15.9974 - val_acc: 0.6653\n",
      "Epoch 345/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7204 - acc: 0.6646 - val_loss: 15.9965 - val_acc: 0.6653\n",
      "Epoch 346/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7240 - acc: 0.6656 - val_loss: 15.9956 - val_acc: 0.6653\n",
      "Epoch 347/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7271 - acc: 0.6625 - val_loss: 15.9956 - val_acc: 0.6653\n",
      "Epoch 348/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7195 - acc: 0.6646 - val_loss: 15.9953 - val_acc: 0.6653\n",
      "Epoch 349/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7134 - acc: 0.6605 - val_loss: 15.9945 - val_acc: 0.6653\n",
      "Epoch 350/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7100 - acc: 0.6615 - val_loss: 15.9938 - val_acc: 0.6653\n",
      "Epoch 351/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7331 - acc: 0.6646 - val_loss: 15.9934 - val_acc: 0.6653\n",
      "Epoch 352/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6909 - acc: 0.6636 - val_loss: 15.9930 - val_acc: 0.6653\n",
      "Epoch 353/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7207 - acc: 0.6636 - val_loss: 15.9927 - val_acc: 0.6653\n",
      "Epoch 354/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7271 - acc: 0.6656 - val_loss: 15.9919 - val_acc: 0.6653\n",
      "Epoch 355/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7179 - acc: 0.6667 - val_loss: 15.9910 - val_acc: 0.6653\n",
      "Epoch 356/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6972 - acc: 0.6656 - val_loss: 15.9907 - val_acc: 0.6653\n",
      "Epoch 357/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7075 - acc: 0.6625 - val_loss: 15.9904 - val_acc: 0.6653\n",
      "Epoch 358/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7109 - acc: 0.6646 - val_loss: 15.9900 - val_acc: 0.6653\n",
      "Epoch 359/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6996 - acc: 0.6646 - val_loss: 15.9895 - val_acc: 0.6653\n",
      "Epoch 360/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7011 - acc: 0.6667 - val_loss: 15.9890 - val_acc: 0.6653\n",
      "Epoch 361/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7165 - acc: 0.6605 - val_loss: 15.9882 - val_acc: 0.6653\n",
      "Epoch 362/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7084 - acc: 0.6615 - val_loss: 15.9873 - val_acc: 0.6653\n",
      "Epoch 363/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6959 - acc: 0.6646 - val_loss: 15.9868 - val_acc: 0.6653\n",
      "Epoch 364/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7017 - acc: 0.6646 - val_loss: 15.9862 - val_acc: 0.6653\n",
      "Epoch 365/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7126 - acc: 0.6636 - val_loss: 15.9853 - val_acc: 0.6653\n",
      "Epoch 366/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6817 - acc: 0.6667 - val_loss: 15.9845 - val_acc: 0.6653\n",
      "Epoch 367/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7178 - acc: 0.6625 - val_loss: 15.9841 - val_acc: 0.6653\n",
      "Epoch 368/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7025 - acc: 0.6687 - val_loss: 15.9834 - val_acc: 0.6653\n",
      "Epoch 369/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6946 - acc: 0.6594 - val_loss: 15.9830 - val_acc: 0.6653\n",
      "Epoch 370/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7045 - acc: 0.6656 - val_loss: 15.9821 - val_acc: 0.6653\n",
      "Epoch 371/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7064 - acc: 0.6667 - val_loss: 15.9813 - val_acc: 0.6653\n",
      "Epoch 372/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7243 - acc: 0.6636 - val_loss: 15.9807 - val_acc: 0.6653\n",
      "Epoch 373/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7030 - acc: 0.6646 - val_loss: 15.9802 - val_acc: 0.6653\n",
      "Epoch 374/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7064 - acc: 0.6646 - val_loss: 15.9799 - val_acc: 0.6653\n",
      "Epoch 375/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7028 - acc: 0.6625 - val_loss: 15.9793 - val_acc: 0.6653\n",
      "Epoch 376/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7131 - acc: 0.6625 - val_loss: 15.9790 - val_acc: 0.6653\n",
      "Epoch 377/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7103 - acc: 0.6625 - val_loss: 15.9785 - val_acc: 0.6653\n",
      "Epoch 378/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.9143 - acc: 0.65 - 0s - loss: 16.6984 - acc: 0.6646 - val_loss: 15.9783 - val_acc: 0.6653\n",
      "Epoch 379/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.7177 - acc: 0.6667 - val_loss: 15.9777 - val_acc: 0.6653\n",
      "Epoch 380/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7036 - acc: 0.6615 - val_loss: 15.9772 - val_acc: 0.6653\n",
      "Epoch 381/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7137 - acc: 0.6636 - val_loss: 15.9769 - val_acc: 0.6653\n",
      "Epoch 382/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6830 - acc: 0.6667 - val_loss: 15.9765 - val_acc: 0.6653\n",
      "Epoch 383/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7112 - acc: 0.6625 - val_loss: 15.9759 - val_acc: 0.6653\n",
      "Epoch 384/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7017 - acc: 0.6656 - val_loss: 15.9756 - val_acc: 0.6653\n",
      "Epoch 385/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6892 - acc: 0.6656 - val_loss: 15.9751 - val_acc: 0.6653\n",
      "Epoch 386/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6971 - acc: 0.6677 - val_loss: 15.9750 - val_acc: 0.6653\n",
      "Epoch 387/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7038 - acc: 0.6636 - val_loss: 15.9744 - val_acc: 0.6653\n",
      "Epoch 388/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6984 - acc: 0.6646 - val_loss: 15.9737 - val_acc: 0.6653\n",
      "Epoch 389/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7072 - acc: 0.6646 - val_loss: 15.9733 - val_acc: 0.6653\n",
      "Epoch 390/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6960 - acc: 0.6656 - val_loss: 15.9729 - val_acc: 0.6653\n",
      "Epoch 391/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6811 - acc: 0.6646 - val_loss: 15.9727 - val_acc: 0.6653\n",
      "Epoch 392/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6967 - acc: 0.6646 - val_loss: 15.9722 - val_acc: 0.6653\n",
      "Epoch 393/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7093 - acc: 0.6656 - val_loss: 15.9716 - val_acc: 0.6653\n",
      "Epoch 394/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6870 - acc: 0.6667 - val_loss: 15.9715 - val_acc: 0.6653\n",
      "Epoch 395/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7012 - acc: 0.6646 - val_loss: 15.9710 - val_acc: 0.6653\n",
      "Epoch 396/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6959 - acc: 0.6636 - val_loss: 15.9707 - val_acc: 0.6653\n",
      "Epoch 397/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6900 - acc: 0.6636 - val_loss: 15.9707 - val_acc: 0.6653\n",
      "Epoch 398/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6851 - acc: 0.6636 - val_loss: 15.9701 - val_acc: 0.6653\n",
      "Epoch 399/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6952 - acc: 0.6656 - val_loss: 15.9699 - val_acc: 0.6653\n",
      "Epoch 400/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6876 - acc: 0.6605 - val_loss: 15.9695 - val_acc: 0.6653\n",
      "Epoch 401/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6967 - acc: 0.6646 - val_loss: 15.9691 - val_acc: 0.6653\n",
      "Epoch 402/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6799 - acc: 0.6636 - val_loss: 15.9688 - val_acc: 0.6653\n",
      "Epoch 403/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7012 - acc: 0.6646 - val_loss: 15.9682 - val_acc: 0.6653\n",
      "Epoch 404/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7005 - acc: 0.6625 - val_loss: 15.9678 - val_acc: 0.6653\n",
      "Epoch 405/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6903 - acc: 0.6625 - val_loss: 15.9670 - val_acc: 0.6653\n",
      "Epoch 406/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7083 - acc: 0.6677 - val_loss: 15.9669 - val_acc: 0.6653\n",
      "Epoch 407/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6983 - acc: 0.6656 - val_loss: 15.9665 - val_acc: 0.6653\n",
      "Epoch 408/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6839 - acc: 0.6615 - val_loss: 15.9663 - val_acc: 0.6653\n",
      "Epoch 409/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7051 - acc: 0.6656 - val_loss: 15.9659 - val_acc: 0.6653\n",
      "Epoch 410/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6786 - acc: 0.6667 - val_loss: 15.9659 - val_acc: 0.6653\n",
      "Epoch 411/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6877 - acc: 0.6656 - val_loss: 15.9654 - val_acc: 0.6653\n",
      "Epoch 412/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7054 - acc: 0.6667 - val_loss: 15.9646 - val_acc: 0.6653\n",
      "Epoch 413/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6933 - acc: 0.6636 - val_loss: 15.9640 - val_acc: 0.6653\n",
      "Epoch 414/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6833 - acc: 0.6656 - val_loss: 15.9635 - val_acc: 0.6653\n",
      "Epoch 415/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6908 - acc: 0.6636 - val_loss: 15.9629 - val_acc: 0.6653\n",
      "Epoch 416/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6869 - acc: 0.6646 - val_loss: 15.9629 - val_acc: 0.6653\n",
      "Epoch 417/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6962 - acc: 0.6677 - val_loss: 15.9627 - val_acc: 0.6653\n",
      "Epoch 418/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6935 - acc: 0.6656 - val_loss: 15.9623 - val_acc: 0.6653\n",
      "Epoch 419/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6939 - acc: 0.6625 - val_loss: 15.9620 - val_acc: 0.6653\n",
      "Epoch 420/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6974 - acc: 0.6656 - val_loss: 15.9617 - val_acc: 0.6653\n",
      "Epoch 421/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6832 - acc: 0.6698 - val_loss: 15.9614 - val_acc: 0.6653\n",
      "Epoch 422/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6886 - acc: 0.6656 - val_loss: 15.9611 - val_acc: 0.6653\n",
      "Epoch 423/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6810 - acc: 0.6646 - val_loss: 15.9608 - val_acc: 0.6653\n",
      "Epoch 424/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6726 - acc: 0.6656 - val_loss: 15.9607 - val_acc: 0.6653\n",
      "Epoch 425/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7001 - acc: 0.6677 - val_loss: 15.9604 - val_acc: 0.6653\n",
      "Epoch 426/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6829 - acc: 0.6656 - val_loss: 15.9598 - val_acc: 0.6653\n",
      "Epoch 427/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6707 - acc: 0.6656 - val_loss: 15.9589 - val_acc: 0.6653\n",
      "Epoch 428/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6952 - acc: 0.6636 - val_loss: 15.9586 - val_acc: 0.6653\n",
      "Epoch 429/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6830 - acc: 0.6646 - val_loss: 15.9585 - val_acc: 0.6653\n",
      "Epoch 430/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6961 - acc: 0.6667 - val_loss: 15.9579 - val_acc: 0.6653\n",
      "Epoch 431/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6960 - acc: 0.6636 - val_loss: 15.9578 - val_acc: 0.6653\n",
      "Epoch 432/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6942 - acc: 0.6646 - val_loss: 15.9576 - val_acc: 0.6653\n",
      "Epoch 433/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6807 - acc: 0.6615 - val_loss: 15.9573 - val_acc: 0.6653\n",
      "Epoch 434/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6874 - acc: 0.6656 - val_loss: 15.9573 - val_acc: 0.6653\n",
      "Epoch 435/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6970 - acc: 0.6677 - val_loss: 15.9574 - val_acc: 0.6653\n",
      "Epoch 436/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6955 - acc: 0.6636 - val_loss: 15.9570 - val_acc: 0.6653\n",
      "Epoch 437/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6933 - acc: 0.6636 - val_loss: 15.9563 - val_acc: 0.6653\n",
      "Epoch 438/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6715 - acc: 0.6667 - val_loss: 15.9557 - val_acc: 0.6653\n",
      "Epoch 439/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6786 - acc: 0.6646 - val_loss: 15.9556 - val_acc: 0.6653\n",
      "Epoch 440/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6919 - acc: 0.6625 - val_loss: 15.9554 - val_acc: 0.6653\n",
      "Epoch 441/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6749 - acc: 0.6615 - val_loss: 15.9553 - val_acc: 0.6653\n",
      "Epoch 442/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6905 - acc: 0.6636 - val_loss: 15.9548 - val_acc: 0.6653\n",
      "Epoch 443/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6661 - acc: 0.6636 - val_loss: 15.9543 - val_acc: 0.6653\n",
      "Epoch 444/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6752 - acc: 0.6615 - val_loss: 15.9540 - val_acc: 0.6653\n",
      "Epoch 445/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6750 - acc: 0.6636 - val_loss: 15.9537 - val_acc: 0.6653\n",
      "Epoch 446/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6683 - acc: 0.6687 - val_loss: 15.9533 - val_acc: 0.6653\n",
      "Epoch 447/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6818 - acc: 0.6636 - val_loss: 15.9530 - val_acc: 0.6653\n",
      "Epoch 448/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6757 - acc: 0.6594 - val_loss: 15.9525 - val_acc: 0.6653\n",
      "Epoch 449/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6939 - acc: 0.6708 - val_loss: 15.9525 - val_acc: 0.6653\n",
      "Epoch 450/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6858 - acc: 0.6667 - val_loss: 15.9522 - val_acc: 0.6653\n",
      "Epoch 451/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6720 - acc: 0.6677 - val_loss: 15.9521 - val_acc: 0.6653\n",
      "Epoch 452/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6780 - acc: 0.6625 - val_loss: 15.9521 - val_acc: 0.6653\n",
      "Epoch 453/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6975 - acc: 0.6646 - val_loss: 15.9519 - val_acc: 0.6653\n",
      "Epoch 454/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6739 - acc: 0.6594 - val_loss: 15.9513 - val_acc: 0.6653\n",
      "Epoch 455/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6700 - acc: 0.6615 - val_loss: 15.9512 - val_acc: 0.6653\n",
      "Epoch 456/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6670 - acc: 0.6625 - val_loss: 15.9508 - val_acc: 0.6653\n",
      "Epoch 457/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6703 - acc: 0.6594 - val_loss: 15.9506 - val_acc: 0.6653\n",
      "Epoch 458/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6846 - acc: 0.6646 - val_loss: 15.9503 - val_acc: 0.6653\n",
      "Epoch 459/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6889 - acc: 0.6625 - val_loss: 15.9502 - val_acc: 0.6653\n",
      "Epoch 460/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6816 - acc: 0.6656 - val_loss: 15.9498 - val_acc: 0.6653\n",
      "Epoch 461/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6688 - acc: 0.6625 - val_loss: 15.9494 - val_acc: 0.6653\n",
      "Epoch 462/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6831 - acc: 0.6625 - val_loss: 15.9493 - val_acc: 0.6653\n",
      "Epoch 463/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6721 - acc: 0.6677 - val_loss: 15.9492 - val_acc: 0.6653\n",
      "Epoch 464/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6795 - acc: 0.6625 - val_loss: 15.9485 - val_acc: 0.6653\n",
      "Epoch 465/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6784 - acc: 0.6667 - val_loss: 15.9482 - val_acc: 0.6653\n",
      "Epoch 466/5000\n",
      "966/966 [==============================] - 0s - loss: 16.7042 - acc: 0.6615 - val_loss: 15.9477 - val_acc: 0.6653\n",
      "Epoch 467/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6648 - acc: 0.6667 - val_loss: 15.9475 - val_acc: 0.6653\n",
      "Epoch 468/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6816 - acc: 0.6656 - val_loss: 15.9472 - val_acc: 0.6653\n",
      "Epoch 469/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6733 - acc: 0.6677 - val_loss: 15.9470 - val_acc: 0.6653\n",
      "Epoch 470/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6740 - acc: 0.6646 - val_loss: 15.9467 - val_acc: 0.6653\n",
      "Epoch 471/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6859 - acc: 0.6625 - val_loss: 15.9463 - val_acc: 0.6653\n",
      "Epoch 472/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6750 - acc: 0.6646 - val_loss: 15.9461 - val_acc: 0.6653\n",
      "Epoch 473/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6734 - acc: 0.6615 - val_loss: 15.9457 - val_acc: 0.6653\n",
      "Epoch 474/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6723 - acc: 0.6646 - val_loss: 15.9454 - val_acc: 0.6653\n",
      "Epoch 475/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6638 - acc: 0.6636 - val_loss: 15.9455 - val_acc: 0.6653\n",
      "Epoch 476/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6655 - acc: 0.6625 - val_loss: 15.9456 - val_acc: 0.6653\n",
      "Epoch 477/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6691 - acc: 0.6677 - val_loss: 15.9452 - val_acc: 0.6653\n",
      "Epoch 478/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6844 - acc: 0.6646 - val_loss: 15.9450 - val_acc: 0.6653\n",
      "Epoch 479/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6623 - acc: 0.6636 - val_loss: 15.9447 - val_acc: 0.6653\n",
      "Epoch 480/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6699 - acc: 0.6594 - val_loss: 15.9447 - val_acc: 0.6653\n",
      "Epoch 481/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6574 - acc: 0.6625 - val_loss: 15.9443 - val_acc: 0.6653\n",
      "Epoch 482/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6824 - acc: 0.6625 - val_loss: 15.9443 - val_acc: 0.6653\n",
      "Epoch 483/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6819 - acc: 0.6646 - val_loss: 15.9439 - val_acc: 0.6653\n",
      "Epoch 484/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6789 - acc: 0.6667 - val_loss: 15.9437 - val_acc: 0.6653\n",
      "Epoch 485/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6850 - acc: 0.6677 - val_loss: 15.9435 - val_acc: 0.6653\n",
      "Epoch 486/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6810 - acc: 0.6646 - val_loss: 15.9431 - val_acc: 0.6653\n",
      "Epoch 487/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6676 - acc: 0.6636 - val_loss: 15.9429 - val_acc: 0.6653\n",
      "Epoch 488/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6487 - acc: 0.6646 - val_loss: 15.9425 - val_acc: 0.6653\n",
      "Epoch 489/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6670 - acc: 0.6636 - val_loss: 15.9420 - val_acc: 0.6653\n",
      "Epoch 490/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6801 - acc: 0.6615 - val_loss: 15.9418 - val_acc: 0.6653\n",
      "Epoch 491/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6716 - acc: 0.6646 - val_loss: 15.9417 - val_acc: 0.6653\n",
      "Epoch 492/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6674 - acc: 0.6687 - val_loss: 15.9412 - val_acc: 0.6653\n",
      "Epoch 493/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6745 - acc: 0.6636 - val_loss: 15.9414 - val_acc: 0.6653\n",
      "Epoch 494/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6708 - acc: 0.6584 - val_loss: 15.9413 - val_acc: 0.6653\n",
      "Epoch 495/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6694 - acc: 0.6615 - val_loss: 15.9408 - val_acc: 0.6653\n",
      "Epoch 496/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6783 - acc: 0.6667 - val_loss: 15.9405 - val_acc: 0.6653\n",
      "Epoch 497/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6928 - acc: 0.6656 - val_loss: 15.9405 - val_acc: 0.6653\n",
      "Epoch 498/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6789 - acc: 0.6656 - val_loss: 15.9403 - val_acc: 0.6653\n",
      "Epoch 499/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6628 - acc: 0.6605 - val_loss: 15.9400 - val_acc: 0.6653\n",
      "Epoch 500/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6537 - acc: 0.6605 - val_loss: 15.9397 - val_acc: 0.6653\n",
      "Epoch 501/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6820 - acc: 0.6656 - val_loss: 15.9395 - val_acc: 0.6653\n",
      "Epoch 502/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6519 - acc: 0.6625 - val_loss: 15.9394 - val_acc: 0.6653\n",
      "Epoch 503/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6770 - acc: 0.6636 - val_loss: 15.9391 - val_acc: 0.6653\n",
      "Epoch 504/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6651 - acc: 0.6646 - val_loss: 15.9390 - val_acc: 0.6653\n",
      "Epoch 505/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6908 - acc: 0.6646 - val_loss: 15.9388 - val_acc: 0.6653\n",
      "Epoch 506/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6662 - acc: 0.6646 - val_loss: 15.9388 - val_acc: 0.6653\n",
      "Epoch 507/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6748 - acc: 0.6646 - val_loss: 15.9388 - val_acc: 0.6653\n",
      "Epoch 508/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6743 - acc: 0.6646 - val_loss: 15.9387 - val_acc: 0.6653\n",
      "Epoch 509/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6675 - acc: 0.6615 - val_loss: 15.9387 - val_acc: 0.6653\n",
      "Epoch 510/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6700 - acc: 0.6584 - val_loss: 15.9384 - val_acc: 0.6653\n",
      "Epoch 511/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6717 - acc: 0.6615 - val_loss: 15.9384 - val_acc: 0.6653\n",
      "Epoch 512/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6669 - acc: 0.6646 - val_loss: 15.9380 - val_acc: 0.6653\n",
      "Epoch 513/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6682 - acc: 0.6646 - val_loss: 15.9378 - val_acc: 0.6653\n",
      "Epoch 514/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6627 - acc: 0.6646 - val_loss: 15.9374 - val_acc: 0.6653\n",
      "Epoch 515/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6906 - acc: 0.6677 - val_loss: 15.9372 - val_acc: 0.6653\n",
      "Epoch 516/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6621 - acc: 0.6667 - val_loss: 15.9369 - val_acc: 0.6653\n",
      "Epoch 517/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6509 - acc: 0.6656 - val_loss: 15.9368 - val_acc: 0.6653\n",
      "Epoch 518/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6485 - acc: 0.6646 - val_loss: 15.9366 - val_acc: 0.6653\n",
      "Epoch 519/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6342 - acc: 0.6605 - val_loss: 15.9367 - val_acc: 0.6653\n",
      "Epoch 520/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6557 - acc: 0.6677 - val_loss: 15.9366 - val_acc: 0.6653\n",
      "Epoch 521/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6518 - acc: 0.6646 - val_loss: 15.9365 - val_acc: 0.6653\n",
      "Epoch 522/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6734 - acc: 0.6636 - val_loss: 15.9365 - val_acc: 0.6653\n",
      "Epoch 523/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6741 - acc: 0.6636 - val_loss: 15.9363 - val_acc: 0.6653\n",
      "Epoch 524/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6758 - acc: 0.6615 - val_loss: 15.9360 - val_acc: 0.6653\n",
      "Epoch 525/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6644 - acc: 0.6636 - val_loss: 15.9357 - val_acc: 0.6653\n",
      "Epoch 526/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6655 - acc: 0.6667 - val_loss: 15.9356 - val_acc: 0.6653\n",
      "Epoch 527/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6534 - acc: 0.6646 - val_loss: 15.9351 - val_acc: 0.6653\n",
      "Epoch 528/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6533 - acc: 0.6646 - val_loss: 15.9349 - val_acc: 0.6653\n",
      "Epoch 529/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6691 - acc: 0.6636 - val_loss: 15.9343 - val_acc: 0.6653\n",
      "Epoch 530/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6612 - acc: 0.6605 - val_loss: 15.9341 - val_acc: 0.6653\n",
      "Epoch 531/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6572 - acc: 0.6625 - val_loss: 15.9341 - val_acc: 0.6653\n",
      "Epoch 532/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6791 - acc: 0.6646 - val_loss: 15.9340 - val_acc: 0.6653\n",
      "Epoch 533/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6686 - acc: 0.6625 - val_loss: 15.9338 - val_acc: 0.6653\n",
      "Epoch 534/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6470 - acc: 0.6615 - val_loss: 15.9332 - val_acc: 0.6653\n",
      "Epoch 535/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6426 - acc: 0.6625 - val_loss: 15.9331 - val_acc: 0.6653\n",
      "Epoch 536/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6605 - acc: 0.6625 - val_loss: 15.9329 - val_acc: 0.6653\n",
      "Epoch 537/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6866 - acc: 0.6656 - val_loss: 15.9327 - val_acc: 0.6653\n",
      "Epoch 538/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6655 - acc: 0.6636 - val_loss: 15.9325 - val_acc: 0.6653\n",
      "Epoch 539/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6462 - acc: 0.6656 - val_loss: 15.9323 - val_acc: 0.6653\n",
      "Epoch 540/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6699 - acc: 0.6646 - val_loss: 15.9320 - val_acc: 0.6653\n",
      "Epoch 541/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6716 - acc: 0.6615 - val_loss: 15.9319 - val_acc: 0.6653\n",
      "Epoch 542/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6711 - acc: 0.6687 - val_loss: 15.9314 - val_acc: 0.6653\n",
      "Epoch 543/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6557 - acc: 0.6636 - val_loss: 15.9313 - val_acc: 0.6653\n",
      "Epoch 544/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6658 - acc: 0.6636 - val_loss: 15.9310 - val_acc: 0.6653\n",
      "Epoch 545/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6545 - acc: 0.6656 - val_loss: 15.9310 - val_acc: 0.6653\n",
      "Epoch 546/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6362 - acc: 0.6625 - val_loss: 15.9308 - val_acc: 0.6653\n",
      "Epoch 547/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6757 - acc: 0.6636 - val_loss: 15.9304 - val_acc: 0.6653\n",
      "Epoch 548/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6547 - acc: 0.6646 - val_loss: 15.9305 - val_acc: 0.6653\n",
      "Epoch 549/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6569 - acc: 0.6636 - val_loss: 15.9301 - val_acc: 0.6653\n",
      "Epoch 550/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6693 - acc: 0.6687 - val_loss: 15.9293 - val_acc: 0.6653\n",
      "Epoch 551/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6640 - acc: 0.6677 - val_loss: 15.9290 - val_acc: 0.6653\n",
      "Epoch 552/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6561 - acc: 0.6636 - val_loss: 15.9289 - val_acc: 0.6653\n",
      "Epoch 553/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6747 - acc: 0.6646 - val_loss: 15.9287 - val_acc: 0.6653\n",
      "Epoch 554/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6606 - acc: 0.6687 - val_loss: 15.9288 - val_acc: 0.6653\n",
      "Epoch 555/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6403 - acc: 0.6656 - val_loss: 15.9287 - val_acc: 0.6653\n",
      "Epoch 556/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6592 - acc: 0.6656 - val_loss: 15.9285 - val_acc: 0.6653\n",
      "Epoch 557/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6623 - acc: 0.6615 - val_loss: 15.9284 - val_acc: 0.6653\n",
      "Epoch 558/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6670 - acc: 0.6708 - val_loss: 15.9283 - val_acc: 0.6653\n",
      "Epoch 559/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6657 - acc: 0.6615 - val_loss: 15.9280 - val_acc: 0.6653\n",
      "Epoch 560/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6612 - acc: 0.6646 - val_loss: 15.9279 - val_acc: 0.6653\n",
      "Epoch 561/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6667 - acc: 0.6656 - val_loss: 15.9280 - val_acc: 0.6653\n",
      "Epoch 562/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6626 - acc: 0.6625 - val_loss: 15.9280 - val_acc: 0.6653\n",
      "Epoch 563/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6563 - acc: 0.6636 - val_loss: 15.9281 - val_acc: 0.6653\n",
      "Epoch 564/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6746 - acc: 0.6646 - val_loss: 15.9281 - val_acc: 0.6653\n",
      "Epoch 565/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6522 - acc: 0.6718 - val_loss: 15.9277 - val_acc: 0.6653\n",
      "Epoch 566/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6494 - acc: 0.6667 - val_loss: 15.9274 - val_acc: 0.6653\n",
      "Epoch 567/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6468 - acc: 0.6625 - val_loss: 15.9272 - val_acc: 0.6653\n",
      "Epoch 568/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6466 - acc: 0.6646 - val_loss: 15.9270 - val_acc: 0.6653\n",
      "Epoch 569/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6576 - acc: 0.6615 - val_loss: 15.9268 - val_acc: 0.6653\n",
      "Epoch 570/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6489 - acc: 0.6615 - val_loss: 15.9268 - val_acc: 0.6653\n",
      "Epoch 571/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6624 - acc: 0.6636 - val_loss: 15.9267 - val_acc: 0.6653\n",
      "Epoch 572/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6671 - acc: 0.6625 - val_loss: 15.9264 - val_acc: 0.6653\n",
      "Epoch 573/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6537 - acc: 0.6636 - val_loss: 15.9267 - val_acc: 0.6653\n",
      "Epoch 574/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6659 - acc: 0.6573 - val_loss: 15.9264 - val_acc: 0.6653\n",
      "Epoch 575/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6452 - acc: 0.6615 - val_loss: 15.9263 - val_acc: 0.6653\n",
      "Epoch 576/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6505 - acc: 0.6667 - val_loss: 15.9260 - val_acc: 0.6653\n",
      "Epoch 577/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6410 - acc: 0.6677 - val_loss: 15.9259 - val_acc: 0.6653\n",
      "Epoch 578/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6412 - acc: 0.6605 - val_loss: 15.9259 - val_acc: 0.6653\n",
      "Epoch 579/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6527 - acc: 0.6718 - val_loss: 15.9255 - val_acc: 0.6653\n",
      "Epoch 580/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6729 - acc: 0.6605 - val_loss: 15.9253 - val_acc: 0.6653\n",
      "Epoch 581/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6507 - acc: 0.6636 - val_loss: 15.9253 - val_acc: 0.6653\n",
      "Epoch 582/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6561 - acc: 0.6656 - val_loss: 15.9253 - val_acc: 0.6653\n",
      "Epoch 583/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6565 - acc: 0.6636 - val_loss: 15.9252 - val_acc: 0.6653\n",
      "Epoch 584/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6548 - acc: 0.6646 - val_loss: 15.9251 - val_acc: 0.6653\n",
      "Epoch 585/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6599 - acc: 0.6615 - val_loss: 15.9249 - val_acc: 0.6653\n",
      "Epoch 586/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6539 - acc: 0.6625 - val_loss: 15.9248 - val_acc: 0.6653\n",
      "Epoch 587/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6507 - acc: 0.6646 - val_loss: 15.9249 - val_acc: 0.6653\n",
      "Epoch 588/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6654 - acc: 0.6625 - val_loss: 15.9247 - val_acc: 0.6653\n",
      "Epoch 589/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6433 - acc: 0.6625 - val_loss: 15.9247 - val_acc: 0.6653\n",
      "Epoch 590/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6703 - acc: 0.6667 - val_loss: 15.9244 - val_acc: 0.6653\n",
      "Epoch 591/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6536 - acc: 0.6584 - val_loss: 15.9244 - val_acc: 0.6653\n",
      "Epoch 592/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6678 - acc: 0.6625 - val_loss: 15.9241 - val_acc: 0.6653\n",
      "Epoch 593/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6453 - acc: 0.6667 - val_loss: 15.9238 - val_acc: 0.6653\n",
      "Epoch 594/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6576 - acc: 0.6667 - val_loss: 15.9236 - val_acc: 0.6653\n",
      "Epoch 595/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6600 - acc: 0.6615 - val_loss: 15.9237 - val_acc: 0.6653\n",
      "Epoch 596/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6430 - acc: 0.6625 - val_loss: 15.9237 - val_acc: 0.6653\n",
      "Epoch 597/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6267 - acc: 0.6605 - val_loss: 15.9235 - val_acc: 0.6653\n",
      "Epoch 598/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6655 - acc: 0.6687 - val_loss: 15.9234 - val_acc: 0.6653\n",
      "Epoch 599/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6442 - acc: 0.6656 - val_loss: 15.9232 - val_acc: 0.6653\n",
      "Epoch 600/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6512 - acc: 0.6656 - val_loss: 15.9231 - val_acc: 0.6653\n",
      "Epoch 601/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6565 - acc: 0.6625 - val_loss: 15.9230 - val_acc: 0.6653\n",
      "Epoch 602/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6428 - acc: 0.6646 - val_loss: 15.9228 - val_acc: 0.6653\n",
      "Epoch 603/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6455 - acc: 0.6615 - val_loss: 15.9227 - val_acc: 0.6653\n",
      "Epoch 604/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6528 - acc: 0.6667 - val_loss: 15.9224 - val_acc: 0.6653\n",
      "Epoch 605/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6431 - acc: 0.6667 - val_loss: 15.9222 - val_acc: 0.6653\n",
      "Epoch 606/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6451 - acc: 0.6667 - val_loss: 15.9221 - val_acc: 0.6653\n",
      "Epoch 607/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6618 - acc: 0.6677 - val_loss: 15.9217 - val_acc: 0.6653\n",
      "Epoch 608/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6413 - acc: 0.6667 - val_loss: 15.9216 - val_acc: 0.6653\n",
      "Epoch 609/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6580 - acc: 0.6677 - val_loss: 15.9217 - val_acc: 0.6653\n",
      "Epoch 610/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6603 - acc: 0.6656 - val_loss: 15.9216 - val_acc: 0.6653\n",
      "Epoch 611/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6332 - acc: 0.6594 - val_loss: 15.9215 - val_acc: 0.6653\n",
      "Epoch 612/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6492 - acc: 0.6667 - val_loss: 15.9214 - val_acc: 0.6653\n",
      "Epoch 613/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6474 - acc: 0.6594 - val_loss: 15.9214 - val_acc: 0.6653\n",
      "Epoch 614/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6451 - acc: 0.6594 - val_loss: 15.9213 - val_acc: 0.6653\n",
      "Epoch 615/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6459 - acc: 0.6656 - val_loss: 15.9213 - val_acc: 0.6653\n",
      "Epoch 616/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6549 - acc: 0.6594 - val_loss: 15.9210 - val_acc: 0.6653\n",
      "Epoch 617/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6607 - acc: 0.6605 - val_loss: 15.9210 - val_acc: 0.6653\n",
      "Epoch 618/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6603 - acc: 0.6656 - val_loss: 15.9210 - val_acc: 0.6653\n",
      "Epoch 619/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6582 - acc: 0.6656 - val_loss: 15.9209 - val_acc: 0.6653\n",
      "Epoch 620/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6479 - acc: 0.6615 - val_loss: 15.9206 - val_acc: 0.6653\n",
      "Epoch 621/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6549 - acc: 0.6625 - val_loss: 15.9202 - val_acc: 0.6653\n",
      "Epoch 622/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6671 - acc: 0.6594 - val_loss: 15.9200 - val_acc: 0.6653\n",
      "Epoch 623/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6340 - acc: 0.6677 - val_loss: 15.9201 - val_acc: 0.6694\n",
      "Epoch 624/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6428 - acc: 0.6646 - val_loss: 15.9199 - val_acc: 0.6694\n",
      "Epoch 625/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6611 - acc: 0.6698 - val_loss: 15.9197 - val_acc: 0.6694\n",
      "Epoch 626/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6444 - acc: 0.6646 - val_loss: 15.9197 - val_acc: 0.6694\n",
      "Epoch 627/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6584 - acc: 0.6656 - val_loss: 15.9196 - val_acc: 0.6694\n",
      "Epoch 628/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6705 - acc: 0.6594 - val_loss: 15.9193 - val_acc: 0.6694\n",
      "Epoch 629/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6565 - acc: 0.6646 - val_loss: 15.9192 - val_acc: 0.6694\n",
      "Epoch 630/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6663 - acc: 0.6636 - val_loss: 15.9188 - val_acc: 0.6694\n",
      "Epoch 631/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6418 - acc: 0.6646 - val_loss: 15.9183 - val_acc: 0.6694\n",
      "Epoch 632/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6302 - acc: 0.6625 - val_loss: 15.9181 - val_acc: 0.6694\n",
      "Epoch 633/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6462 - acc: 0.6615 - val_loss: 15.9182 - val_acc: 0.6694\n",
      "Epoch 634/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6538 - acc: 0.6605 - val_loss: 15.9180 - val_acc: 0.6694\n",
      "Epoch 635/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6427 - acc: 0.6646 - val_loss: 15.9178 - val_acc: 0.6694\n",
      "Epoch 636/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6522 - acc: 0.6667 - val_loss: 15.9179 - val_acc: 0.6694\n",
      "Epoch 637/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6608 - acc: 0.6615 - val_loss: 15.9178 - val_acc: 0.6694\n",
      "Epoch 638/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6622 - acc: 0.6667 - val_loss: 15.9175 - val_acc: 0.6694\n",
      "Epoch 639/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6305 - acc: 0.6677 - val_loss: 15.9172 - val_acc: 0.6694\n",
      "Epoch 640/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6624 - acc: 0.6646 - val_loss: 15.9171 - val_acc: 0.6694\n",
      "Epoch 641/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6483 - acc: 0.6667 - val_loss: 15.9169 - val_acc: 0.6694\n",
      "Epoch 642/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6472 - acc: 0.6605 - val_loss: 15.9168 - val_acc: 0.6694\n",
      "Epoch 643/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6599 - acc: 0.6573 - val_loss: 15.9167 - val_acc: 0.6694\n",
      "Epoch 644/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6399 - acc: 0.6636 - val_loss: 15.9164 - val_acc: 0.6694\n",
      "Epoch 645/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6523 - acc: 0.6656 - val_loss: 15.9163 - val_acc: 0.6694\n",
      "Epoch 646/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6637 - acc: 0.6615 - val_loss: 15.9163 - val_acc: 0.6694\n",
      "Epoch 647/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6303 - acc: 0.6584 - val_loss: 15.9163 - val_acc: 0.6694\n",
      "Epoch 648/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6443 - acc: 0.6625 - val_loss: 15.9161 - val_acc: 0.6694\n",
      "Epoch 649/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6543 - acc: 0.6605 - val_loss: 15.9162 - val_acc: 0.6694\n",
      "Epoch 650/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6478 - acc: 0.6605 - val_loss: 15.9162 - val_acc: 0.6694\n",
      "Epoch 651/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6419 - acc: 0.6605 - val_loss: 15.9163 - val_acc: 0.6694\n",
      "Epoch 652/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6636 - acc: 0.6605 - val_loss: 15.9161 - val_acc: 0.6694\n",
      "Epoch 653/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6626 - acc: 0.6615 - val_loss: 15.9157 - val_acc: 0.6694\n",
      "Epoch 654/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6469 - acc: 0.6605 - val_loss: 15.9155 - val_acc: 0.6694\n",
      "Epoch 655/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6587 - acc: 0.6646 - val_loss: 15.9153 - val_acc: 0.6694\n",
      "Epoch 656/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6486 - acc: 0.6646 - val_loss: 15.9152 - val_acc: 0.6694\n",
      "Epoch 657/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6399 - acc: 0.6667 - val_loss: 15.9151 - val_acc: 0.6694\n",
      "Epoch 658/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6550 - acc: 0.6605 - val_loss: 15.9150 - val_acc: 0.6694\n",
      "Epoch 659/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6666 - acc: 0.6594 - val_loss: 15.9149 - val_acc: 0.6694\n",
      "Epoch 660/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6719 - acc: 0.6605 - val_loss: 15.9147 - val_acc: 0.6694\n",
      "Epoch 661/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6540 - acc: 0.6656 - val_loss: 15.9145 - val_acc: 0.6694\n",
      "Epoch 662/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6301 - acc: 0.6625 - val_loss: 15.9141 - val_acc: 0.6694\n",
      "Epoch 663/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6541 - acc: 0.6698 - val_loss: 15.9142 - val_acc: 0.6694\n",
      "Epoch 664/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6496 - acc: 0.6605 - val_loss: 15.9140 - val_acc: 0.6694\n",
      "Epoch 665/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6601 - acc: 0.6625 - val_loss: 15.9140 - val_acc: 0.6694\n",
      "Epoch 666/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6392 - acc: 0.6563 - val_loss: 15.9143 - val_acc: 0.6694\n",
      "Epoch 667/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6498 - acc: 0.6646 - val_loss: 15.9144 - val_acc: 0.6694\n",
      "Epoch 668/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6544 - acc: 0.6594 - val_loss: 15.9144 - val_acc: 0.6694\n",
      "Epoch 669/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6583 - acc: 0.6584 - val_loss: 15.9141 - val_acc: 0.6694\n",
      "Epoch 670/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6282 - acc: 0.6615 - val_loss: 15.9140 - val_acc: 0.6694\n",
      "Epoch 671/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6429 - acc: 0.6646 - val_loss: 15.9140 - val_acc: 0.6694\n",
      "Epoch 672/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6415 - acc: 0.6605 - val_loss: 15.9139 - val_acc: 0.6694\n",
      "Epoch 673/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6458 - acc: 0.6594 - val_loss: 15.9139 - val_acc: 0.6694\n",
      "Epoch 674/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6401 - acc: 0.6594 - val_loss: 15.9138 - val_acc: 0.6694\n",
      "Epoch 675/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6429 - acc: 0.6646 - val_loss: 15.9137 - val_acc: 0.6694\n",
      "Epoch 676/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6417 - acc: 0.6605 - val_loss: 15.9134 - val_acc: 0.6694\n",
      "Epoch 677/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6441 - acc: 0.6687 - val_loss: 15.9134 - val_acc: 0.6694\n",
      "Epoch 678/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6420 - acc: 0.6594 - val_loss: 15.9134 - val_acc: 0.6694\n",
      "Epoch 679/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6468 - acc: 0.6615 - val_loss: 15.9132 - val_acc: 0.6694\n",
      "Epoch 680/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6602 - acc: 0.6625 - val_loss: 15.9133 - val_acc: 0.6694\n",
      "Epoch 681/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6549 - acc: 0.6636 - val_loss: 15.9131 - val_acc: 0.6694\n",
      "Epoch 682/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6386 - acc: 0.6625 - val_loss: 15.9130 - val_acc: 0.6694\n",
      "Epoch 683/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6344 - acc: 0.6636 - val_loss: 15.9130 - val_acc: 0.6694\n",
      "Epoch 684/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6194 - acc: 0.6605 - val_loss: 15.9131 - val_acc: 0.6694\n",
      "Epoch 685/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6529 - acc: 0.6625 - val_loss: 15.9128 - val_acc: 0.6694\n",
      "Epoch 686/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6469 - acc: 0.6605 - val_loss: 15.9125 - val_acc: 0.6694\n",
      "Epoch 687/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6533 - acc: 0.6605 - val_loss: 15.9124 - val_acc: 0.6694\n",
      "Epoch 688/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6417 - acc: 0.6625 - val_loss: 15.9124 - val_acc: 0.6694\n",
      "Epoch 689/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6311 - acc: 0.6646 - val_loss: 15.9124 - val_acc: 0.6694\n",
      "Epoch 690/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6535 - acc: 0.6625 - val_loss: 15.9124 - val_acc: 0.6694\n",
      "Epoch 691/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6595 - acc: 0.6636 - val_loss: 15.9124 - val_acc: 0.6694\n",
      "Epoch 692/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6501 - acc: 0.6615 - val_loss: 15.9123 - val_acc: 0.6694\n",
      "Epoch 693/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6401 - acc: 0.6594 - val_loss: 15.9122 - val_acc: 0.6694\n",
      "Epoch 694/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6361 - acc: 0.6594 - val_loss: 15.9122 - val_acc: 0.6694\n",
      "Epoch 695/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6317 - acc: 0.6573 - val_loss: 15.9119 - val_acc: 0.6694\n",
      "Epoch 696/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6377 - acc: 0.6584 - val_loss: 15.9116 - val_acc: 0.6694\n",
      "Epoch 697/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6334 - acc: 0.6687 - val_loss: 15.9115 - val_acc: 0.6694\n",
      "Epoch 698/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6544 - acc: 0.6594 - val_loss: 15.9112 - val_acc: 0.6694\n",
      "Epoch 699/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6615 - acc: 0.6656 - val_loss: 15.9110 - val_acc: 0.6694\n",
      "Epoch 700/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6439 - acc: 0.6625 - val_loss: 15.9108 - val_acc: 0.6694\n",
      "Epoch 701/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6495 - acc: 0.6625 - val_loss: 15.9107 - val_acc: 0.6694\n",
      "Epoch 702/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6610 - acc: 0.6636 - val_loss: 15.9107 - val_acc: 0.6694\n",
      "Epoch 703/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6420 - acc: 0.6625 - val_loss: 15.9107 - val_acc: 0.6694\n",
      "Epoch 704/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6351 - acc: 0.6625 - val_loss: 15.9105 - val_acc: 0.6694\n",
      "Epoch 705/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6218 - acc: 0.6563 - val_loss: 15.9106 - val_acc: 0.6694\n",
      "Epoch 706/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6237 - acc: 0.6625 - val_loss: 15.9104 - val_acc: 0.6694\n",
      "Epoch 707/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6230 - acc: 0.6625 - val_loss: 15.9102 - val_acc: 0.6694\n",
      "Epoch 708/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6379 - acc: 0.6573 - val_loss: 15.9101 - val_acc: 0.6694\n",
      "Epoch 709/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6492 - acc: 0.6605 - val_loss: 15.9100 - val_acc: 0.6694\n",
      "Epoch 710/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6417 - acc: 0.6646 - val_loss: 15.9099 - val_acc: 0.6694\n",
      "Epoch 711/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6560 - acc: 0.6636 - val_loss: 15.9097 - val_acc: 0.6694\n",
      "Epoch 712/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6283 - acc: 0.6687 - val_loss: 15.9097 - val_acc: 0.6694\n",
      "Epoch 713/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6202 - acc: 0.6625 - val_loss: 15.9096 - val_acc: 0.6694\n",
      "Epoch 714/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6306 - acc: 0.6625 - val_loss: 15.9096 - val_acc: 0.6694\n",
      "Epoch 715/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6377 - acc: 0.6636 - val_loss: 15.9097 - val_acc: 0.6694\n",
      "Epoch 716/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6201 - acc: 0.6605 - val_loss: 15.9094 - val_acc: 0.6694\n",
      "Epoch 717/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6304 - acc: 0.6656 - val_loss: 15.9095 - val_acc: 0.6694\n",
      "Epoch 718/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6480 - acc: 0.6636 - val_loss: 15.9094 - val_acc: 0.6694\n",
      "Epoch 719/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6444 - acc: 0.6615 - val_loss: 15.9091 - val_acc: 0.6694\n",
      "Epoch 720/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6322 - acc: 0.6605 - val_loss: 15.9092 - val_acc: 0.6694\n",
      "Epoch 721/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6227 - acc: 0.6625 - val_loss: 15.9090 - val_acc: 0.6694\n",
      "Epoch 722/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6441 - acc: 0.6573 - val_loss: 15.9089 - val_acc: 0.6694\n",
      "Epoch 723/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6282 - acc: 0.6605 - val_loss: 15.9088 - val_acc: 0.6694\n",
      "Epoch 724/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6491 - acc: 0.6605 - val_loss: 15.9086 - val_acc: 0.6694\n",
      "Epoch 725/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6283 - acc: 0.6646 - val_loss: 15.9086 - val_acc: 0.6694\n",
      "Epoch 726/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6258 - acc: 0.6615 - val_loss: 15.9088 - val_acc: 0.6694\n",
      "Epoch 727/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6298 - acc: 0.6646 - val_loss: 15.9089 - val_acc: 0.6694\n",
      "Epoch 728/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6282 - acc: 0.6625 - val_loss: 15.9088 - val_acc: 0.6694\n",
      "Epoch 729/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6383 - acc: 0.6615 - val_loss: 15.9089 - val_acc: 0.6694\n",
      "Epoch 730/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6266 - acc: 0.6656 - val_loss: 15.9089 - val_acc: 0.6694\n",
      "Epoch 731/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6496 - acc: 0.6573 - val_loss: 15.9089 - val_acc: 0.6694\n",
      "Epoch 732/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6382 - acc: 0.6615 - val_loss: 15.9086 - val_acc: 0.6694\n",
      "Epoch 733/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6331 - acc: 0.6636 - val_loss: 15.9086 - val_acc: 0.6694\n",
      "Epoch 734/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6375 - acc: 0.6636 - val_loss: 15.9083 - val_acc: 0.6694\n",
      "Epoch 735/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6303 - acc: 0.6542 - val_loss: 15.9081 - val_acc: 0.6694\n",
      "Epoch 736/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6362 - acc: 0.6594 - val_loss: 15.9078 - val_acc: 0.6694\n",
      "Epoch 737/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6413 - acc: 0.6636 - val_loss: 15.9076 - val_acc: 0.6694\n",
      "Epoch 738/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6475 - acc: 0.6625 - val_loss: 15.9075 - val_acc: 0.6694\n",
      "Epoch 739/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6333 - acc: 0.6615 - val_loss: 15.9074 - val_acc: 0.6694\n",
      "Epoch 740/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6337 - acc: 0.6584 - val_loss: 15.9073 - val_acc: 0.6694\n",
      "Epoch 741/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6365 - acc: 0.6594 - val_loss: 15.9071 - val_acc: 0.6694\n",
      "Epoch 742/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6351 - acc: 0.6636 - val_loss: 15.9070 - val_acc: 0.6694\n",
      "Epoch 743/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6153 - acc: 0.6615 - val_loss: 15.9068 - val_acc: 0.6694\n",
      "Epoch 744/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6377 - acc: 0.6594 - val_loss: 15.9068 - val_acc: 0.6694\n",
      "Epoch 745/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6333 - acc: 0.6594 - val_loss: 15.9068 - val_acc: 0.6694\n",
      "Epoch 746/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6221 - acc: 0.6646 - val_loss: 15.9066 - val_acc: 0.6694\n",
      "Epoch 747/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6712 - acc: 0.6625 - val_loss: 15.9066 - val_acc: 0.6694\n",
      "Epoch 748/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6376 - acc: 0.6615 - val_loss: 15.9065 - val_acc: 0.6694\n",
      "Epoch 749/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6481 - acc: 0.6584 - val_loss: 15.9064 - val_acc: 0.6694\n",
      "Epoch 750/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6345 - acc: 0.6656 - val_loss: 15.9062 - val_acc: 0.6694\n",
      "Epoch 751/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6360 - acc: 0.6656 - val_loss: 15.9060 - val_acc: 0.6694\n",
      "Epoch 752/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6346 - acc: 0.6646 - val_loss: 15.9057 - val_acc: 0.6694\n",
      "Epoch 753/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6325 - acc: 0.6553 - val_loss: 15.9055 - val_acc: 0.6694\n",
      "Epoch 754/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6384 - acc: 0.6636 - val_loss: 15.9054 - val_acc: 0.6694\n",
      "Epoch 755/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6302 - acc: 0.6636 - val_loss: 15.9053 - val_acc: 0.6694\n",
      "Epoch 756/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6361 - acc: 0.6625 - val_loss: 15.9049 - val_acc: 0.6694\n",
      "Epoch 757/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6263 - acc: 0.6594 - val_loss: 15.9046 - val_acc: 0.6694\n",
      "Epoch 758/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6499 - acc: 0.6636 - val_loss: 15.9046 - val_acc: 0.6694\n",
      "Epoch 759/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6407 - acc: 0.6522 - val_loss: 15.9045 - val_acc: 0.6694\n",
      "Epoch 760/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6335 - acc: 0.6625 - val_loss: 15.9044 - val_acc: 0.6694\n",
      "Epoch 761/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6255 - acc: 0.6625 - val_loss: 15.9044 - val_acc: 0.6694\n",
      "Epoch 762/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6507 - acc: 0.6605 - val_loss: 15.9042 - val_acc: 0.6694\n",
      "Epoch 763/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6300 - acc: 0.6646 - val_loss: 15.9043 - val_acc: 0.6694\n",
      "Epoch 764/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6247 - acc: 0.6594 - val_loss: 15.9045 - val_acc: 0.6694\n",
      "Epoch 765/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6252 - acc: 0.6584 - val_loss: 15.9047 - val_acc: 0.6694\n",
      "Epoch 766/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6497 - acc: 0.6501 - val_loss: 15.9048 - val_acc: 0.6694\n",
      "Epoch 767/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6534 - acc: 0.6584 - val_loss: 15.9047 - val_acc: 0.6694\n",
      "Epoch 768/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6271 - acc: 0.6605 - val_loss: 15.9048 - val_acc: 0.6694\n",
      "Epoch 769/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6457 - acc: 0.6656 - val_loss: 15.9046 - val_acc: 0.6694\n",
      "Epoch 770/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6354 - acc: 0.6687 - val_loss: 15.9045 - val_acc: 0.6694\n",
      "Epoch 771/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6324 - acc: 0.6594 - val_loss: 15.9045 - val_acc: 0.6694\n",
      "Epoch 772/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6441 - acc: 0.6573 - val_loss: 15.9043 - val_acc: 0.6694\n",
      "Epoch 773/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6412 - acc: 0.6625 - val_loss: 15.9042 - val_acc: 0.6694\n",
      "Epoch 774/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6209 - acc: 0.6615 - val_loss: 15.9039 - val_acc: 0.6694\n",
      "Epoch 775/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6269 - acc: 0.6584 - val_loss: 15.9040 - val_acc: 0.6694\n",
      "Epoch 776/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6415 - acc: 0.6594 - val_loss: 15.9038 - val_acc: 0.6694\n",
      "Epoch 777/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6299 - acc: 0.6636 - val_loss: 15.9036 - val_acc: 0.6694\n",
      "Epoch 778/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6250 - acc: 0.6615 - val_loss: 15.9037 - val_acc: 0.6694\n",
      "Epoch 779/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6427 - acc: 0.6584 - val_loss: 15.9036 - val_acc: 0.6694\n",
      "Epoch 780/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6188 - acc: 0.6615 - val_loss: 15.9034 - val_acc: 0.6694\n",
      "Epoch 781/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6317 - acc: 0.6625 - val_loss: 15.9031 - val_acc: 0.6694\n",
      "Epoch 782/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6110 - acc: 0.6677 - val_loss: 15.9029 - val_acc: 0.6694\n",
      "Epoch 783/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6394 - acc: 0.6615 - val_loss: 15.9029 - val_acc: 0.6694\n",
      "Epoch 784/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6491 - acc: 0.6656 - val_loss: 15.9028 - val_acc: 0.6694\n",
      "Epoch 785/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6418 - acc: 0.6615 - val_loss: 15.9026 - val_acc: 0.6694\n",
      "Epoch 786/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6338 - acc: 0.6605 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 787/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6163 - acc: 0.6594 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 788/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6381 - acc: 0.6615 - val_loss: 15.9023 - val_acc: 0.6694\n",
      "Epoch 789/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6343 - acc: 0.6532 - val_loss: 15.9022 - val_acc: 0.6694\n",
      "Epoch 790/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6367 - acc: 0.6605 - val_loss: 15.9023 - val_acc: 0.6694\n",
      "Epoch 791/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6334 - acc: 0.6573 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 792/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6375 - acc: 0.6687 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 793/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6159 - acc: 0.6615 - val_loss: 15.9025 - val_acc: 0.6694\n",
      "Epoch 794/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6373 - acc: 0.6656 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 795/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6258 - acc: 0.6636 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 796/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6467 - acc: 0.6615 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 797/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6308 - acc: 0.6636 - val_loss: 15.9024 - val_acc: 0.6694\n",
      "Epoch 798/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6311 - acc: 0.6656 - val_loss: 15.9021 - val_acc: 0.6694\n",
      "Epoch 799/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6361 - acc: 0.6584 - val_loss: 15.9018 - val_acc: 0.6694\n",
      "Epoch 800/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6358 - acc: 0.6636 - val_loss: 15.9019 - val_acc: 0.6694\n",
      "Epoch 801/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6339 - acc: 0.6656 - val_loss: 15.9017 - val_acc: 0.6694\n",
      "Epoch 802/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6375 - acc: 0.6615 - val_loss: 15.9018 - val_acc: 0.6694\n",
      "Epoch 803/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6313 - acc: 0.6594 - val_loss: 15.9017 - val_acc: 0.6694\n",
      "Epoch 804/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6361 - acc: 0.6718 - val_loss: 15.9017 - val_acc: 0.6694\n",
      "Epoch 805/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6403 - acc: 0.6656 - val_loss: 15.9017 - val_acc: 0.6694\n",
      "Epoch 806/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6392 - acc: 0.6605 - val_loss: 15.9016 - val_acc: 0.6694\n",
      "Epoch 807/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6167 - acc: 0.6584 - val_loss: 15.9016 - val_acc: 0.6694\n",
      "Epoch 808/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6312 - acc: 0.6677 - val_loss: 15.9016 - val_acc: 0.6694\n",
      "Epoch 809/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6360 - acc: 0.6636 - val_loss: 15.9016 - val_acc: 0.6694\n",
      "Epoch 810/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6111 - acc: 0.6532 - val_loss: 15.9014 - val_acc: 0.6694\n",
      "Epoch 811/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6300 - acc: 0.6584 - val_loss: 15.9011 - val_acc: 0.6694\n",
      "Epoch 812/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6288 - acc: 0.6615 - val_loss: 15.9008 - val_acc: 0.6694\n",
      "Epoch 813/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6229 - acc: 0.6563 - val_loss: 15.9007 - val_acc: 0.6694\n",
      "Epoch 814/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6321 - acc: 0.6615 - val_loss: 15.9006 - val_acc: 0.6694\n",
      "Epoch 815/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6271 - acc: 0.6573 - val_loss: 15.9007 - val_acc: 0.6694\n",
      "Epoch 816/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6379 - acc: 0.6594 - val_loss: 15.9005 - val_acc: 0.6694\n",
      "Epoch 817/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6347 - acc: 0.6594 - val_loss: 15.9004 - val_acc: 0.6694\n",
      "Epoch 818/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6282 - acc: 0.6594 - val_loss: 15.9004 - val_acc: 0.6694\n",
      "Epoch 819/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6271 - acc: 0.6646 - val_loss: 15.9004 - val_acc: 0.6694\n",
      "Epoch 820/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6299 - acc: 0.6656 - val_loss: 15.9002 - val_acc: 0.6694\n",
      "Epoch 821/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6132 - acc: 0.6615 - val_loss: 15.9002 - val_acc: 0.6694\n",
      "Epoch 822/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6307 - acc: 0.6605 - val_loss: 15.9003 - val_acc: 0.6694\n",
      "Epoch 823/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6264 - acc: 0.6553 - val_loss: 15.9001 - val_acc: 0.6694\n",
      "Epoch 824/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6175 - acc: 0.6656 - val_loss: 15.8999 - val_acc: 0.6694\n",
      "Epoch 825/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6334 - acc: 0.6646 - val_loss: 15.9000 - val_acc: 0.6694\n",
      "Epoch 826/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6474 - acc: 0.6667 - val_loss: 15.9002 - val_acc: 0.6694\n",
      "Epoch 827/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6270 - acc: 0.6625 - val_loss: 15.9000 - val_acc: 0.6694\n",
      "Epoch 828/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6358 - acc: 0.6573 - val_loss: 15.8999 - val_acc: 0.6694\n",
      "Epoch 829/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6262 - acc: 0.6667 - val_loss: 15.8997 - val_acc: 0.6694\n",
      "Epoch 830/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6228 - acc: 0.6656 - val_loss: 15.8995 - val_acc: 0.6694\n",
      "Epoch 831/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6246 - acc: 0.6636 - val_loss: 15.8994 - val_acc: 0.6694\n",
      "Epoch 832/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6212 - acc: 0.6615 - val_loss: 15.8993 - val_acc: 0.6694\n",
      "Epoch 833/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6330 - acc: 0.6646 - val_loss: 15.8992 - val_acc: 0.6694\n",
      "Epoch 834/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6264 - acc: 0.6656 - val_loss: 15.8991 - val_acc: 0.6694\n",
      "Epoch 835/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6360 - acc: 0.6625 - val_loss: 15.8988 - val_acc: 0.6694\n",
      "Epoch 836/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6413 - acc: 0.6532 - val_loss: 15.8987 - val_acc: 0.6694\n",
      "Epoch 837/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6380 - acc: 0.6605 - val_loss: 15.8988 - val_acc: 0.6694\n",
      "Epoch 838/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6184 - acc: 0.6625 - val_loss: 15.8988 - val_acc: 0.6694\n",
      "Epoch 839/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6298 - acc: 0.6636 - val_loss: 15.8990 - val_acc: 0.6694\n",
      "Epoch 840/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6313 - acc: 0.6615 - val_loss: 15.8991 - val_acc: 0.6694\n",
      "Epoch 841/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6141 - acc: 0.6625 - val_loss: 15.8991 - val_acc: 0.6694\n",
      "Epoch 842/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6391 - acc: 0.6646 - val_loss: 15.8992 - val_acc: 0.6694\n",
      "Epoch 843/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6171 - acc: 0.6542 - val_loss: 15.8992 - val_acc: 0.6694\n",
      "Epoch 844/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6183 - acc: 0.6594 - val_loss: 15.8991 - val_acc: 0.6694\n",
      "Epoch 845/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6275 - acc: 0.6667 - val_loss: 15.8991 - val_acc: 0.6694\n",
      "Epoch 846/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6287 - acc: 0.6636 - val_loss: 15.8988 - val_acc: 0.6694\n",
      "Epoch 847/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6359 - acc: 0.6625 - val_loss: 15.8986 - val_acc: 0.6694\n",
      "Epoch 848/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6263 - acc: 0.6667 - val_loss: 15.8984 - val_acc: 0.6694\n",
      "Epoch 849/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6248 - acc: 0.6646 - val_loss: 15.8984 - val_acc: 0.6694\n",
      "Epoch 850/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6232 - acc: 0.6625 - val_loss: 15.8982 - val_acc: 0.6694\n",
      "Epoch 851/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6101 - acc: 0.6646 - val_loss: 15.8982 - val_acc: 0.6694\n",
      "Epoch 852/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6193 - acc: 0.6594 - val_loss: 15.8982 - val_acc: 0.6694\n",
      "Epoch 853/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6404 - acc: 0.6636 - val_loss: 15.8981 - val_acc: 0.6694\n",
      "Epoch 854/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6165 - acc: 0.6646 - val_loss: 15.8982 - val_acc: 0.6694\n",
      "Epoch 855/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6273 - acc: 0.6625 - val_loss: 15.8981 - val_acc: 0.6694\n",
      "Epoch 856/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6235 - acc: 0.6584 - val_loss: 15.8980 - val_acc: 0.6694\n",
      "Epoch 857/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6374 - acc: 0.6677 - val_loss: 15.8981 - val_acc: 0.6694\n",
      "Epoch 858/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6244 - acc: 0.6573 - val_loss: 15.8980 - val_acc: 0.6694\n",
      "Epoch 859/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6295 - acc: 0.6615 - val_loss: 15.8980 - val_acc: 0.6694\n",
      "Epoch 860/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6434 - acc: 0.6656 - val_loss: 15.8979 - val_acc: 0.6694\n",
      "Epoch 861/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6175 - acc: 0.6646 - val_loss: 15.8979 - val_acc: 0.6694\n",
      "Epoch 862/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6400 - acc: 0.6636 - val_loss: 15.8975 - val_acc: 0.6694\n",
      "Epoch 863/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6483 - acc: 0.6584 - val_loss: 15.8973 - val_acc: 0.6694\n",
      "Epoch 864/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6409 - acc: 0.6594 - val_loss: 15.8972 - val_acc: 0.6694\n",
      "Epoch 865/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6265 - acc: 0.6573 - val_loss: 15.8971 - val_acc: 0.6694\n",
      "Epoch 866/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6395 - acc: 0.6563 - val_loss: 15.8969 - val_acc: 0.6694\n",
      "Epoch 867/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6275 - acc: 0.6636 - val_loss: 15.8970 - val_acc: 0.6694\n",
      "Epoch 868/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6267 - acc: 0.6646 - val_loss: 15.8969 - val_acc: 0.6694\n",
      "Epoch 869/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6188 - acc: 0.6594 - val_loss: 15.8968 - val_acc: 0.6694\n",
      "Epoch 870/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6397 - acc: 0.6718 - val_loss: 15.8968 - val_acc: 0.6694\n",
      "Epoch 871/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6155 - acc: 0.6584 - val_loss: 15.8968 - val_acc: 0.6694\n",
      "Epoch 872/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6397 - acc: 0.6636 - val_loss: 15.8965 - val_acc: 0.6694\n",
      "Epoch 873/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6297 - acc: 0.6646 - val_loss: 15.8964 - val_acc: 0.6694\n",
      "Epoch 874/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6299 - acc: 0.6646 - val_loss: 15.8964 - val_acc: 0.6694\n",
      "Epoch 875/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6348 - acc: 0.6594 - val_loss: 15.8963 - val_acc: 0.6694\n",
      "Epoch 876/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6420 - acc: 0.6667 - val_loss: 15.8962 - val_acc: 0.6694\n",
      "Epoch 877/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6333 - acc: 0.6646 - val_loss: 15.8961 - val_acc: 0.6694\n",
      "Epoch 878/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6392 - acc: 0.6573 - val_loss: 15.8959 - val_acc: 0.6694\n",
      "Epoch 879/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6365 - acc: 0.6584 - val_loss: 15.8958 - val_acc: 0.6694\n",
      "Epoch 880/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6433 - acc: 0.6594 - val_loss: 15.8958 - val_acc: 0.6694\n",
      "Epoch 881/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6297 - acc: 0.6594 - val_loss: 15.8958 - val_acc: 0.6694\n",
      "Epoch 882/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6088 - acc: 0.6573 - val_loss: 15.8960 - val_acc: 0.6694\n",
      "Epoch 883/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6236 - acc: 0.6584 - val_loss: 15.8961 - val_acc: 0.6694\n",
      "Epoch 884/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6157 - acc: 0.6605 - val_loss: 15.8961 - val_acc: 0.6694\n",
      "Epoch 885/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6169 - acc: 0.6594 - val_loss: 15.8962 - val_acc: 0.6694\n",
      "Epoch 886/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6310 - acc: 0.6625 - val_loss: 15.8961 - val_acc: 0.6694\n",
      "Epoch 887/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6262 - acc: 0.6615 - val_loss: 15.8960 - val_acc: 0.6694\n",
      "Epoch 888/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6218 - acc: 0.6615 - val_loss: 15.8960 - val_acc: 0.6694\n",
      "Epoch 889/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6413 - acc: 0.6584 - val_loss: 15.8960 - val_acc: 0.6694\n",
      "Epoch 890/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6287 - acc: 0.6584 - val_loss: 15.8958 - val_acc: 0.6694\n",
      "Epoch 891/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6322 - acc: 0.6636 - val_loss: 15.8956 - val_acc: 0.6694\n",
      "Epoch 892/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6187 - acc: 0.6625 - val_loss: 15.8954 - val_acc: 0.6694\n",
      "Epoch 893/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6314 - acc: 0.6553 - val_loss: 15.8952 - val_acc: 0.6694\n",
      "Epoch 894/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6232 - acc: 0.6553 - val_loss: 15.8952 - val_acc: 0.6694\n",
      "Epoch 895/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6265 - acc: 0.6605 - val_loss: 15.8952 - val_acc: 0.6694\n",
      "Epoch 896/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6246 - acc: 0.6553 - val_loss: 15.8953 - val_acc: 0.6694\n",
      "Epoch 897/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6220 - acc: 0.6615 - val_loss: 15.8951 - val_acc: 0.6694\n",
      "Epoch 898/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6285 - acc: 0.6625 - val_loss: 15.8949 - val_acc: 0.6694\n",
      "Epoch 899/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6134 - acc: 0.6584 - val_loss: 15.8950 - val_acc: 0.6694\n",
      "Epoch 900/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6174 - acc: 0.6605 - val_loss: 15.8949 - val_acc: 0.6694\n",
      "Epoch 901/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6158 - acc: 0.6625 - val_loss: 15.8950 - val_acc: 0.6694\n",
      "Epoch 902/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6295 - acc: 0.6594 - val_loss: 15.8948 - val_acc: 0.6694\n",
      "Epoch 903/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6332 - acc: 0.6553 - val_loss: 15.8948 - val_acc: 0.6694\n",
      "Epoch 904/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6409 - acc: 0.6605 - val_loss: 15.8948 - val_acc: 0.6694\n",
      "Epoch 905/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6345 - acc: 0.6646 - val_loss: 15.8946 - val_acc: 0.6694\n",
      "Epoch 906/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6151 - acc: 0.6542 - val_loss: 15.8946 - val_acc: 0.6694\n",
      "Epoch 907/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6202 - acc: 0.6594 - val_loss: 15.8945 - val_acc: 0.6694\n",
      "Epoch 908/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6228 - acc: 0.6636 - val_loss: 15.8945 - val_acc: 0.6694\n",
      "Epoch 909/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6301 - acc: 0.6605 - val_loss: 15.8946 - val_acc: 0.6694\n",
      "Epoch 910/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6330 - acc: 0.6636 - val_loss: 15.8947 - val_acc: 0.6694\n",
      "Epoch 911/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6529 - acc: 0.6594 - val_loss: 15.8946 - val_acc: 0.6694\n",
      "Epoch 912/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6103 - acc: 0.6636 - val_loss: 15.8944 - val_acc: 0.6694\n",
      "Epoch 913/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6244 - acc: 0.6584 - val_loss: 15.8943 - val_acc: 0.6694\n",
      "Epoch 914/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6165 - acc: 0.6584 - val_loss: 15.8940 - val_acc: 0.6694\n",
      "Epoch 915/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6219 - acc: 0.6646 - val_loss: 15.8939 - val_acc: 0.6694\n",
      "Epoch 916/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6277 - acc: 0.6573 - val_loss: 15.8939 - val_acc: 0.6694\n",
      "Epoch 917/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6124 - acc: 0.6594 - val_loss: 15.8939 - val_acc: 0.6694\n",
      "Epoch 918/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6514 - acc: 0.6584 - val_loss: 15.8938 - val_acc: 0.6694\n",
      "Epoch 919/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6152 - acc: 0.6573 - val_loss: 15.8937 - val_acc: 0.6694\n",
      "Epoch 920/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6584 - val_loss: 15.8935 - val_acc: 0.6694\n",
      "Epoch 921/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6225 - acc: 0.6656 - val_loss: 15.8936 - val_acc: 0.6694\n",
      "Epoch 922/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6021 - acc: 0.6615 - val_loss: 15.8936 - val_acc: 0.6694\n",
      "Epoch 923/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6218 - acc: 0.6584 - val_loss: 15.8936 - val_acc: 0.6694\n",
      "Epoch 924/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6279 - acc: 0.6646 - val_loss: 15.8937 - val_acc: 0.6694\n",
      "Epoch 925/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6240 - acc: 0.6522 - val_loss: 15.8937 - val_acc: 0.6694\n",
      "Epoch 926/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6318 - acc: 0.6501 - val_loss: 15.8937 - val_acc: 0.6694\n",
      "Epoch 927/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6189 - acc: 0.6656 - val_loss: 15.8938 - val_acc: 0.6694\n",
      "Epoch 928/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6358 - acc: 0.6542 - val_loss: 15.8937 - val_acc: 0.6694\n",
      "Epoch 929/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6227 - acc: 0.6646 - val_loss: 15.8935 - val_acc: 0.6694\n",
      "Epoch 930/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6333 - acc: 0.6605 - val_loss: 15.8933 - val_acc: 0.6694\n",
      "Epoch 931/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6196 - acc: 0.6584 - val_loss: 15.8932 - val_acc: 0.6694\n",
      "Epoch 932/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6305 - acc: 0.6532 - val_loss: 15.8932 - val_acc: 0.6694\n",
      "Epoch 933/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6253 - acc: 0.6615 - val_loss: 15.8933 - val_acc: 0.6694\n",
      "Epoch 934/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6236 - acc: 0.6605 - val_loss: 15.8933 - val_acc: 0.6694\n",
      "Epoch 935/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6211 - acc: 0.6615 - val_loss: 15.8932 - val_acc: 0.6694\n",
      "Epoch 936/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6573 - val_loss: 15.8931 - val_acc: 0.6694\n",
      "Epoch 937/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6084 - acc: 0.6605 - val_loss: 15.8931 - val_acc: 0.6694\n",
      "Epoch 938/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6164 - acc: 0.6594 - val_loss: 15.8931 - val_acc: 0.6694\n",
      "Epoch 939/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6284 - acc: 0.6646 - val_loss: 15.8930 - val_acc: 0.6694\n",
      "Epoch 940/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6376 - acc: 0.6615 - val_loss: 15.8930 - val_acc: 0.6694\n",
      "Epoch 941/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6234 - acc: 0.6491 - val_loss: 15.8927 - val_acc: 0.6694\n",
      "Epoch 942/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6231 - acc: 0.6656 - val_loss: 15.8927 - val_acc: 0.6694\n",
      "Epoch 943/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6158 - acc: 0.6542 - val_loss: 15.8924 - val_acc: 0.6694\n",
      "Epoch 944/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6202 - acc: 0.6615 - val_loss: 15.8924 - val_acc: 0.6694\n",
      "Epoch 945/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6083 - acc: 0.6605 - val_loss: 15.8924 - val_acc: 0.6694\n",
      "Epoch 946/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6325 - acc: 0.6584 - val_loss: 15.8922 - val_acc: 0.6694\n",
      "Epoch 947/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6115 - acc: 0.6532 - val_loss: 15.8923 - val_acc: 0.6694\n",
      "Epoch 948/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6289 - acc: 0.6636 - val_loss: 15.8923 - val_acc: 0.6694\n",
      "Epoch 949/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6384 - acc: 0.6636 - val_loss: 15.8921 - val_acc: 0.6694\n",
      "Epoch 950/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6334 - acc: 0.6605 - val_loss: 15.8920 - val_acc: 0.6694\n",
      "Epoch 951/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6201 - acc: 0.6563 - val_loss: 15.8919 - val_acc: 0.6694\n",
      "Epoch 952/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6148 - acc: 0.6698 - val_loss: 15.8917 - val_acc: 0.6694\n",
      "Epoch 953/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6188 - acc: 0.6594 - val_loss: 15.8917 - val_acc: 0.6694\n",
      "Epoch 954/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6103 - acc: 0.6636 - val_loss: 15.8916 - val_acc: 0.6694\n",
      "Epoch 955/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6273 - acc: 0.6646 - val_loss: 15.8914 - val_acc: 0.6694\n",
      "Epoch 956/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6137 - acc: 0.6615 - val_loss: 15.8914 - val_acc: 0.6694\n",
      "Epoch 957/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6126 - acc: 0.6667 - val_loss: 15.8913 - val_acc: 0.6694\n",
      "Epoch 958/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6199 - acc: 0.6656 - val_loss: 15.8913 - val_acc: 0.6694\n",
      "Epoch 959/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6311 - acc: 0.6615 - val_loss: 15.8912 - val_acc: 0.6694\n",
      "Epoch 960/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6162 - acc: 0.6594 - val_loss: 15.8912 - val_acc: 0.6694\n",
      "Epoch 961/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6166 - acc: 0.6625 - val_loss: 15.8913 - val_acc: 0.6694\n",
      "Epoch 962/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6039 - acc: 0.6584 - val_loss: 15.8911 - val_acc: 0.6694\n",
      "Epoch 963/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6327 - acc: 0.6615 - val_loss: 15.8910 - val_acc: 0.6694\n",
      "Epoch 964/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6134 - acc: 0.6615 - val_loss: 15.8911 - val_acc: 0.6694\n",
      "Epoch 965/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6281 - acc: 0.6584 - val_loss: 15.8911 - val_acc: 0.6694\n",
      "Epoch 966/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6118 - acc: 0.6553 - val_loss: 15.8912 - val_acc: 0.6694\n",
      "Epoch 967/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6150 - acc: 0.6573 - val_loss: 15.8913 - val_acc: 0.6694\n",
      "Epoch 968/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6238 - acc: 0.6615 - val_loss: 15.8912 - val_acc: 0.6694\n",
      "Epoch 969/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6119 - acc: 0.6605 - val_loss: 15.8912 - val_acc: 0.6694\n",
      "Epoch 970/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6202 - acc: 0.6667 - val_loss: 15.8909 - val_acc: 0.6694\n",
      "Epoch 971/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6289 - acc: 0.6573 - val_loss: 15.8907 - val_acc: 0.6694\n",
      "Epoch 972/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6269 - acc: 0.6615 - val_loss: 15.8907 - val_acc: 0.6694\n",
      "Epoch 973/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6069 - acc: 0.6594 - val_loss: 15.8905 - val_acc: 0.6694\n",
      "Epoch 974/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6240 - acc: 0.6563 - val_loss: 15.8907 - val_acc: 0.6694\n",
      "Epoch 975/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6169 - acc: 0.6625 - val_loss: 15.8907 - val_acc: 0.6694\n",
      "Epoch 976/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6375 - acc: 0.6584 - val_loss: 15.8906 - val_acc: 0.6694\n",
      "Epoch 977/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6409 - acc: 0.6563 - val_loss: 15.8905 - val_acc: 0.6694\n",
      "Epoch 978/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6362 - acc: 0.6615 - val_loss: 15.8903 - val_acc: 0.6694\n",
      "Epoch 979/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6244 - acc: 0.6594 - val_loss: 15.8904 - val_acc: 0.6694\n",
      "Epoch 980/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6302 - acc: 0.6594 - val_loss: 15.8903 - val_acc: 0.6694\n",
      "Epoch 981/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6307 - acc: 0.6667 - val_loss: 15.8902 - val_acc: 0.6694\n",
      "Epoch 982/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6698 - val_loss: 15.8902 - val_acc: 0.6694\n",
      "Epoch 983/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6220 - acc: 0.6542 - val_loss: 15.8901 - val_acc: 0.6694\n",
      "Epoch 984/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6216 - acc: 0.6625 - val_loss: 15.8901 - val_acc: 0.6694\n",
      "Epoch 985/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6231 - acc: 0.6646 - val_loss: 15.8900 - val_acc: 0.6694\n",
      "Epoch 986/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6261 - acc: 0.6636 - val_loss: 15.8900 - val_acc: 0.6694\n",
      "Epoch 987/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6060 - acc: 0.6667 - val_loss: 15.8899 - val_acc: 0.6694\n",
      "Epoch 988/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6117 - acc: 0.6553 - val_loss: 15.8898 - val_acc: 0.6694\n",
      "Epoch 989/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6102 - acc: 0.6615 - val_loss: 15.8899 - val_acc: 0.6694\n",
      "Epoch 990/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6288 - acc: 0.6605 - val_loss: 15.8899 - val_acc: 0.6694\n",
      "Epoch 991/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6160 - acc: 0.6615 - val_loss: 15.8897 - val_acc: 0.6694\n",
      "Epoch 992/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6295 - acc: 0.6667 - val_loss: 15.8896 - val_acc: 0.6694\n",
      "Epoch 993/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6156 - acc: 0.6594 - val_loss: 15.8896 - val_acc: 0.6694\n",
      "Epoch 994/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6046 - acc: 0.6573 - val_loss: 15.8896 - val_acc: 0.6694\n",
      "Epoch 995/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6145 - acc: 0.6625 - val_loss: 15.8898 - val_acc: 0.6694\n",
      "Epoch 996/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6217 - acc: 0.6667 - val_loss: 15.8897 - val_acc: 0.6694\n",
      "Epoch 997/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6221 - acc: 0.6667 - val_loss: 15.8897 - val_acc: 0.6694\n",
      "Epoch 998/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6274 - acc: 0.6615 - val_loss: 15.8898 - val_acc: 0.6694\n",
      "Epoch 999/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6174 - acc: 0.6667 - val_loss: 15.8896 - val_acc: 0.6694\n",
      "Epoch 1000/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6260 - acc: 0.6594 - val_loss: 15.8895 - val_acc: 0.6694\n",
      "Epoch 1001/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6279 - acc: 0.6667 - val_loss: 15.8894 - val_acc: 0.6694\n",
      "Epoch 1002/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6101 - acc: 0.6542 - val_loss: 15.8891 - val_acc: 0.6694\n",
      "Epoch 1003/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6274 - acc: 0.6708 - val_loss: 15.8889 - val_acc: 0.6694\n",
      "Epoch 1004/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6116 - acc: 0.6605 - val_loss: 15.8888 - val_acc: 0.6694\n",
      "Epoch 1005/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6677 - val_loss: 15.8886 - val_acc: 0.6694\n",
      "Epoch 1006/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6249 - acc: 0.6636 - val_loss: 15.8887 - val_acc: 0.6694\n",
      "Epoch 1007/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6102 - acc: 0.6646 - val_loss: 15.8886 - val_acc: 0.6694\n",
      "Epoch 1008/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6266 - acc: 0.6708 - val_loss: 15.8884 - val_acc: 0.6694\n",
      "Epoch 1009/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6122 - acc: 0.6646 - val_loss: 15.8882 - val_acc: 0.6694\n",
      "Epoch 1010/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6252 - acc: 0.6636 - val_loss: 15.8880 - val_acc: 0.6694\n",
      "Epoch 1011/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6160 - acc: 0.6522 - val_loss: 15.8881 - val_acc: 0.6694\n",
      "Epoch 1012/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6440 - acc: 0.6615 - val_loss: 15.8881 - val_acc: 0.6694\n",
      "Epoch 1013/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6208 - acc: 0.6646 - val_loss: 15.8879 - val_acc: 0.6694\n",
      "Epoch 1014/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6399 - acc: 0.6646 - val_loss: 15.8877 - val_acc: 0.6694\n",
      "Epoch 1015/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6307 - acc: 0.6646 - val_loss: 15.8877 - val_acc: 0.6694\n",
      "Epoch 1016/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6267 - acc: 0.6687 - val_loss: 15.8875 - val_acc: 0.6694\n",
      "Epoch 1017/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6347 - acc: 0.6615 - val_loss: 15.8874 - val_acc: 0.6694\n",
      "Epoch 1018/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6584 - val_loss: 15.8873 - val_acc: 0.6694\n",
      "Epoch 1019/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6127 - acc: 0.6656 - val_loss: 15.8872 - val_acc: 0.6694\n",
      "Epoch 1020/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6210 - acc: 0.6656 - val_loss: 15.8872 - val_acc: 0.6694\n",
      "Epoch 1021/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6113 - acc: 0.6656 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1022/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6268 - acc: 0.6687 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1023/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6168 - acc: 0.6563 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1024/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6311 - acc: 0.6646 - val_loss: 15.8872 - val_acc: 0.6694\n",
      "Epoch 1025/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6317 - acc: 0.6584 - val_loss: 15.8873 - val_acc: 0.6694\n",
      "Epoch 1026/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6204 - acc: 0.6625 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1027/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6022 - acc: 0.6625 - val_loss: 15.8872 - val_acc: 0.6694\n",
      "Epoch 1028/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6299 - acc: 0.6625 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1029/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6206 - acc: 0.6553 - val_loss: 15.8871 - val_acc: 0.6694\n",
      "Epoch 1030/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6452 - acc: 0.6605 - val_loss: 15.8868 - val_acc: 0.6694\n",
      "Epoch 1031/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6285 - acc: 0.6563 - val_loss: 15.8869 - val_acc: 0.6694\n",
      "Epoch 1032/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6300 - acc: 0.6625 - val_loss: 15.8867 - val_acc: 0.6694\n",
      "Epoch 1033/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6209 - acc: 0.6563 - val_loss: 15.8867 - val_acc: 0.6694\n",
      "Epoch 1034/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6220 - acc: 0.6615 - val_loss: 15.8866 - val_acc: 0.6694\n",
      "Epoch 1035/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6136 - acc: 0.6656 - val_loss: 15.8866 - val_acc: 0.6694\n",
      "Epoch 1036/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6184 - acc: 0.6667 - val_loss: 15.8864 - val_acc: 0.6694\n",
      "Epoch 1037/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6136 - acc: 0.6625 - val_loss: 15.8863 - val_acc: 0.6694\n",
      "Epoch 1038/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6247 - acc: 0.6656 - val_loss: 15.8862 - val_acc: 0.6694\n",
      "Epoch 1039/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6122 - acc: 0.6636 - val_loss: 15.8863 - val_acc: 0.6694\n",
      "Epoch 1040/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6584 - val_loss: 15.8862 - val_acc: 0.6694\n",
      "Epoch 1041/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6173 - acc: 0.6491 - val_loss: 15.8862 - val_acc: 0.6694\n",
      "Epoch 1042/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6231 - acc: 0.6646 - val_loss: 15.8861 - val_acc: 0.6694\n",
      "Epoch 1043/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6149 - acc: 0.6605 - val_loss: 15.8860 - val_acc: 0.6694\n",
      "Epoch 1044/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6103 - acc: 0.6625 - val_loss: 15.8859 - val_acc: 0.6694\n",
      "Epoch 1045/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6074 - acc: 0.6542 - val_loss: 15.8858 - val_acc: 0.6694\n",
      "Epoch 1046/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6177 - acc: 0.6625 - val_loss: 15.8858 - val_acc: 0.6694\n",
      "Epoch 1047/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6108 - acc: 0.6615 - val_loss: 15.8858 - val_acc: 0.6694\n",
      "Epoch 1048/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6148 - acc: 0.6687 - val_loss: 15.8858 - val_acc: 0.6694\n",
      "Epoch 1049/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6203 - acc: 0.6615 - val_loss: 15.8857 - val_acc: 0.6694\n",
      "Epoch 1050/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6207 - acc: 0.6625 - val_loss: 15.8857 - val_acc: 0.6694\n",
      "Epoch 1051/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6316 - acc: 0.6605 - val_loss: 15.8857 - val_acc: 0.6694\n",
      "Epoch 1052/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6243 - acc: 0.6636 - val_loss: 15.8856 - val_acc: 0.6694\n",
      "Epoch 1053/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6298 - acc: 0.6625 - val_loss: 15.8856 - val_acc: 0.6694\n",
      "Epoch 1054/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6052 - acc: 0.6605 - val_loss: 15.8855 - val_acc: 0.6694\n",
      "Epoch 1055/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6018 - acc: 0.6584 - val_loss: 15.8855 - val_acc: 0.6694\n",
      "Epoch 1056/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6371 - acc: 0.6646 - val_loss: 15.8855 - val_acc: 0.6694\n",
      "Epoch 1057/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6278 - acc: 0.6584 - val_loss: 15.8856 - val_acc: 0.6694\n",
      "Epoch 1058/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6020 - acc: 0.6677 - val_loss: 15.8855 - val_acc: 0.6694\n",
      "Epoch 1059/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6201 - acc: 0.6594 - val_loss: 15.8853 - val_acc: 0.6694\n",
      "Epoch 1060/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6179 - acc: 0.6584 - val_loss: 15.8854 - val_acc: 0.6694\n",
      "Epoch 1061/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6076 - acc: 0.6605 - val_loss: 15.8853 - val_acc: 0.6694\n",
      "Epoch 1062/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6311 - acc: 0.6656 - val_loss: 15.8852 - val_acc: 0.6694\n",
      "Epoch 1063/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6263 - acc: 0.6646 - val_loss: 15.8853 - val_acc: 0.6694\n",
      "Epoch 1064/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6219 - acc: 0.6615 - val_loss: 15.8852 - val_acc: 0.6694\n",
      "Epoch 1065/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6174 - acc: 0.6677 - val_loss: 15.8851 - val_acc: 0.6694\n",
      "Epoch 1066/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6059 - acc: 0.6615 - val_loss: 15.8851 - val_acc: 0.6694\n",
      "Epoch 1067/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6152 - acc: 0.6636 - val_loss: 15.8847 - val_acc: 0.6694\n",
      "Epoch 1068/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6139 - acc: 0.6625 - val_loss: 15.8846 - val_acc: 0.6694\n",
      "Epoch 1069/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6103 - acc: 0.6636 - val_loss: 15.8847 - val_acc: 0.6694\n",
      "Epoch 1070/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6267 - acc: 0.6625 - val_loss: 15.8845 - val_acc: 0.6694\n",
      "Epoch 1071/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6230 - acc: 0.6563 - val_loss: 15.8844 - val_acc: 0.6694\n",
      "Epoch 1072/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6196 - acc: 0.6573 - val_loss: 15.8843 - val_acc: 0.6694\n",
      "Epoch 1073/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6109 - acc: 0.6646 - val_loss: 15.8841 - val_acc: 0.6694\n",
      "Epoch 1074/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6055 - acc: 0.6615 - val_loss: 15.8840 - val_acc: 0.6694\n",
      "Epoch 1075/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6263 - acc: 0.6667 - val_loss: 15.8839 - val_acc: 0.6694\n",
      "Epoch 1076/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6057 - acc: 0.6594 - val_loss: 15.8839 - val_acc: 0.6694\n",
      "Epoch 1077/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6135 - acc: 0.6656 - val_loss: 15.8840 - val_acc: 0.6694\n",
      "Epoch 1078/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6151 - acc: 0.6573 - val_loss: 15.8839 - val_acc: 0.6694\n",
      "Epoch 1079/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6098 - acc: 0.6687 - val_loss: 15.8839 - val_acc: 0.6694\n",
      "Epoch 1080/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6254 - acc: 0.6573 - val_loss: 15.8838 - val_acc: 0.6694\n",
      "Epoch 1081/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6116 - acc: 0.6656 - val_loss: 15.8837 - val_acc: 0.6694\n",
      "Epoch 1082/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6164 - acc: 0.6636 - val_loss: 15.8837 - val_acc: 0.6694\n",
      "Epoch 1083/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6128 - acc: 0.6563 - val_loss: 15.8838 - val_acc: 0.6694\n",
      "Epoch 1084/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6274 - acc: 0.6563 - val_loss: 15.8838 - val_acc: 0.6694\n",
      "Epoch 1085/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6109 - acc: 0.6615 - val_loss: 15.8836 - val_acc: 0.6694\n",
      "Epoch 1086/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6109 - acc: 0.6625 - val_loss: 15.8835 - val_acc: 0.6694\n",
      "Epoch 1087/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6084 - acc: 0.6553 - val_loss: 15.8834 - val_acc: 0.6694\n",
      "Epoch 1088/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6128 - acc: 0.6625 - val_loss: 15.8833 - val_acc: 0.6694\n",
      "Epoch 1089/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6237 - acc: 0.6625 - val_loss: 15.8834 - val_acc: 0.6694\n",
      "Epoch 1090/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6156 - acc: 0.6718 - val_loss: 15.8833 - val_acc: 0.6694\n",
      "Epoch 1091/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6124 - acc: 0.6667 - val_loss: 15.8832 - val_acc: 0.6694\n",
      "Epoch 1092/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6625 - val_loss: 15.8832 - val_acc: 0.6694\n",
      "Epoch 1093/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6081 - acc: 0.6573 - val_loss: 15.8832 - val_acc: 0.6694\n",
      "Epoch 1094/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6115 - acc: 0.6594 - val_loss: 15.8832 - val_acc: 0.6694\n",
      "Epoch 1095/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.5972 - acc: 0.66 - 0s - loss: 16.6085 - acc: 0.6656 - val_loss: 15.8831 - val_acc: 0.6694\n",
      "Epoch 1096/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6245 - acc: 0.6625 - val_loss: 15.8830 - val_acc: 0.6694\n",
      "Epoch 1097/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6181 - acc: 0.6594 - val_loss: 15.8830 - val_acc: 0.6694\n",
      "Epoch 1098/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6247 - acc: 0.6605 - val_loss: 15.8829 - val_acc: 0.6694\n",
      "Epoch 1099/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6181 - acc: 0.6594 - val_loss: 15.8826 - val_acc: 0.6694\n",
      "Epoch 1100/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6087 - acc: 0.6563 - val_loss: 15.8826 - val_acc: 0.6694\n",
      "Epoch 1101/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6046 - acc: 0.6646 - val_loss: 15.8826 - val_acc: 0.6694\n",
      "Epoch 1102/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6090 - acc: 0.6594 - val_loss: 15.8825 - val_acc: 0.6694\n",
      "Epoch 1103/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5980 - acc: 0.6573 - val_loss: 15.8824 - val_acc: 0.6694\n",
      "Epoch 1104/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6163 - acc: 0.6636 - val_loss: 15.8822 - val_acc: 0.6694\n",
      "Epoch 1105/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6271 - acc: 0.6625 - val_loss: 15.8822 - val_acc: 0.6694\n",
      "Epoch 1106/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6128 - acc: 0.6677 - val_loss: 15.8822 - val_acc: 0.6694\n",
      "Epoch 1107/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6277 - acc: 0.6677 - val_loss: 15.8822 - val_acc: 0.6694\n",
      "Epoch 1108/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6034 - acc: 0.6625 - val_loss: 15.8821 - val_acc: 0.6694\n",
      "Epoch 1109/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6318 - acc: 0.6656 - val_loss: 15.8820 - val_acc: 0.6694\n",
      "Epoch 1110/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6213 - acc: 0.6687 - val_loss: 15.8822 - val_acc: 0.6694\n",
      "Epoch 1111/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6413 - acc: 0.6605 - val_loss: 15.8821 - val_acc: 0.6694\n",
      "Epoch 1112/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6193 - acc: 0.6605 - val_loss: 15.8821 - val_acc: 0.6694\n",
      "Epoch 1113/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6287 - acc: 0.6573 - val_loss: 15.8820 - val_acc: 0.6694\n",
      "Epoch 1114/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6667 - val_loss: 15.8819 - val_acc: 0.6694\n",
      "Epoch 1115/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6625 - val_loss: 15.8819 - val_acc: 0.6694\n",
      "Epoch 1116/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6123 - acc: 0.6584 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1117/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6079 - acc: 0.6718 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1118/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6216 - acc: 0.6605 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1119/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5994 - acc: 0.6573 - val_loss: 15.8815 - val_acc: 0.6694\n",
      "Epoch 1120/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5970 - acc: 0.6594 - val_loss: 15.8815 - val_acc: 0.6694\n",
      "Epoch 1121/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6176 - acc: 0.6636 - val_loss: 15.8815 - val_acc: 0.6694\n",
      "Epoch 1122/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6636 - val_loss: 15.8816 - val_acc: 0.6694\n",
      "Epoch 1123/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6064 - acc: 0.6625 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1124/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6056 - acc: 0.6532 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1125/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6011 - acc: 0.6615 - val_loss: 15.8818 - val_acc: 0.6694\n",
      "Epoch 1126/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6155 - acc: 0.6605 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1127/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5989 - acc: 0.6553 - val_loss: 15.8818 - val_acc: 0.6694\n",
      "Epoch 1128/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6687 - val_loss: 15.8818 - val_acc: 0.6694\n",
      "Epoch 1129/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6044 - acc: 0.6667 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1130/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6584 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1131/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6141 - acc: 0.6584 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1132/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6667 - val_loss: 15.8817 - val_acc: 0.6694\n",
      "Epoch 1133/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6157 - acc: 0.6636 - val_loss: 15.8814 - val_acc: 0.6694\n",
      "Epoch 1134/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6183 - acc: 0.6584 - val_loss: 15.8814 - val_acc: 0.6694\n",
      "Epoch 1135/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6274 - acc: 0.6646 - val_loss: 15.8812 - val_acc: 0.6694\n",
      "Epoch 1136/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6168 - acc: 0.6636 - val_loss: 15.8812 - val_acc: 0.6694\n",
      "Epoch 1137/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6124 - acc: 0.6625 - val_loss: 15.8811 - val_acc: 0.6694\n",
      "Epoch 1138/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6075 - acc: 0.6636 - val_loss: 15.8808 - val_acc: 0.6694\n",
      "Epoch 1139/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6140 - acc: 0.6646 - val_loss: 15.8808 - val_acc: 0.6694\n",
      "Epoch 1140/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6033 - acc: 0.6636 - val_loss: 15.8807 - val_acc: 0.6694\n",
      "Epoch 1141/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6070 - acc: 0.6563 - val_loss: 15.8807 - val_acc: 0.6694\n",
      "Epoch 1142/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6096 - acc: 0.6522 - val_loss: 15.8808 - val_acc: 0.6694\n",
      "Epoch 1143/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6136 - acc: 0.6594 - val_loss: 15.8807 - val_acc: 0.6694\n",
      "Epoch 1144/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6112 - acc: 0.6573 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1145/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6230 - acc: 0.6563 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1146/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6058 - acc: 0.6646 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1147/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6198 - acc: 0.6584 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1148/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6150 - acc: 0.6667 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1149/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6210 - acc: 0.6625 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1150/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6158 - acc: 0.6646 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1151/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6108 - acc: 0.6594 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1152/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6187 - acc: 0.6667 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1153/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6158 - acc: 0.6573 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1154/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6147 - acc: 0.6636 - val_loss: 15.8806 - val_acc: 0.6694\n",
      "Epoch 1155/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6310 - acc: 0.6698 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1156/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6553 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1157/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6097 - acc: 0.6605 - val_loss: 15.8804 - val_acc: 0.6694\n",
      "Epoch 1158/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6594 - val_loss: 15.8803 - val_acc: 0.6694\n",
      "Epoch 1159/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6283 - acc: 0.6615 - val_loss: 15.8804 - val_acc: 0.6694\n",
      "Epoch 1160/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6197 - acc: 0.6677 - val_loss: 15.8803 - val_acc: 0.6694\n",
      "Epoch 1161/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6301 - acc: 0.6573 - val_loss: 15.8804 - val_acc: 0.6694\n",
      "Epoch 1162/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6151 - acc: 0.6656 - val_loss: 15.8805 - val_acc: 0.6694\n",
      "Epoch 1163/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5978 - acc: 0.6625 - val_loss: 15.8804 - val_acc: 0.6694\n",
      "Epoch 1164/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6646 - val_loss: 15.8803 - val_acc: 0.6694\n",
      "Epoch 1165/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5934 - acc: 0.6615 - val_loss: 15.8803 - val_acc: 0.6694\n",
      "Epoch 1166/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6172 - acc: 0.6605 - val_loss: 15.8802 - val_acc: 0.6694\n",
      "Epoch 1167/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6111 - acc: 0.6522 - val_loss: 15.8801 - val_acc: 0.6694\n",
      "Epoch 1168/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6625 - val_loss: 15.8801 - val_acc: 0.6694\n",
      "Epoch 1169/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6625 - val_loss: 15.8800 - val_acc: 0.6694\n",
      "Epoch 1170/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6319 - acc: 0.6677 - val_loss: 15.8800 - val_acc: 0.6694\n",
      "Epoch 1171/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6124 - acc: 0.6636 - val_loss: 15.8801 - val_acc: 0.6694\n",
      "Epoch 1172/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6203 - acc: 0.6646 - val_loss: 15.8800 - val_acc: 0.6694\n",
      "Epoch 1173/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6010 - acc: 0.6615 - val_loss: 15.8799 - val_acc: 0.6694\n",
      "Epoch 1174/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6310 - acc: 0.6522 - val_loss: 15.8799 - val_acc: 0.6694\n",
      "Epoch 1175/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5948 - acc: 0.6687 - val_loss: 15.8799 - val_acc: 0.6694\n",
      "Epoch 1176/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6615 - val_loss: 15.8798 - val_acc: 0.6694\n",
      "Epoch 1177/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6060 - acc: 0.6553 - val_loss: 15.8797 - val_acc: 0.6694\n",
      "Epoch 1178/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6170 - acc: 0.6656 - val_loss: 15.8797 - val_acc: 0.6694\n",
      "Epoch 1179/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6188 - acc: 0.6646 - val_loss: 15.8797 - val_acc: 0.6694\n",
      "Epoch 1180/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6584 - val_loss: 15.8796 - val_acc: 0.6694\n",
      "Epoch 1181/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6315 - acc: 0.6646 - val_loss: 15.8794 - val_acc: 0.6694\n",
      "Epoch 1182/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6051 - acc: 0.6615 - val_loss: 15.8795 - val_acc: 0.6694\n",
      "Epoch 1183/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6113 - acc: 0.6656 - val_loss: 15.8793 - val_acc: 0.6694\n",
      "Epoch 1184/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6127 - acc: 0.6667 - val_loss: 15.8793 - val_acc: 0.6694\n",
      "Epoch 1185/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6605 - val_loss: 15.8792 - val_acc: 0.6694\n",
      "Epoch 1186/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6281 - acc: 0.6625 - val_loss: 15.8791 - val_acc: 0.6694\n",
      "Epoch 1187/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6188 - acc: 0.6646 - val_loss: 15.8791 - val_acc: 0.6694\n",
      "Epoch 1188/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6111 - acc: 0.6615 - val_loss: 15.8790 - val_acc: 0.6694\n",
      "Epoch 1189/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6153 - acc: 0.6677 - val_loss: 15.8790 - val_acc: 0.6694\n",
      "Epoch 1190/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6270 - acc: 0.6584 - val_loss: 15.8790 - val_acc: 0.6694\n",
      "Epoch 1191/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6140 - acc: 0.6667 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1192/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6220 - acc: 0.6584 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1193/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6150 - acc: 0.6594 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1194/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6145 - acc: 0.6594 - val_loss: 15.8790 - val_acc: 0.6694\n",
      "Epoch 1195/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6063 - acc: 0.6615 - val_loss: 15.8790 - val_acc: 0.6694\n",
      "Epoch 1196/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6136 - acc: 0.6553 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1197/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6245 - acc: 0.6573 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1198/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5930 - acc: 0.6646 - val_loss: 15.8789 - val_acc: 0.6694\n",
      "Epoch 1199/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6553 - val_loss: 15.8788 - val_acc: 0.6694\n",
      "Epoch 1200/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6563 - val_loss: 15.8788 - val_acc: 0.6694\n",
      "Epoch 1201/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6168 - acc: 0.6511 - val_loss: 15.8787 - val_acc: 0.6694\n",
      "Epoch 1202/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6625 - val_loss: 15.8787 - val_acc: 0.6736\n",
      "Epoch 1203/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5998 - acc: 0.6615 - val_loss: 15.8788 - val_acc: 0.6736\n",
      "Epoch 1204/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6187 - acc: 0.6584 - val_loss: 15.8788 - val_acc: 0.6736\n",
      "Epoch 1205/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6153 - acc: 0.6605 - val_loss: 15.8787 - val_acc: 0.6736\n",
      "Epoch 1206/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6021 - acc: 0.6573 - val_loss: 15.8786 - val_acc: 0.6736\n",
      "Epoch 1207/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5974 - acc: 0.6584 - val_loss: 15.8786 - val_acc: 0.6736\n",
      "Epoch 1208/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6043 - acc: 0.6687 - val_loss: 15.8785 - val_acc: 0.6736\n",
      "Epoch 1209/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6002 - acc: 0.6646 - val_loss: 15.8784 - val_acc: 0.6736\n",
      "Epoch 1210/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6156 - acc: 0.6656 - val_loss: 15.8783 - val_acc: 0.6736\n",
      "Epoch 1211/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6646 - val_loss: 15.8783 - val_acc: 0.6736\n",
      "Epoch 1212/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6218 - acc: 0.6594 - val_loss: 15.8784 - val_acc: 0.6736\n",
      "Epoch 1213/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6594 - val_loss: 15.8782 - val_acc: 0.6736\n",
      "Epoch 1214/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6256 - acc: 0.6605 - val_loss: 15.8782 - val_acc: 0.6736\n",
      "Epoch 1215/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6187 - acc: 0.6594 - val_loss: 15.8782 - val_acc: 0.6736\n",
      "Epoch 1216/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6146 - acc: 0.6698 - val_loss: 15.8781 - val_acc: 0.6736\n",
      "Epoch 1217/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6584 - val_loss: 15.8782 - val_acc: 0.6736\n",
      "Epoch 1218/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6081 - acc: 0.6677 - val_loss: 15.8780 - val_acc: 0.6736\n",
      "Epoch 1219/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6133 - acc: 0.6605 - val_loss: 15.8779 - val_acc: 0.6736\n",
      "Epoch 1220/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6184 - acc: 0.6563 - val_loss: 15.8780 - val_acc: 0.6736\n",
      "Epoch 1221/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5940 - acc: 0.6625 - val_loss: 15.8781 - val_acc: 0.6736\n",
      "Epoch 1222/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6646 - val_loss: 15.8781 - val_acc: 0.6736\n",
      "Epoch 1223/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6046 - acc: 0.6615 - val_loss: 15.8779 - val_acc: 0.6736\n",
      "Epoch 1224/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6089 - acc: 0.6542 - val_loss: 15.8778 - val_acc: 0.6736\n",
      "Epoch 1225/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6066 - acc: 0.6605 - val_loss: 15.8777 - val_acc: 0.6736\n",
      "Epoch 1226/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6098 - acc: 0.6656 - val_loss: 15.8777 - val_acc: 0.6736\n",
      "Epoch 1227/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6154 - acc: 0.6584 - val_loss: 15.8775 - val_acc: 0.6736\n",
      "Epoch 1228/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6605 - val_loss: 15.8775 - val_acc: 0.6736\n",
      "Epoch 1229/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6039 - acc: 0.6656 - val_loss: 15.8774 - val_acc: 0.6736\n",
      "Epoch 1230/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5901 - acc: 0.6563 - val_loss: 15.8774 - val_acc: 0.6736\n",
      "Epoch 1231/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6118 - acc: 0.6656 - val_loss: 15.8775 - val_acc: 0.6736\n",
      "Epoch 1232/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6067 - acc: 0.6646 - val_loss: 15.8774 - val_acc: 0.6736\n",
      "Epoch 1233/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6066 - acc: 0.6553 - val_loss: 15.8774 - val_acc: 0.6736\n",
      "Epoch 1234/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6074 - acc: 0.6656 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1235/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6065 - acc: 0.6667 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1236/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6024 - acc: 0.6636 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1237/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6087 - acc: 0.6625 - val_loss: 15.8772 - val_acc: 0.6736\n",
      "Epoch 1238/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6011 - acc: 0.6625 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1239/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6119 - acc: 0.6636 - val_loss: 15.8772 - val_acc: 0.6736\n",
      "Epoch 1240/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6211 - acc: 0.6573 - val_loss: 15.8772 - val_acc: 0.6736\n",
      "Epoch 1241/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6573 - val_loss: 15.8774 - val_acc: 0.6736\n",
      "Epoch 1242/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6085 - acc: 0.6605 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1243/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6092 - acc: 0.6605 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1244/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5985 - acc: 0.6584 - val_loss: 15.8773 - val_acc: 0.6736\n",
      "Epoch 1245/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6010 - acc: 0.6718 - val_loss: 15.8771 - val_acc: 0.6736\n",
      "Epoch 1246/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6292 - acc: 0.6615 - val_loss: 15.8770 - val_acc: 0.6736\n",
      "Epoch 1247/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6040 - acc: 0.6553 - val_loss: 15.8771 - val_acc: 0.6736\n",
      "Epoch 1248/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6051 - acc: 0.6636 - val_loss: 15.8770 - val_acc: 0.6736\n",
      "Epoch 1249/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6113 - acc: 0.6687 - val_loss: 15.8768 - val_acc: 0.6736\n",
      "Epoch 1250/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6008 - acc: 0.6605 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1251/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6033 - acc: 0.6677 - val_loss: 15.8768 - val_acc: 0.6736\n",
      "Epoch 1252/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6083 - acc: 0.6553 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1253/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6605 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1254/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6138 - acc: 0.6584 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1255/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6625 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1256/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6667 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1257/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6010 - acc: 0.6605 - val_loss: 15.8768 - val_acc: 0.6736\n",
      "Epoch 1258/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6045 - acc: 0.6605 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1259/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6207 - acc: 0.6563 - val_loss: 15.8767 - val_acc: 0.6736\n",
      "Epoch 1260/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6189 - acc: 0.6511 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1261/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6240 - acc: 0.6656 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1262/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6272 - acc: 0.6573 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1263/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6205 - acc: 0.6615 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1264/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6677 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1265/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6077 - acc: 0.6563 - val_loss: 15.8764 - val_acc: 0.6736\n",
      "Epoch 1266/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6584 - val_loss: 15.8764 - val_acc: 0.6736\n",
      "Epoch 1267/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6152 - acc: 0.6605 - val_loss: 15.8764 - val_acc: 0.6736\n",
      "Epoch 1268/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6594 - val_loss: 15.8764 - val_acc: 0.6736\n",
      "Epoch 1269/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6011 - acc: 0.6656 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1270/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6625 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1271/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6058 - acc: 0.6563 - val_loss: 15.8766 - val_acc: 0.6736\n",
      "Epoch 1272/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6114 - acc: 0.6636 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1273/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6230 - acc: 0.6553 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1274/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6149 - acc: 0.6584 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1275/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6105 - acc: 0.6677 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1276/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6708 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1277/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6636 - val_loss: 15.8765 - val_acc: 0.6736\n",
      "Epoch 1278/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6677 - val_loss: 15.8762 - val_acc: 0.6736\n",
      "Epoch 1279/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6563 - val_loss: 15.8762 - val_acc: 0.6736\n",
      "Epoch 1280/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6266 - acc: 0.6563 - val_loss: 15.8760 - val_acc: 0.6736\n",
      "Epoch 1281/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6137 - acc: 0.6605 - val_loss: 15.8759 - val_acc: 0.6736\n",
      "Epoch 1282/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5951 - acc: 0.6522 - val_loss: 15.8759 - val_acc: 0.6736\n",
      "Epoch 1283/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6625 - val_loss: 15.8759 - val_acc: 0.6736\n",
      "Epoch 1284/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6006 - acc: 0.6636 - val_loss: 15.8759 - val_acc: 0.6736\n",
      "Epoch 1285/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5924 - acc: 0.6687 - val_loss: 15.8758 - val_acc: 0.6736\n",
      "Epoch 1286/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6594 - val_loss: 15.8756 - val_acc: 0.6736\n",
      "Epoch 1287/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6075 - acc: 0.6532 - val_loss: 15.8756 - val_acc: 0.6736\n",
      "Epoch 1288/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6625 - val_loss: 15.8757 - val_acc: 0.6736\n",
      "Epoch 1289/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6563 - val_loss: 15.8755 - val_acc: 0.6736\n",
      "Epoch 1290/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5967 - acc: 0.6615 - val_loss: 15.8754 - val_acc: 0.6736\n",
      "Epoch 1291/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6176 - acc: 0.6584 - val_loss: 15.8753 - val_acc: 0.6736\n",
      "Epoch 1292/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6039 - acc: 0.6656 - val_loss: 15.8752 - val_acc: 0.6736\n",
      "Epoch 1293/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6594 - val_loss: 15.8751 - val_acc: 0.6736\n",
      "Epoch 1294/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6189 - acc: 0.6605 - val_loss: 15.8749 - val_acc: 0.6736\n",
      "Epoch 1295/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6007 - acc: 0.6625 - val_loss: 15.8749 - val_acc: 0.6736\n",
      "Epoch 1296/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6584 - val_loss: 15.8747 - val_acc: 0.6736\n",
      "Epoch 1297/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6127 - acc: 0.6605 - val_loss: 15.8747 - val_acc: 0.6736\n",
      "Epoch 1298/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6038 - acc: 0.6615 - val_loss: 15.8746 - val_acc: 0.6736\n",
      "Epoch 1299/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6096 - acc: 0.6615 - val_loss: 15.8744 - val_acc: 0.6736\n",
      "Epoch 1300/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6239 - acc: 0.6605 - val_loss: 15.8744 - val_acc: 0.6736\n",
      "Epoch 1301/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6163 - acc: 0.6563 - val_loss: 15.8743 - val_acc: 0.6736\n",
      "Epoch 1302/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6018 - acc: 0.6708 - val_loss: 15.8741 - val_acc: 0.6736\n",
      "Epoch 1303/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6007 - acc: 0.6667 - val_loss: 15.8740 - val_acc: 0.6736\n",
      "Epoch 1304/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6046 - acc: 0.6636 - val_loss: 15.8739 - val_acc: 0.6736\n",
      "Epoch 1305/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6636 - val_loss: 15.8739 - val_acc: 0.6736\n",
      "Epoch 1306/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6155 - acc: 0.6625 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1307/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6105 - acc: 0.6636 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1308/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6698 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1309/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6296 - acc: 0.6646 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1310/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6698 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1311/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6157 - acc: 0.6594 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1312/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6096 - acc: 0.6573 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1313/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6625 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1314/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5955 - acc: 0.6687 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1315/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6039 - acc: 0.6636 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1316/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6220 - acc: 0.6584 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1317/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6636 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1318/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6102 - acc: 0.6605 - val_loss: 15.8739 - val_acc: 0.6736\n",
      "Epoch 1319/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6003 - acc: 0.6625 - val_loss: 15.8739 - val_acc: 0.6736\n",
      "Epoch 1320/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6167 - acc: 0.6605 - val_loss: 15.8739 - val_acc: 0.6736\n",
      "Epoch 1321/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6656 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1322/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6573 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1323/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6017 - acc: 0.6625 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1324/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6090 - acc: 0.6636 - val_loss: 15.8738 - val_acc: 0.6736\n",
      "Epoch 1325/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5969 - acc: 0.6615 - val_loss: 15.8737 - val_acc: 0.6736\n",
      "Epoch 1326/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6076 - acc: 0.6656 - val_loss: 15.8736 - val_acc: 0.6736\n",
      "Epoch 1327/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6256 - acc: 0.6667 - val_loss: 15.8735 - val_acc: 0.6736\n",
      "Epoch 1328/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6155 - acc: 0.6677 - val_loss: 15.8734 - val_acc: 0.6736\n",
      "Epoch 1329/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6071 - acc: 0.6594 - val_loss: 15.8733 - val_acc: 0.6736\n",
      "Epoch 1330/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6135 - acc: 0.6636 - val_loss: 15.8732 - val_acc: 0.6736\n",
      "Epoch 1331/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6150 - acc: 0.6636 - val_loss: 15.8732 - val_acc: 0.6736\n",
      "Epoch 1332/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6231 - acc: 0.6594 - val_loss: 15.8731 - val_acc: 0.6736\n",
      "Epoch 1333/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6103 - acc: 0.6532 - val_loss: 15.8731 - val_acc: 0.6736\n",
      "Epoch 1334/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6066 - acc: 0.6625 - val_loss: 15.8733 - val_acc: 0.6736\n",
      "Epoch 1335/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5970 - acc: 0.6646 - val_loss: 15.8733 - val_acc: 0.6736\n",
      "Epoch 1336/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6014 - acc: 0.6656 - val_loss: 15.8732 - val_acc: 0.6736\n",
      "Epoch 1337/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6667 - val_loss: 15.8732 - val_acc: 0.6736\n",
      "Epoch 1338/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6573 - val_loss: 15.8731 - val_acc: 0.6736\n",
      "Epoch 1339/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6511 - val_loss: 15.8730 - val_acc: 0.6736\n",
      "Epoch 1340/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6044 - acc: 0.6646 - val_loss: 15.8730 - val_acc: 0.6736\n",
      "Epoch 1341/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6217 - acc: 0.6625 - val_loss: 15.8729 - val_acc: 0.6736\n",
      "Epoch 1342/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6553 - val_loss: 15.8729 - val_acc: 0.6736\n",
      "Epoch 1343/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6163 - acc: 0.6615 - val_loss: 15.8729 - val_acc: 0.6736\n",
      "Epoch 1344/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5990 - acc: 0.6563 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1345/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6594 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1346/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6137 - acc: 0.6677 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1347/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6594 - val_loss: 15.8729 - val_acc: 0.6736\n",
      "Epoch 1348/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6594 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1349/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6615 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1350/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6067 - acc: 0.6646 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1351/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5977 - acc: 0.6605 - val_loss: 15.8727 - val_acc: 0.6736\n",
      "Epoch 1352/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6605 - val_loss: 15.8726 - val_acc: 0.6736\n",
      "Epoch 1353/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6563 - val_loss: 15.8727 - val_acc: 0.6736\n",
      "Epoch 1354/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6563 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1355/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6625 - val_loss: 15.8728 - val_acc: 0.6736\n",
      "Epoch 1356/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6094 - acc: 0.6636 - val_loss: 15.8727 - val_acc: 0.6736\n",
      "Epoch 1357/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6213 - acc: 0.6636 - val_loss: 15.8727 - val_acc: 0.6736\n",
      "Epoch 1358/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6594 - val_loss: 15.8726 - val_acc: 0.6736\n",
      "Epoch 1359/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6708 - val_loss: 15.8726 - val_acc: 0.6736\n",
      "Epoch 1360/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6656 - val_loss: 15.8724 - val_acc: 0.6736\n",
      "Epoch 1361/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5963 - acc: 0.6636 - val_loss: 15.8724 - val_acc: 0.6736\n",
      "Epoch 1362/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6677 - val_loss: 15.8724 - val_acc: 0.6736\n",
      "Epoch 1363/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6020 - acc: 0.6656 - val_loss: 15.8723 - val_acc: 0.6736\n",
      "Epoch 1364/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6038 - acc: 0.6615 - val_loss: 15.8723 - val_acc: 0.6736\n",
      "Epoch 1365/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6667 - val_loss: 15.8723 - val_acc: 0.6736\n",
      "Epoch 1366/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6060 - acc: 0.6615 - val_loss: 15.8722 - val_acc: 0.6736\n",
      "Epoch 1367/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6065 - acc: 0.6656 - val_loss: 15.8722 - val_acc: 0.6736\n",
      "Epoch 1368/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6015 - acc: 0.6553 - val_loss: 15.8722 - val_acc: 0.6736\n",
      "Epoch 1369/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6594 - val_loss: 15.8722 - val_acc: 0.6736\n",
      "Epoch 1370/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5999 - acc: 0.6625 - val_loss: 15.8720 - val_acc: 0.6736\n",
      "Epoch 1371/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6292 - acc: 0.6553 - val_loss: 15.8720 - val_acc: 0.6736\n",
      "Epoch 1372/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5999 - acc: 0.6584 - val_loss: 15.8719 - val_acc: 0.6736\n",
      "Epoch 1373/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6038 - acc: 0.6511 - val_loss: 15.8718 - val_acc: 0.6736\n",
      "Epoch 1374/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6006 - acc: 0.6625 - val_loss: 15.8717 - val_acc: 0.6736\n",
      "Epoch 1375/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6687 - val_loss: 15.8717 - val_acc: 0.6736\n",
      "Epoch 1376/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5988 - acc: 0.6615 - val_loss: 15.8718 - val_acc: 0.6736\n",
      "Epoch 1377/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6206 - acc: 0.6605 - val_loss: 15.8718 - val_acc: 0.6736\n",
      "Epoch 1378/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6146 - acc: 0.6646 - val_loss: 15.8717 - val_acc: 0.6736\n",
      "Epoch 1379/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5973 - acc: 0.6563 - val_loss: 15.8716 - val_acc: 0.6736\n",
      "Epoch 1380/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6129 - acc: 0.6646 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1381/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6594 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1382/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6615 - val_loss: 15.8713 - val_acc: 0.6736\n",
      "Epoch 1383/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6139 - acc: 0.6584 - val_loss: 15.8713 - val_acc: 0.6736\n",
      "Epoch 1384/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5981 - acc: 0.6563 - val_loss: 15.8713 - val_acc: 0.6736\n",
      "Epoch 1385/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6573 - val_loss: 15.8714 - val_acc: 0.6736\n",
      "Epoch 1386/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6646 - val_loss: 15.8713 - val_acc: 0.6736\n",
      "Epoch 1387/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5976 - acc: 0.6615 - val_loss: 15.8714 - val_acc: 0.6736\n",
      "Epoch 1388/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5971 - acc: 0.6636 - val_loss: 15.8714 - val_acc: 0.6736\n",
      "Epoch 1389/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6636 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1390/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5931 - acc: 0.6656 - val_loss: 15.8714 - val_acc: 0.6736\n",
      "Epoch 1391/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6047 - acc: 0.6625 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1392/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5956 - acc: 0.6667 - val_loss: 15.8716 - val_acc: 0.6736\n",
      "Epoch 1393/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6004 - acc: 0.6553 - val_loss: 15.8716 - val_acc: 0.6736\n",
      "Epoch 1394/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6168 - acc: 0.6605 - val_loss: 15.8716 - val_acc: 0.6736\n",
      "Epoch 1395/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5984 - acc: 0.6625 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1396/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6646 - val_loss: 15.8716 - val_acc: 0.6736\n",
      "Epoch 1397/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6677 - val_loss: 15.8715 - val_acc: 0.6736\n",
      "Epoch 1398/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6636 - val_loss: 15.8714 - val_acc: 0.6736\n",
      "Epoch 1399/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6083 - acc: 0.6636 - val_loss: 15.8711 - val_acc: 0.6736\n",
      "Epoch 1400/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6084 - acc: 0.6646 - val_loss: 15.8711 - val_acc: 0.6736\n",
      "Epoch 1401/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6656 - val_loss: 15.8711 - val_acc: 0.6736\n",
      "Epoch 1402/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5955 - acc: 0.6563 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1403/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6178 - acc: 0.6594 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1404/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6636 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1405/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6113 - acc: 0.6656 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1406/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6059 - acc: 0.6584 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1407/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6138 - acc: 0.6677 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1408/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6116 - acc: 0.6636 - val_loss: 15.8708 - val_acc: 0.6736\n",
      "Epoch 1409/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6109 - acc: 0.6636 - val_loss: 15.8708 - val_acc: 0.6736\n",
      "Epoch 1410/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6584 - val_loss: 15.8708 - val_acc: 0.6736\n",
      "Epoch 1411/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5978 - acc: 0.6677 - val_loss: 15.8707 - val_acc: 0.6736\n",
      "Epoch 1412/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6594 - val_loss: 15.8709 - val_acc: 0.6736\n",
      "Epoch 1413/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5977 - acc: 0.6646 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1414/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5987 - acc: 0.6563 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1415/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6625 - val_loss: 15.8709 - val_acc: 0.6736\n",
      "Epoch 1416/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6625 - val_loss: 15.8709 - val_acc: 0.6736\n",
      "Epoch 1417/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6656 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1418/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6061 - acc: 0.6594 - val_loss: 15.8710 - val_acc: 0.6736\n",
      "Epoch 1419/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6127 - acc: 0.6615 - val_loss: 15.8709 - val_acc: 0.6736\n",
      "Epoch 1420/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6687 - val_loss: 15.8707 - val_acc: 0.6736\n",
      "Epoch 1421/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5981 - acc: 0.6594 - val_loss: 15.8707 - val_acc: 0.6736\n",
      "Epoch 1422/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6162 - acc: 0.6573 - val_loss: 15.8706 - val_acc: 0.6736\n",
      "Epoch 1423/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6636 - val_loss: 15.8706 - val_acc: 0.6736\n",
      "Epoch 1424/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6615 - val_loss: 15.8704 - val_acc: 0.6736\n",
      "Epoch 1425/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6004 - acc: 0.6522 - val_loss: 15.8704 - val_acc: 0.6736\n",
      "Epoch 1426/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6605 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1427/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5969 - acc: 0.6584 - val_loss: 15.8704 - val_acc: 0.6736\n",
      "Epoch 1428/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6213 - acc: 0.6646 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1429/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6000 - acc: 0.6677 - val_loss: 15.8702 - val_acc: 0.6736\n",
      "Epoch 1430/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5909 - acc: 0.6667 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1431/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6605 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1432/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6101 - acc: 0.6625 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1433/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6687 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1434/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6063 - acc: 0.6708 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1435/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6625 - val_loss: 15.8702 - val_acc: 0.6736\n",
      "Epoch 1436/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6687 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1437/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5896 - acc: 0.6584 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1438/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6656 - val_loss: 15.8702 - val_acc: 0.6736\n",
      "Epoch 1439/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6677 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1440/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6064 - acc: 0.6625 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1441/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6636 - val_loss: 15.8704 - val_acc: 0.6736\n",
      "Epoch 1442/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6605 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1443/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5963 - acc: 0.6573 - val_loss: 15.8703 - val_acc: 0.6736\n",
      "Epoch 1444/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6198 - acc: 0.6708 - val_loss: 15.8702 - val_acc: 0.6736\n",
      "Epoch 1445/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5983 - acc: 0.6677 - val_loss: 15.8701 - val_acc: 0.6736\n",
      "Epoch 1446/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6687 - val_loss: 15.8700 - val_acc: 0.6736\n",
      "Epoch 1447/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6137 - acc: 0.6563 - val_loss: 15.8700 - val_acc: 0.6736\n",
      "Epoch 1448/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6034 - acc: 0.6615 - val_loss: 15.8700 - val_acc: 0.6736\n",
      "Epoch 1449/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6113 - acc: 0.6532 - val_loss: 15.8699 - val_acc: 0.6736\n",
      "Epoch 1450/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5926 - acc: 0.6677 - val_loss: 15.8699 - val_acc: 0.6736\n",
      "Epoch 1451/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6636 - val_loss: 15.8698 - val_acc: 0.6736\n",
      "Epoch 1452/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6126 - acc: 0.6667 - val_loss: 15.8698 - val_acc: 0.6736\n",
      "Epoch 1453/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6605 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1454/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6677 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1455/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6615 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1456/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6656 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1457/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6667 - val_loss: 15.8698 - val_acc: 0.6736\n",
      "Epoch 1458/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6046 - acc: 0.6656 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1459/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5974 - acc: 0.6615 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1460/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6021 - acc: 0.6605 - val_loss: 15.8697 - val_acc: 0.6736\n",
      "Epoch 1461/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5990 - acc: 0.6667 - val_loss: 15.8696 - val_acc: 0.6736\n",
      "Epoch 1462/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6594 - val_loss: 15.8695 - val_acc: 0.6736\n",
      "Epoch 1463/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5977 - acc: 0.6687 - val_loss: 15.8695 - val_acc: 0.6736\n",
      "Epoch 1464/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5963 - acc: 0.6625 - val_loss: 15.8695 - val_acc: 0.6736\n",
      "Epoch 1465/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6092 - acc: 0.6615 - val_loss: 15.8696 - val_acc: 0.6736\n",
      "Epoch 1466/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6071 - acc: 0.6646 - val_loss: 15.8695 - val_acc: 0.6736\n",
      "Epoch 1467/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6667 - val_loss: 15.8695 - val_acc: 0.6736\n",
      "Epoch 1468/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6677 - val_loss: 15.8694 - val_acc: 0.6736\n",
      "Epoch 1469/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6044 - acc: 0.6646 - val_loss: 15.8694 - val_acc: 0.6736\n",
      "Epoch 1470/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6042 - acc: 0.6573 - val_loss: 15.8692 - val_acc: 0.6736\n",
      "Epoch 1471/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6052 - acc: 0.6594 - val_loss: 15.8692 - val_acc: 0.6736\n",
      "Epoch 1472/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6148 - acc: 0.6573 - val_loss: 15.8691 - val_acc: 0.6736\n",
      "Epoch 1473/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6081 - acc: 0.6667 - val_loss: 15.8691 - val_acc: 0.6736\n",
      "Epoch 1474/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6224 - acc: 0.6573 - val_loss: 15.8691 - val_acc: 0.6736\n",
      "Epoch 1475/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6625 - val_loss: 15.8691 - val_acc: 0.6736\n",
      "Epoch 1476/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6140 - acc: 0.6563 - val_loss: 15.8690 - val_acc: 0.6736\n",
      "Epoch 1477/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6139 - acc: 0.6605 - val_loss: 15.8690 - val_acc: 0.6736\n",
      "Epoch 1478/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6677 - val_loss: 15.8688 - val_acc: 0.6736\n",
      "Epoch 1479/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6718 - val_loss: 15.8689 - val_acc: 0.6736\n",
      "Epoch 1480/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6636 - val_loss: 15.8689 - val_acc: 0.6736\n",
      "Epoch 1481/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6245 - acc: 0.6532 - val_loss: 15.8687 - val_acc: 0.6736\n",
      "Epoch 1482/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6667 - val_loss: 15.8685 - val_acc: 0.6736\n",
      "Epoch 1483/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5934 - acc: 0.6646 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1484/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6625 - val_loss: 15.8685 - val_acc: 0.6736\n",
      "Epoch 1485/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6055 - acc: 0.6584 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1486/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6043 - acc: 0.6615 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1487/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6687 - val_loss: 15.8685 - val_acc: 0.6736\n",
      "Epoch 1488/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5997 - acc: 0.6656 - val_loss: 15.8685 - val_acc: 0.6736\n",
      "Epoch 1489/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6289 - acc: 0.6656 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1490/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6636 - val_loss: 15.8685 - val_acc: 0.6736\n",
      "Epoch 1491/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6074 - acc: 0.6625 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1492/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6084 - acc: 0.6615 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1493/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6625 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1494/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5951 - acc: 0.6625 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1495/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6615 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1496/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6584 - val_loss: 15.8682 - val_acc: 0.6736\n",
      "Epoch 1497/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6015 - acc: 0.6522 - val_loss: 15.8682 - val_acc: 0.6736\n",
      "Epoch 1498/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6001 - acc: 0.6625 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1499/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6615 - val_loss: 15.8682 - val_acc: 0.6736\n",
      "Epoch 1500/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6090 - acc: 0.6646 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1501/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6573 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1502/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6563 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1503/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6656 - val_loss: 15.8684 - val_acc: 0.6736\n",
      "Epoch 1504/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6062 - acc: 0.6646 - val_loss: 15.8683 - val_acc: 0.6736\n",
      "Epoch 1505/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6062 - acc: 0.6687 - val_loss: 15.8682 - val_acc: 0.6736\n",
      "Epoch 1506/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6056 - acc: 0.6605 - val_loss: 15.8681 - val_acc: 0.6736\n",
      "Epoch 1507/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6667 - val_loss: 15.8681 - val_acc: 0.6736\n",
      "Epoch 1508/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6192 - acc: 0.6615 - val_loss: 15.8681 - val_acc: 0.6736\n",
      "Epoch 1509/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6636 - val_loss: 15.8681 - val_acc: 0.6736\n",
      "Epoch 1510/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6584 - val_loss: 15.8681 - val_acc: 0.6736\n",
      "Epoch 1511/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6010 - acc: 0.6646 - val_loss: 15.8680 - val_acc: 0.6736\n",
      "Epoch 1512/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6698 - val_loss: 15.8680 - val_acc: 0.6736\n",
      "Epoch 1513/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6636 - val_loss: 15.8679 - val_acc: 0.6736\n",
      "Epoch 1514/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6542 - val_loss: 15.8679 - val_acc: 0.6736\n",
      "Epoch 1515/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6043 - acc: 0.6625 - val_loss: 15.8677 - val_acc: 0.6736\n",
      "Epoch 1516/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6130 - acc: 0.6615 - val_loss: 15.8677 - val_acc: 0.6736\n",
      "Epoch 1517/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6037 - acc: 0.6605 - val_loss: 15.8677 - val_acc: 0.6736\n",
      "Epoch 1518/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5971 - acc: 0.6698 - val_loss: 15.8676 - val_acc: 0.6736\n",
      "Epoch 1519/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5990 - acc: 0.6636 - val_loss: 15.8675 - val_acc: 0.6736\n",
      "Epoch 1520/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5983 - acc: 0.6708 - val_loss: 15.8674 - val_acc: 0.6736\n",
      "Epoch 1521/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6677 - val_loss: 15.8672 - val_acc: 0.6736\n",
      "Epoch 1522/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6077 - acc: 0.6594 - val_loss: 15.8671 - val_acc: 0.6736\n",
      "Epoch 1523/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6646 - val_loss: 15.8671 - val_acc: 0.6736\n",
      "Epoch 1524/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6089 - acc: 0.6594 - val_loss: 15.8671 - val_acc: 0.6736\n",
      "Epoch 1525/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6553 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1526/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6646 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1527/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6615 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1528/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6065 - acc: 0.6636 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1529/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6656 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1530/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6001 - acc: 0.6470 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1531/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6749 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1532/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6698 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1533/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6059 - acc: 0.6677 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1534/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6636 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1535/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6085 - acc: 0.6625 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1536/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6615 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1537/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6532 - val_loss: 15.8670 - val_acc: 0.6736\n",
      "Epoch 1538/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6164 - acc: 0.6584 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1539/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6615 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1540/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6068 - acc: 0.6615 - val_loss: 15.8669 - val_acc: 0.6736\n",
      "Epoch 1541/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6084 - acc: 0.6615 - val_loss: 15.8668 - val_acc: 0.6736\n",
      "Epoch 1542/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6085 - acc: 0.6594 - val_loss: 15.8667 - val_acc: 0.6736\n",
      "Epoch 1543/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6167 - acc: 0.6625 - val_loss: 15.8666 - val_acc: 0.6736\n",
      "Epoch 1544/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6016 - acc: 0.6563 - val_loss: 15.8666 - val_acc: 0.6736\n",
      "Epoch 1545/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6178 - acc: 0.6615 - val_loss: 15.8666 - val_acc: 0.6736\n",
      "Epoch 1546/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6117 - acc: 0.6667 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1547/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6194 - acc: 0.6646 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1548/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6729 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1549/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5980 - acc: 0.6522 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1550/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6615 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1551/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5950 - acc: 0.6584 - val_loss: 15.8664 - val_acc: 0.6736\n",
      "Epoch 1552/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6071 - acc: 0.6718 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1553/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6605 - val_loss: 15.8661 - val_acc: 0.6736\n",
      "Epoch 1554/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6646 - val_loss: 15.8663 - val_acc: 0.6736\n",
      "Epoch 1555/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6594 - val_loss: 15.8661 - val_acc: 0.6736\n",
      "Epoch 1556/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6034 - acc: 0.6656 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1557/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6057 - acc: 0.6625 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1558/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6006 - acc: 0.6698 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1559/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6049 - acc: 0.6687 - val_loss: 15.8663 - val_acc: 0.6736\n",
      "Epoch 1560/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5930 - acc: 0.6594 - val_loss: 15.8663 - val_acc: 0.6736\n",
      "Epoch 1561/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6625 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1562/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5994 - acc: 0.6563 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1563/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6058 - acc: 0.6636 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1564/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6594 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1565/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5981 - acc: 0.6522 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1566/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5955 - acc: 0.6594 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1567/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6708 - val_loss: 15.8663 - val_acc: 0.6736\n",
      "Epoch 1568/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5984 - acc: 0.6605 - val_loss: 15.8662 - val_acc: 0.6736\n",
      "Epoch 1569/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6625 - val_loss: 15.8663 - val_acc: 0.6736\n",
      "Epoch 1570/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6125 - acc: 0.6625 - val_loss: 15.8661 - val_acc: 0.6736\n",
      "Epoch 1571/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6605 - val_loss: 15.8660 - val_acc: 0.6736\n",
      "Epoch 1572/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6099 - acc: 0.6677 - val_loss: 15.8659 - val_acc: 0.6736\n",
      "Epoch 1573/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6007 - acc: 0.6625 - val_loss: 15.8659 - val_acc: 0.6736\n",
      "Epoch 1574/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5926 - acc: 0.6584 - val_loss: 15.8659 - val_acc: 0.6736\n",
      "Epoch 1575/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6708 - val_loss: 15.8658 - val_acc: 0.6736\n",
      "Epoch 1576/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6739 - val_loss: 15.8658 - val_acc: 0.6736\n",
      "Epoch 1577/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6646 - val_loss: 15.8658 - val_acc: 0.6736\n",
      "Epoch 1578/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6584 - val_loss: 15.8658 - val_acc: 0.6736\n",
      "Epoch 1579/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6708 - val_loss: 15.8657 - val_acc: 0.6736\n",
      "Epoch 1580/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6615 - val_loss: 15.8657 - val_acc: 0.6736\n",
      "Epoch 1581/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6646 - val_loss: 15.8657 - val_acc: 0.6736\n",
      "Epoch 1582/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6615 - val_loss: 15.8657 - val_acc: 0.6736\n",
      "Epoch 1583/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6187 - acc: 0.6573 - val_loss: 15.8656 - val_acc: 0.6736\n",
      "Epoch 1584/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6656 - val_loss: 15.8655 - val_acc: 0.6736\n",
      "Epoch 1585/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6801 - val_loss: 15.8654 - val_acc: 0.6736\n",
      "Epoch 1586/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6044 - acc: 0.6615 - val_loss: 15.8652 - val_acc: 0.6736\n",
      "Epoch 1587/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5954 - acc: 0.6646 - val_loss: 15.8652 - val_acc: 0.6736\n",
      "Epoch 1588/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6594 - val_loss: 15.8651 - val_acc: 0.6736\n",
      "Epoch 1589/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5858 - acc: 0.6542 - val_loss: 15.8651 - val_acc: 0.6736\n",
      "Epoch 1590/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5927 - acc: 0.6646 - val_loss: 15.8650 - val_acc: 0.6736\n",
      "Epoch 1591/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6573 - val_loss: 15.8650 - val_acc: 0.6736\n",
      "Epoch 1592/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6018 - acc: 0.6636 - val_loss: 15.8648 - val_acc: 0.6736\n",
      "Epoch 1593/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6615 - val_loss: 15.8648 - val_acc: 0.6736\n",
      "Epoch 1594/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5984 - acc: 0.6677 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1595/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5945 - acc: 0.6646 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1596/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6048 - acc: 0.6615 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1597/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6677 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1598/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6687 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1599/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6687 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1600/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6069 - acc: 0.6625 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1601/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6646 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1602/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6615 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1603/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6563 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1604/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6615 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1605/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6677 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1606/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5937 - acc: 0.6636 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1607/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5927 - acc: 0.6667 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1608/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6594 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1609/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6698 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1610/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5985 - acc: 0.6594 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1611/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6718 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1612/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6079 - acc: 0.6594 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1613/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6037 - acc: 0.6605 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1614/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6020 - acc: 0.6573 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1615/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6625 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1616/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6088 - acc: 0.6584 - val_loss: 15.8647 - val_acc: 0.6736\n",
      "Epoch 1617/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6667 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1618/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6110 - acc: 0.6718 - val_loss: 15.8646 - val_acc: 0.6736\n",
      "Epoch 1619/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6646 - val_loss: 15.8645 - val_acc: 0.6736\n",
      "Epoch 1620/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6584 - val_loss: 15.8644 - val_acc: 0.6736\n",
      "Epoch 1621/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6698 - val_loss: 15.8643 - val_acc: 0.6736\n",
      "Epoch 1622/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6121 - acc: 0.6615 - val_loss: 15.8643 - val_acc: 0.6736\n",
      "Epoch 1623/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6677 - val_loss: 15.8642 - val_acc: 0.6736\n",
      "Epoch 1624/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6099 - acc: 0.6542 - val_loss: 15.8641 - val_acc: 0.6736\n",
      "Epoch 1625/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6677 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1626/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5936 - acc: 0.6563 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1627/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6718 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1628/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6019 - acc: 0.6667 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1629/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6625 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1630/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6698 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1631/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6646 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1632/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5989 - acc: 0.6594 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1633/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6573 - val_loss: 15.8641 - val_acc: 0.6736\n",
      "Epoch 1634/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6087 - acc: 0.6605 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1635/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5965 - acc: 0.6749 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1636/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6002 - acc: 0.6584 - val_loss: 15.8641 - val_acc: 0.6736\n",
      "Epoch 1637/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6718 - val_loss: 15.8641 - val_acc: 0.6736\n",
      "Epoch 1638/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6687 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1639/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6636 - val_loss: 15.8640 - val_acc: 0.6736\n",
      "Epoch 1640/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6646 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1641/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6605 - val_loss: 15.8639 - val_acc: 0.6736\n",
      "Epoch 1642/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6594 - val_loss: 15.8638 - val_acc: 0.6736\n",
      "Epoch 1643/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6112 - acc: 0.6646 - val_loss: 15.8638 - val_acc: 0.6736\n",
      "Epoch 1644/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6039 - acc: 0.6532 - val_loss: 15.8636 - val_acc: 0.6736\n",
      "Epoch 1645/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6584 - val_loss: 15.8636 - val_acc: 0.6736\n",
      "Epoch 1646/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6594 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1647/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6605 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1648/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6667 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1649/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6064 - acc: 0.6594 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1650/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6573 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1651/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6636 - val_loss: 15.8637 - val_acc: 0.6736\n",
      "Epoch 1652/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6708 - val_loss: 15.8636 - val_acc: 0.6736\n",
      "Epoch 1653/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6106 - acc: 0.6687 - val_loss: 15.8635 - val_acc: 0.6736\n",
      "Epoch 1654/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5987 - acc: 0.6656 - val_loss: 15.8634 - val_acc: 0.6736\n",
      "Epoch 1655/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5967 - acc: 0.6615 - val_loss: 15.8634 - val_acc: 0.6736\n",
      "Epoch 1656/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5957 - acc: 0.6646 - val_loss: 15.8635 - val_acc: 0.6736\n",
      "Epoch 1657/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5936 - acc: 0.6646 - val_loss: 15.8635 - val_acc: 0.6736\n",
      "Epoch 1658/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6014 - acc: 0.6708 - val_loss: 15.8634 - val_acc: 0.6736\n",
      "Epoch 1659/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5976 - acc: 0.6615 - val_loss: 15.8633 - val_acc: 0.6736\n",
      "Epoch 1660/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6152 - acc: 0.6646 - val_loss: 15.8633 - val_acc: 0.6736\n",
      "Epoch 1661/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6646 - val_loss: 15.8634 - val_acc: 0.6736\n",
      "Epoch 1662/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6636 - val_loss: 15.8633 - val_acc: 0.6736\n",
      "Epoch 1663/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6646 - val_loss: 15.8633 - val_acc: 0.6736\n",
      "Epoch 1664/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6083 - acc: 0.6656 - val_loss: 15.8632 - val_acc: 0.6736\n",
      "Epoch 1665/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6636 - val_loss: 15.8631 - val_acc: 0.6736\n",
      "Epoch 1666/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6667 - val_loss: 15.8632 - val_acc: 0.6736\n",
      "Epoch 1667/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6625 - val_loss: 15.8632 - val_acc: 0.6736\n",
      "Epoch 1668/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6085 - acc: 0.6615 - val_loss: 15.8632 - val_acc: 0.6736\n",
      "Epoch 1669/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6687 - val_loss: 15.8631 - val_acc: 0.6736\n",
      "Epoch 1670/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6625 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1671/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6625 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1672/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6615 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1673/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6698 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1674/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6007 - acc: 0.6470 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1675/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6090 - acc: 0.6656 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1676/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6036 - acc: 0.6646 - val_loss: 15.8631 - val_acc: 0.6736\n",
      "Epoch 1677/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6594 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1678/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5851 - acc: 0.6656 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1679/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5988 - acc: 0.6729 - val_loss: 15.8630 - val_acc: 0.6736\n",
      "Epoch 1680/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6067 - acc: 0.6687 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1681/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6646 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1682/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6718 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1683/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6522 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1684/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6522 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1685/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6646 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1686/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6164 - acc: 0.6636 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1687/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6636 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1688/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5878 - acc: 0.6563 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1689/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6636 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1690/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6156 - acc: 0.6677 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1691/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6112 - acc: 0.6636 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1692/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6656 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1693/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6047 - acc: 0.6594 - val_loss: 15.8629 - val_acc: 0.6736\n",
      "Epoch 1694/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5862 - acc: 0.6605 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1695/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6625 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1696/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6615 - val_loss: 15.8628 - val_acc: 0.6736\n",
      "Epoch 1697/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6605 - val_loss: 15.8626 - val_acc: 0.6736\n",
      "Epoch 1698/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6185 - acc: 0.6667 - val_loss: 15.8625 - val_acc: 0.6736\n",
      "Epoch 1699/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5987 - acc: 0.6584 - val_loss: 15.8623 - val_acc: 0.6736\n",
      "Epoch 1700/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5950 - acc: 0.6687 - val_loss: 15.8622 - val_acc: 0.6736\n",
      "Epoch 1701/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6749 - val_loss: 15.8622 - val_acc: 0.6736\n",
      "Epoch 1702/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5987 - acc: 0.6542 - val_loss: 15.8621 - val_acc: 0.6736\n",
      "Epoch 1703/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6615 - val_loss: 15.8621 - val_acc: 0.6736\n",
      "Epoch 1704/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6157 - acc: 0.6687 - val_loss: 15.8619 - val_acc: 0.6736\n",
      "Epoch 1705/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6656 - val_loss: 15.8618 - val_acc: 0.6736\n",
      "Epoch 1706/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6182 - acc: 0.6625 - val_loss: 15.8617 - val_acc: 0.6736\n",
      "Epoch 1707/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6615 - val_loss: 15.8618 - val_acc: 0.6736\n",
      "Epoch 1708/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6708 - val_loss: 15.8618 - val_acc: 0.6736\n",
      "Epoch 1709/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6092 - acc: 0.6677 - val_loss: 15.8618 - val_acc: 0.6736\n",
      "Epoch 1710/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5976 - acc: 0.6625 - val_loss: 15.8617 - val_acc: 0.6736\n",
      "Epoch 1711/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6625 - val_loss: 15.8617 - val_acc: 0.6736\n",
      "Epoch 1712/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6573 - val_loss: 15.8617 - val_acc: 0.6736\n",
      "Epoch 1713/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6075 - acc: 0.6770 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1714/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6698 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1715/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6542 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1716/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6625 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1717/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6646 - val_loss: 15.8617 - val_acc: 0.6736\n",
      "Epoch 1718/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6097 - acc: 0.6729 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1719/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6667 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1720/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6000 - acc: 0.6667 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1721/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5992 - acc: 0.6677 - val_loss: 15.8615 - val_acc: 0.6736\n",
      "Epoch 1722/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6594 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1723/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6066 - acc: 0.6646 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1724/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5821 - acc: 0.6698 - val_loss: 15.8616 - val_acc: 0.6736\n",
      "Epoch 1725/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6573 - val_loss: 15.8616 - val_acc: 0.6777\n",
      "Epoch 1726/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5990 - acc: 0.6677 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1727/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6127 - acc: 0.6605 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1728/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5978 - acc: 0.6625 - val_loss: 15.8619 - val_acc: 0.6777\n",
      "Epoch 1729/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6185 - acc: 0.6646 - val_loss: 15.8619 - val_acc: 0.6777\n",
      "Epoch 1730/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6656 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1731/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6563 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1732/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5898 - acc: 0.6656 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1733/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6573 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1734/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6656 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1735/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5921 - acc: 0.6584 - val_loss: 15.8616 - val_acc: 0.6777\n",
      "Epoch 1736/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6677 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1737/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5967 - acc: 0.6573 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1738/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6584 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1739/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6698 - val_loss: 15.8617 - val_acc: 0.6777\n",
      "Epoch 1740/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6615 - val_loss: 15.8619 - val_acc: 0.6777\n",
      "Epoch 1741/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6584 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1742/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6584 - val_loss: 15.8618 - val_acc: 0.6777\n",
      "Epoch 1743/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6636 - val_loss: 15.8616 - val_acc: 0.6777\n",
      "Epoch 1744/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6646 - val_loss: 15.8616 - val_acc: 0.6777\n",
      "Epoch 1745/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6615 - val_loss: 15.8615 - val_acc: 0.6777\n",
      "Epoch 1746/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6059 - acc: 0.6563 - val_loss: 15.8614 - val_acc: 0.6777\n",
      "Epoch 1747/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6656 - val_loss: 15.8612 - val_acc: 0.6777\n",
      "Epoch 1748/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6000 - acc: 0.6594 - val_loss: 15.8612 - val_acc: 0.6777\n",
      "Epoch 1749/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6646 - val_loss: 15.8611 - val_acc: 0.6777\n",
      "Epoch 1750/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6615 - val_loss: 15.8611 - val_acc: 0.6777\n",
      "Epoch 1751/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6061 - acc: 0.6584 - val_loss: 15.8610 - val_acc: 0.6777\n",
      "Epoch 1752/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6016 - acc: 0.6573 - val_loss: 15.8610 - val_acc: 0.6777\n",
      "Epoch 1753/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6646 - val_loss: 15.8610 - val_acc: 0.6777\n",
      "Epoch 1754/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6594 - val_loss: 15.8610 - val_acc: 0.6777\n",
      "Epoch 1755/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6080 - acc: 0.6636 - val_loss: 15.8609 - val_acc: 0.6777\n",
      "Epoch 1756/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6687 - val_loss: 15.8608 - val_acc: 0.6777\n",
      "Epoch 1757/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5950 - acc: 0.6656 - val_loss: 15.8607 - val_acc: 0.6777\n",
      "Epoch 1758/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6076 - acc: 0.6636 - val_loss: 15.8606 - val_acc: 0.6777\n",
      "Epoch 1759/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6276 - acc: 0.6625 - val_loss: 15.8605 - val_acc: 0.6777\n",
      "Epoch 1760/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6739 - val_loss: 15.8605 - val_acc: 0.6777\n",
      "Epoch 1761/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6099 - acc: 0.6708 - val_loss: 15.8604 - val_acc: 0.6777\n",
      "Epoch 1762/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6687 - val_loss: 15.8603 - val_acc: 0.6777\n",
      "Epoch 1763/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5851 - acc: 0.6584 - val_loss: 15.8603 - val_acc: 0.6777\n",
      "Epoch 1764/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6646 - val_loss: 15.8602 - val_acc: 0.6777\n",
      "Epoch 1765/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.6664 - acc: 0.65 - 0s - loss: 16.5936 - acc: 0.6656 - val_loss: 15.8601 - val_acc: 0.6777\n",
      "Epoch 1766/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6687 - val_loss: 15.8602 - val_acc: 0.6777\n",
      "Epoch 1767/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6636 - val_loss: 15.8601 - val_acc: 0.6777\n",
      "Epoch 1768/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6041 - acc: 0.6739 - val_loss: 15.8600 - val_acc: 0.6777\n",
      "Epoch 1769/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6646 - val_loss: 15.8599 - val_acc: 0.6777\n",
      "Epoch 1770/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5972 - acc: 0.6718 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1771/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5981 - acc: 0.6687 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1772/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6687 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1773/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6781 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1774/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5931 - acc: 0.6615 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1775/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5984 - acc: 0.6584 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1776/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5976 - acc: 0.6605 - val_loss: 15.8595 - val_acc: 0.6777\n",
      "Epoch 1777/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6708 - val_loss: 15.8595 - val_acc: 0.6777\n",
      "Epoch 1778/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6625 - val_loss: 15.8594 - val_acc: 0.6777\n",
      "Epoch 1779/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5927 - acc: 0.6687 - val_loss: 15.8594 - val_acc: 0.6777\n",
      "Epoch 1780/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6625 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1781/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5937 - acc: 0.6667 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1782/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6594 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1783/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5898 - acc: 0.6573 - val_loss: 15.8595 - val_acc: 0.6777\n",
      "Epoch 1784/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6605 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1785/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6625 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1786/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6667 - val_loss: 15.8595 - val_acc: 0.6777\n",
      "Epoch 1787/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6636 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1788/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6573 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1789/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6646 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1790/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6166 - acc: 0.6636 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1791/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6165 - acc: 0.6573 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1792/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6729 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1793/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5969 - acc: 0.6667 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1794/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6128 - acc: 0.6636 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1795/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6584 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1796/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5924 - acc: 0.6615 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1797/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6656 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1798/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6636 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1799/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6677 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1800/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5951 - acc: 0.6677 - val_loss: 15.8596 - val_acc: 0.6777\n",
      "Epoch 1801/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6008 - acc: 0.6615 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1802/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6090 - acc: 0.6563 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1803/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6656 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1804/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6118 - acc: 0.6594 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1805/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6615 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1806/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6667 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1807/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6687 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1808/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5984 - acc: 0.6636 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1809/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6542 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1810/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6636 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1811/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6605 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1812/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5994 - acc: 0.6698 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1813/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6677 - val_loss: 15.8599 - val_acc: 0.6777\n",
      "Epoch 1814/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6636 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1815/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6667 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1816/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6056 - acc: 0.6532 - val_loss: 15.8599 - val_acc: 0.6777\n",
      "Epoch 1817/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6625 - val_loss: 15.8600 - val_acc: 0.6777\n",
      "Epoch 1818/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6594 - val_loss: 15.8600 - val_acc: 0.6777\n",
      "Epoch 1819/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5948 - acc: 0.6698 - val_loss: 15.8599 - val_acc: 0.6777\n",
      "Epoch 1820/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6656 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1821/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6615 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1822/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6594 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1823/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6677 - val_loss: 15.8598 - val_acc: 0.6777\n",
      "Epoch 1824/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5989 - acc: 0.6636 - val_loss: 15.8597 - val_acc: 0.6777\n",
      "Epoch 1825/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6656 - val_loss: 15.8595 - val_acc: 0.6777\n",
      "Epoch 1826/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6718 - val_loss: 15.8594 - val_acc: 0.6777\n",
      "Epoch 1827/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6646 - val_loss: 15.8594 - val_acc: 0.6777\n",
      "Epoch 1828/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6687 - val_loss: 15.8594 - val_acc: 0.6777\n",
      "Epoch 1829/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5870 - acc: 0.6708 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1830/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6729 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1831/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6004 - acc: 0.6594 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1832/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6708 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1833/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6584 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1834/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5936 - acc: 0.6656 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1835/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6625 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1836/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6698 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1837/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6646 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1838/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6563 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1839/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5940 - acc: 0.6708 - val_loss: 15.8593 - val_acc: 0.6777\n",
      "Epoch 1840/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5841 - acc: 0.6646 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1841/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6067 - acc: 0.6532 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1842/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6625 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1843/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6687 - val_loss: 15.8591 - val_acc: 0.6777\n",
      "Epoch 1844/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6615 - val_loss: 15.8591 - val_acc: 0.6777\n",
      "Epoch 1845/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6036 - acc: 0.6615 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1846/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6687 - val_loss: 15.8592 - val_acc: 0.6777\n",
      "Epoch 1847/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6625 - val_loss: 15.8591 - val_acc: 0.6777\n",
      "Epoch 1848/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5956 - acc: 0.6584 - val_loss: 15.8590 - val_acc: 0.6777\n",
      "Epoch 1849/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6584 - val_loss: 15.8589 - val_acc: 0.6777\n",
      "Epoch 1850/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6646 - val_loss: 15.8590 - val_acc: 0.6777\n",
      "Epoch 1851/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6656 - val_loss: 15.8590 - val_acc: 0.6777\n",
      "Epoch 1852/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6625 - val_loss: 15.8590 - val_acc: 0.6777\n",
      "Epoch 1853/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6116 - acc: 0.6677 - val_loss: 15.8589 - val_acc: 0.6777\n",
      "Epoch 1854/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6667 - val_loss: 15.8588 - val_acc: 0.6777\n",
      "Epoch 1855/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6656 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1856/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6636 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1857/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5999 - acc: 0.6677 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1858/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6729 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1859/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6646 - val_loss: 15.8586 - val_acc: 0.6777\n",
      "Epoch 1860/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6667 - val_loss: 15.8586 - val_acc: 0.6777\n",
      "Epoch 1861/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6718 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1862/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5870 - acc: 0.6615 - val_loss: 15.8587 - val_acc: 0.6777\n",
      "Epoch 1863/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5921 - acc: 0.6677 - val_loss: 15.8586 - val_acc: 0.6777\n",
      "Epoch 1864/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6584 - val_loss: 15.8586 - val_acc: 0.6777\n",
      "Epoch 1865/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6729 - val_loss: 15.8586 - val_acc: 0.6777\n",
      "Epoch 1866/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6089 - acc: 0.6584 - val_loss: 15.8585 - val_acc: 0.6777\n",
      "Epoch 1867/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6030 - acc: 0.6573 - val_loss: 15.8584 - val_acc: 0.6777\n",
      "Epoch 1868/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6584 - val_loss: 15.8585 - val_acc: 0.6777\n",
      "Epoch 1869/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6646 - val_loss: 15.8585 - val_acc: 0.6777\n",
      "Epoch 1870/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6625 - val_loss: 15.8585 - val_acc: 0.6777\n",
      "Epoch 1871/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6667 - val_loss: 15.8585 - val_acc: 0.6777\n",
      "Epoch 1872/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5973 - acc: 0.6656 - val_loss: 15.8584 - val_acc: 0.6777\n",
      "Epoch 1873/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5932 - acc: 0.6656 - val_loss: 15.8583 - val_acc: 0.6777\n",
      "Epoch 1874/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6625 - val_loss: 15.8584 - val_acc: 0.6777\n",
      "Epoch 1875/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6646 - val_loss: 15.8582 - val_acc: 0.6777\n",
      "Epoch 1876/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6708 - val_loss: 15.8582 - val_acc: 0.6777\n",
      "Epoch 1877/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6656 - val_loss: 15.8582 - val_acc: 0.6777\n",
      "Epoch 1878/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5896 - acc: 0.6594 - val_loss: 15.8582 - val_acc: 0.6777\n",
      "Epoch 1879/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6101 - acc: 0.6573 - val_loss: 15.8583 - val_acc: 0.6777\n",
      "Epoch 1880/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6021 - acc: 0.6677 - val_loss: 15.8582 - val_acc: 0.6736\n",
      "Epoch 1881/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6677 - val_loss: 15.8581 - val_acc: 0.6777\n",
      "Epoch 1882/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6066 - acc: 0.6636 - val_loss: 15.8581 - val_acc: 0.6777\n",
      "Epoch 1883/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6636 - val_loss: 15.8580 - val_acc: 0.6777\n",
      "Epoch 1884/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6002 - acc: 0.6636 - val_loss: 15.8580 - val_acc: 0.6736\n",
      "Epoch 1885/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6636 - val_loss: 15.8579 - val_acc: 0.6736\n",
      "Epoch 1886/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6625 - val_loss: 15.8579 - val_acc: 0.6736\n",
      "Epoch 1887/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6584 - val_loss: 15.8580 - val_acc: 0.6736\n",
      "Epoch 1888/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6625 - val_loss: 15.8579 - val_acc: 0.6736\n",
      "Epoch 1889/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6573 - val_loss: 15.8578 - val_acc: 0.6736\n",
      "Epoch 1890/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6584 - val_loss: 15.8578 - val_acc: 0.6736\n",
      "Epoch 1891/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6573 - val_loss: 15.8578 - val_acc: 0.6736\n",
      "Epoch 1892/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6625 - val_loss: 15.8578 - val_acc: 0.6736\n",
      "Epoch 1893/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6729 - val_loss: 15.8577 - val_acc: 0.6736\n",
      "Epoch 1894/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6615 - val_loss: 15.8577 - val_acc: 0.6736\n",
      "Epoch 1895/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6729 - val_loss: 15.8576 - val_acc: 0.6736\n",
      "Epoch 1896/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6542 - val_loss: 15.8576 - val_acc: 0.6736\n",
      "Epoch 1897/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6677 - val_loss: 15.8577 - val_acc: 0.6736\n",
      "Epoch 1898/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6625 - val_loss: 15.8577 - val_acc: 0.6736\n",
      "Epoch 1899/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6718 - val_loss: 15.8576 - val_acc: 0.6736\n",
      "Epoch 1900/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5887 - acc: 0.6646 - val_loss: 15.8576 - val_acc: 0.6736\n",
      "Epoch 1901/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6605 - val_loss: 15.8575 - val_acc: 0.6736\n",
      "Epoch 1902/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6615 - val_loss: 15.8575 - val_acc: 0.6736\n",
      "Epoch 1903/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6615 - val_loss: 15.8575 - val_acc: 0.6736\n",
      "Epoch 1904/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6615 - val_loss: 15.8575 - val_acc: 0.6736\n",
      "Epoch 1905/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6625 - val_loss: 15.8575 - val_acc: 0.6736\n",
      "Epoch 1906/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6605 - val_loss: 15.8574 - val_acc: 0.6736\n",
      "Epoch 1907/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6718 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1908/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6542 - val_loss: 15.8574 - val_acc: 0.6736\n",
      "Epoch 1909/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6594 - val_loss: 15.8574 - val_acc: 0.6736\n",
      "Epoch 1910/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6584 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1911/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6584 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1912/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6636 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1913/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6594 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1914/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5962 - acc: 0.6677 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1915/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6677 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1916/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6677 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1917/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6687 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1918/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6656 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1919/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6047 - acc: 0.6625 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1920/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6119 - acc: 0.6646 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1921/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5950 - acc: 0.6698 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1922/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6615 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1923/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6153 - acc: 0.6625 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1924/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6677 - val_loss: 15.8569 - val_acc: 0.6736\n",
      "Epoch 1925/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6061 - acc: 0.6594 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1926/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6605 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1927/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6573 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1928/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6656 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1929/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6656 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1930/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6615 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1931/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6646 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1932/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6001 - acc: 0.6667 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1933/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6636 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1934/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6667 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1935/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6040 - acc: 0.6584 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1936/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6636 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1937/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6636 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1938/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6646 - val_loss: 15.8573 - val_acc: 0.6736\n",
      "Epoch 1939/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6687 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1940/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6594 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1941/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6749 - val_loss: 15.8572 - val_acc: 0.6736\n",
      "Epoch 1942/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6718 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1943/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5927 - acc: 0.6615 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1944/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6014 - acc: 0.6636 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1945/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6594 - val_loss: 15.8570 - val_acc: 0.6736\n",
      "Epoch 1946/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6625 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1947/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6037 - acc: 0.6636 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1948/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6636 - val_loss: 15.8571 - val_acc: 0.6736\n",
      "Epoch 1949/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6729 - val_loss: 15.8569 - val_acc: 0.6736\n",
      "Epoch 1950/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6656 - val_loss: 15.8569 - val_acc: 0.6736\n",
      "Epoch 1951/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6594 - val_loss: 15.8568 - val_acc: 0.6736\n",
      "Epoch 1952/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5932 - acc: 0.6677 - val_loss: 15.8568 - val_acc: 0.6736\n",
      "Epoch 1953/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6646 - val_loss: 15.8568 - val_acc: 0.6736\n",
      "Epoch 1954/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6667 - val_loss: 15.8567 - val_acc: 0.6736\n",
      "Epoch 1955/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5947 - acc: 0.6698 - val_loss: 15.8566 - val_acc: 0.6736\n",
      "Epoch 1956/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6625 - val_loss: 15.8566 - val_acc: 0.6736\n",
      "Epoch 1957/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6051 - acc: 0.6542 - val_loss: 15.8565 - val_acc: 0.6736\n",
      "Epoch 1958/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6656 - val_loss: 15.8564 - val_acc: 0.6736\n",
      "Epoch 1959/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6656 - val_loss: 15.8564 - val_acc: 0.6736\n",
      "Epoch 1960/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5813 - acc: 0.6625 - val_loss: 15.8565 - val_acc: 0.6736\n",
      "Epoch 1961/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6677 - val_loss: 15.8564 - val_acc: 0.6736\n",
      "Epoch 1962/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6656 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1963/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6667 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1964/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6636 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1965/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6002 - acc: 0.6615 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1966/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5908 - acc: 0.6687 - val_loss: 15.8564 - val_acc: 0.6736\n",
      "Epoch 1967/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6687 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1968/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6615 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1969/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6636 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1970/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6687 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1971/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6687 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1972/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6605 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1973/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6625 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1974/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5858 - acc: 0.6636 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1975/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6615 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1976/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6615 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1977/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6146 - acc: 0.6687 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1978/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6749 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1979/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6708 - val_loss: 15.8563 - val_acc: 0.6736\n",
      "Epoch 1980/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5945 - acc: 0.6636 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1981/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6708 - val_loss: 15.8562 - val_acc: 0.6736\n",
      "Epoch 1982/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6729 - val_loss: 15.8560 - val_acc: 0.6736\n",
      "Epoch 1983/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6038 - acc: 0.6584 - val_loss: 15.8560 - val_acc: 0.6736\n",
      "Epoch 1984/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6636 - val_loss: 15.8560 - val_acc: 0.6736\n",
      "Epoch 1985/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6563 - val_loss: 15.8560 - val_acc: 0.6736\n",
      "Epoch 1986/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6625 - val_loss: 15.8560 - val_acc: 0.6736\n",
      "Epoch 1987/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6646 - val_loss: 15.8559 - val_acc: 0.6736\n",
      "Epoch 1988/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6563 - val_loss: 15.8558 - val_acc: 0.6736\n",
      "Epoch 1989/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6667 - val_loss: 15.8558 - val_acc: 0.6736\n",
      "Epoch 1990/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6625 - val_loss: 15.8559 - val_acc: 0.6736\n",
      "Epoch 1991/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5973 - acc: 0.6646 - val_loss: 15.8559 - val_acc: 0.6736\n",
      "Epoch 1992/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6625 - val_loss: 15.8558 - val_acc: 0.6736\n",
      "Epoch 1993/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6636 - val_loss: 15.8558 - val_acc: 0.6736\n",
      "Epoch 1994/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6636 - val_loss: 15.8556 - val_acc: 0.6736\n",
      "Epoch 1995/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6667 - val_loss: 15.8555 - val_acc: 0.6736\n",
      "Epoch 1996/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6667 - val_loss: 15.8554 - val_acc: 0.6736\n",
      "Epoch 1997/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6698 - val_loss: 15.8553 - val_acc: 0.6736\n",
      "Epoch 1998/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6656 - val_loss: 15.8553 - val_acc: 0.6736\n",
      "Epoch 1999/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6708 - val_loss: 15.8553 - val_acc: 0.6736\n",
      "Epoch 2000/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6677 - val_loss: 15.8553 - val_acc: 0.6736\n",
      "Epoch 2001/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6677 - val_loss: 15.8552 - val_acc: 0.6736\n",
      "Epoch 2002/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6615 - val_loss: 15.8552 - val_acc: 0.6736\n",
      "Epoch 2003/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6646 - val_loss: 15.8552 - val_acc: 0.6736\n",
      "Epoch 2004/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6656 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2005/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6656 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2006/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6625 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2007/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6687 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2008/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6625 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2009/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5839 - acc: 0.6584 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2010/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5958 - acc: 0.6615 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2011/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6594 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2012/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6636 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2013/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6594 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2014/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6729 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2015/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6656 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2016/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6605 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2017/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6056 - acc: 0.6646 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2018/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5608 - acc: 0.6656 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2019/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6553 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2020/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6594 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2021/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6677 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2022/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6646 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2023/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6553 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2024/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6687 - val_loss: 15.8551 - val_acc: 0.6736\n",
      "Epoch 2025/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6625 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2026/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6677 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2027/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6656 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2028/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5998 - acc: 0.6698 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2029/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6667 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2030/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6646 - val_loss: 15.8550 - val_acc: 0.6736\n",
      "Epoch 2031/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6646 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2032/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6667 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2033/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6058 - acc: 0.6739 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2034/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6615 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2035/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6003 - acc: 0.6677 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2036/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6646 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2037/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6625 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2038/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6095 - acc: 0.6636 - val_loss: 15.8547 - val_acc: 0.6736\n",
      "Epoch 2039/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6646 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2040/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6594 - val_loss: 15.8547 - val_acc: 0.6736\n",
      "Epoch 2041/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6095 - acc: 0.6646 - val_loss: 15.8547 - val_acc: 0.6736\n",
      "Epoch 2042/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6677 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2043/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5850 - acc: 0.6605 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2044/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5767 - acc: 0.6677 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2045/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6144 - acc: 0.6656 - val_loss: 15.8549 - val_acc: 0.6736\n",
      "Epoch 2046/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6687 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2047/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5954 - acc: 0.6532 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2048/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6042 - acc: 0.6584 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2049/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6656 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2050/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6563 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2051/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6615 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2052/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6615 - val_loss: 15.8548 - val_acc: 0.6736\n",
      "Epoch 2053/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5985 - acc: 0.6749 - val_loss: 15.8547 - val_acc: 0.6736\n",
      "Epoch 2054/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6615 - val_loss: 15.8547 - val_acc: 0.6736\n",
      "Epoch 2055/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6708 - val_loss: 15.8546 - val_acc: 0.6736\n",
      "Epoch 2056/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6054 - acc: 0.6636 - val_loss: 15.8544 - val_acc: 0.6736\n",
      "Epoch 2057/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6625 - val_loss: 15.8543 - val_acc: 0.6736\n",
      "Epoch 2058/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6594 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2059/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6687 - val_loss: 15.8543 - val_acc: 0.6736\n",
      "Epoch 2060/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5980 - acc: 0.6770 - val_loss: 15.8543 - val_acc: 0.6736\n",
      "Epoch 2061/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6123 - acc: 0.6573 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2062/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6656 - val_loss: 15.8543 - val_acc: 0.6736\n",
      "Epoch 2063/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6615 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2064/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6594 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2065/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6594 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2066/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6698 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2067/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6687 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2068/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5957 - acc: 0.6625 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2069/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5880 - acc: 0.6625 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2070/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6656 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2071/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6605 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2072/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6687 - val_loss: 15.8542 - val_acc: 0.6736\n",
      "Epoch 2073/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6605 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2074/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5596 - acc: 0.6656 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2075/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6718 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2076/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6677 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2077/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6553 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2078/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5962 - acc: 0.6615 - val_loss: 15.8541 - val_acc: 0.6736\n",
      "Epoch 2079/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5948 - acc: 0.6542 - val_loss: 15.8540 - val_acc: 0.6736\n",
      "Epoch 2080/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6687 - val_loss: 15.8540 - val_acc: 0.6736\n",
      "Epoch 2081/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6078 - acc: 0.6605 - val_loss: 15.8539 - val_acc: 0.6736\n",
      "Epoch 2082/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6010 - acc: 0.6605 - val_loss: 15.8539 - val_acc: 0.6736\n",
      "Epoch 2083/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6031 - acc: 0.6636 - val_loss: 15.8538 - val_acc: 0.6736\n",
      "Epoch 2084/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6636 - val_loss: 15.8538 - val_acc: 0.6736\n",
      "Epoch 2085/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6708 - val_loss: 15.8537 - val_acc: 0.6736\n",
      "Epoch 2086/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6646 - val_loss: 15.8537 - val_acc: 0.6736\n",
      "Epoch 2087/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6698 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2088/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6646 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2089/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6656 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2090/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6584 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2091/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5967 - acc: 0.6636 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2092/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6594 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2093/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6687 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2094/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6615 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2095/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6605 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2096/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6605 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2097/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5951 - acc: 0.6636 - val_loss: 15.8534 - val_acc: 0.6736\n",
      "Epoch 2098/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5954 - acc: 0.6636 - val_loss: 15.8534 - val_acc: 0.6736\n",
      "Epoch 2099/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6584 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2100/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6615 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2101/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6625 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2102/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6687 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2103/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6646 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2104/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6584 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2105/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6677 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2106/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6677 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2107/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6646 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2108/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6677 - val_loss: 15.8536 - val_acc: 0.6736\n",
      "Epoch 2109/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5941 - acc: 0.6646 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2110/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5994 - acc: 0.6625 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2111/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6615 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2112/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6667 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2113/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6698 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2114/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6625 - val_loss: 15.8535 - val_acc: 0.6736\n",
      "Epoch 2115/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6615 - val_loss: 15.8534 - val_acc: 0.6736\n",
      "Epoch 2116/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6718 - val_loss: 15.8533 - val_acc: 0.6736\n",
      "Epoch 2117/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6005 - acc: 0.6573 - val_loss: 15.8534 - val_acc: 0.6736\n",
      "Epoch 2118/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5936 - acc: 0.6594 - val_loss: 15.8534 - val_acc: 0.6736\n",
      "Epoch 2119/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6615 - val_loss: 15.8533 - val_acc: 0.6736\n",
      "Epoch 2120/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6646 - val_loss: 15.8533 - val_acc: 0.6736\n",
      "Epoch 2121/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6646 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2122/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5931 - acc: 0.6563 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2123/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6646 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2124/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6025 - acc: 0.6667 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2125/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6573 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2126/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6563 - val_loss: 15.8532 - val_acc: 0.6736\n",
      "Epoch 2127/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6677 - val_loss: 15.8531 - val_acc: 0.6736\n",
      "Epoch 2128/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6656 - val_loss: 15.8530 - val_acc: 0.6736\n",
      "Epoch 2129/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6625 - val_loss: 15.8530 - val_acc: 0.6736\n",
      "Epoch 2130/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6636 - val_loss: 15.8530 - val_acc: 0.6736\n",
      "Epoch 2131/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6667 - val_loss: 15.8530 - val_acc: 0.6736\n",
      "Epoch 2132/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6698 - val_loss: 15.8530 - val_acc: 0.6736\n",
      "Epoch 2133/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6625 - val_loss: 15.8529 - val_acc: 0.6736\n",
      "Epoch 2134/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6594 - val_loss: 15.8528 - val_acc: 0.6736\n",
      "Epoch 2135/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6667 - val_loss: 15.8527 - val_acc: 0.6736\n",
      "Epoch 2136/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6687 - val_loss: 15.8527 - val_acc: 0.6736\n",
      "Epoch 2137/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6656 - val_loss: 15.8527 - val_acc: 0.6736\n",
      "Epoch 2138/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6594 - val_loss: 15.8527 - val_acc: 0.6736\n",
      "Epoch 2139/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6698 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2140/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6667 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2141/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5898 - acc: 0.6615 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2142/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6677 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2143/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6698 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2144/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6522 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2145/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6698 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2146/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6677 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2147/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5889 - acc: 0.6615 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2148/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6656 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2149/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6584 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2150/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6708 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2151/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5957 - acc: 0.6605 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2152/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6687 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2153/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6749 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2154/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6646 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2155/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6729 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2156/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6636 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2157/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5841 - acc: 0.6718 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2158/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6667 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2159/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6615 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2160/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6594 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2161/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6687 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2162/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6646 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2163/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6687 - val_loss: 15.8526 - val_acc: 0.6736\n",
      "Epoch 2164/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6687 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2165/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5932 - acc: 0.6687 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2166/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6553 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2167/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5889 - acc: 0.6573 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2168/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5676 - acc: 0.6698 - val_loss: 15.8525 - val_acc: 0.6736\n",
      "Epoch 2169/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6584 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2170/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6058 - acc: 0.6573 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2171/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6000 - acc: 0.6656 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2172/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6625 - val_loss: 15.8524 - val_acc: 0.6736\n",
      "Epoch 2173/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6667 - val_loss: 15.8523 - val_acc: 0.6736\n",
      "Epoch 2174/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6698 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2175/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6615 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2176/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6020 - acc: 0.6667 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2177/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6687 - val_loss: 15.8522 - val_acc: 0.6736\n",
      "Epoch 2178/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6646 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2179/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5952 - acc: 0.6563 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2180/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6667 - val_loss: 15.8520 - val_acc: 0.6736\n",
      "Epoch 2181/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6573 - val_loss: 15.8521 - val_acc: 0.6736\n",
      "Epoch 2182/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6687 - val_loss: 15.8520 - val_acc: 0.6736\n",
      "Epoch 2183/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6584 - val_loss: 15.8520 - val_acc: 0.6736\n",
      "Epoch 2184/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6656 - val_loss: 15.8519 - val_acc: 0.6736\n",
      "Epoch 2185/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6625 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2186/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6656 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2187/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6615 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2188/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6594 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2189/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6677 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2190/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6656 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2191/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6687 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2192/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6018 - acc: 0.6636 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2193/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6739 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2194/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5915 - acc: 0.6729 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2195/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6615 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2196/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6573 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2197/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6646 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2198/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6625 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2199/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6625 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2200/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6708 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2201/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6708 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2202/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5940 - acc: 0.6646 - val_loss: 15.8518 - val_acc: 0.6736\n",
      "Epoch 2203/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6563 - val_loss: 15.8517 - val_acc: 0.6736\n",
      "Epoch 2204/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5980 - acc: 0.6667 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2205/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6615 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2206/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5887 - acc: 0.6687 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2207/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6636 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2208/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6739 - val_loss: 15.8516 - val_acc: 0.6736\n",
      "Epoch 2209/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6636 - val_loss: 15.8515 - val_acc: 0.6736\n",
      "Epoch 2210/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6667 - val_loss: 15.8514 - val_acc: 0.6736\n",
      "Epoch 2211/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6625 - val_loss: 15.8514 - val_acc: 0.6736\n",
      "Epoch 2212/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6584 - val_loss: 15.8514 - val_acc: 0.6736\n",
      "Epoch 2213/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6677 - val_loss: 15.8514 - val_acc: 0.6736\n",
      "Epoch 2214/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5969 - acc: 0.6687 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2215/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6739 - val_loss: 15.8512 - val_acc: 0.6736\n",
      "Epoch 2216/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6026 - acc: 0.6615 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2217/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6718 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2218/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6687 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2219/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6553 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2220/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6677 - val_loss: 15.8513 - val_acc: 0.6736\n",
      "Epoch 2221/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6089 - acc: 0.6584 - val_loss: 15.8512 - val_acc: 0.6736\n",
      "Epoch 2222/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6636 - val_loss: 15.8511 - val_acc: 0.6736\n",
      "Epoch 2223/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6625 - val_loss: 15.8511 - val_acc: 0.6736\n",
      "Epoch 2224/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5852 - acc: 0.6625 - val_loss: 15.8510 - val_acc: 0.6736\n",
      "Epoch 2225/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6646 - val_loss: 15.8509 - val_acc: 0.6736\n",
      "Epoch 2226/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6739 - val_loss: 15.8509 - val_acc: 0.6736\n",
      "Epoch 2227/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6573 - val_loss: 15.8509 - val_acc: 0.6736\n",
      "Epoch 2228/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6646 - val_loss: 15.8508 - val_acc: 0.6736\n",
      "Epoch 2229/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6625 - val_loss: 15.8508 - val_acc: 0.6736\n",
      "Epoch 2230/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5941 - acc: 0.6667 - val_loss: 15.8507 - val_acc: 0.6736\n",
      "Epoch 2231/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5930 - acc: 0.6708 - val_loss: 15.8507 - val_acc: 0.6736\n",
      "Epoch 2232/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6625 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2233/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6584 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2234/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5896 - acc: 0.6667 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2235/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6532 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2236/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6594 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2237/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6605 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2238/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6656 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2239/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6563 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2240/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6064 - acc: 0.6646 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2241/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6718 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2242/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6625 - val_loss: 15.8503 - val_acc: 0.6736\n",
      "Epoch 2243/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6687 - val_loss: 15.8503 - val_acc: 0.6736\n",
      "Epoch 2244/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6646 - val_loss: 15.8503 - val_acc: 0.6736\n",
      "Epoch 2245/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6636 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2246/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6594 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2247/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6625 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2248/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6615 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2249/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6718 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2250/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5668 - acc: 0.6646 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2251/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6708 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2252/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6636 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2253/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6532 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2254/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6667 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2255/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5899 - acc: 0.6646 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2256/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6594 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2257/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6646 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2258/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6677 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2259/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6594 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2260/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6646 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2261/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6677 - val_loss: 15.8507 - val_acc: 0.6736\n",
      "Epoch 2262/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6677 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2263/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6677 - val_loss: 15.8507 - val_acc: 0.6736\n",
      "Epoch 2264/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6004 - acc: 0.6729 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2265/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5962 - acc: 0.6615 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2266/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6677 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2267/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6605 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2268/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5956 - acc: 0.6615 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2269/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6625 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2270/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6646 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2271/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6656 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2272/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5963 - acc: 0.6615 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2273/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6605 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2274/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6625 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2275/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6687 - val_loss: 15.8506 - val_acc: 0.6736\n",
      "Epoch 2276/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6150 - acc: 0.6677 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2277/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6636 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2278/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5971 - acc: 0.6749 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2279/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6625 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2280/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6667 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2281/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6019 - acc: 0.6718 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2282/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6594 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2283/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6542 - val_loss: 15.8505 - val_acc: 0.6736\n",
      "Epoch 2284/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6708 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2285/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5957 - acc: 0.6656 - val_loss: 15.8504 - val_acc: 0.6736\n",
      "Epoch 2286/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6636 - val_loss: 15.8503 - val_acc: 0.6736\n",
      "Epoch 2287/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5852 - acc: 0.6656 - val_loss: 15.8503 - val_acc: 0.6736\n",
      "Epoch 2288/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6542 - val_loss: 15.8502 - val_acc: 0.6736\n",
      "Epoch 2289/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6636 - val_loss: 15.8502 - val_acc: 0.6736\n",
      "Epoch 2290/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5924 - acc: 0.6573 - val_loss: 15.8501 - val_acc: 0.6736\n",
      "Epoch 2291/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6698 - val_loss: 15.8501 - val_acc: 0.6736\n",
      "Epoch 2292/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6677 - val_loss: 15.8501 - val_acc: 0.6736\n",
      "Epoch 2293/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6625 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2294/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6542 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2295/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6594 - val_loss: 15.8501 - val_acc: 0.6736\n",
      "Epoch 2296/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5958 - acc: 0.6625 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2297/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6656 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2298/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6573 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2299/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6698 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2300/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6594 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2301/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6677 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2302/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6667 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2303/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6021 - acc: 0.6687 - val_loss: 15.8498 - val_acc: 0.6736\n",
      "Epoch 2304/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6615 - val_loss: 15.8498 - val_acc: 0.6736\n",
      "Epoch 2305/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6667 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2306/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6594 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2307/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6573 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2308/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6646 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2309/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6584 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2310/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6636 - val_loss: 15.8500 - val_acc: 0.6736\n",
      "Epoch 2311/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6646 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2312/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6573 - val_loss: 15.8499 - val_acc: 0.6736\n",
      "Epoch 2313/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5851 - acc: 0.6605 - val_loss: 15.8498 - val_acc: 0.6736\n",
      "Epoch 2314/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6708 - val_loss: 15.8498 - val_acc: 0.6736\n",
      "Epoch 2315/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6656 - val_loss: 15.8498 - val_acc: 0.6736\n",
      "Epoch 2316/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6605 - val_loss: 15.8497 - val_acc: 0.6736\n",
      "Epoch 2317/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6605 - val_loss: 15.8496 - val_acc: 0.6736\n",
      "Epoch 2318/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6625 - val_loss: 15.8496 - val_acc: 0.6736\n",
      "Epoch 2319/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6625 - val_loss: 15.8496 - val_acc: 0.6736\n",
      "Epoch 2320/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5954 - acc: 0.6646 - val_loss: 15.8495 - val_acc: 0.6736\n",
      "Epoch 2321/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6646 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2322/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5878 - acc: 0.6656 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2323/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6532 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2324/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6656 - val_loss: 15.8495 - val_acc: 0.6736\n",
      "Epoch 2325/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6615 - val_loss: 15.8495 - val_acc: 0.6736\n",
      "Epoch 2326/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5955 - acc: 0.6646 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2327/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5990 - acc: 0.6646 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2328/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6646 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2329/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6625 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2330/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5941 - acc: 0.6677 - val_loss: 15.8492 - val_acc: 0.6736\n",
      "Epoch 2331/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6594 - val_loss: 15.8492 - val_acc: 0.6736\n",
      "Epoch 2332/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6636 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2333/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6594 - val_loss: 15.8492 - val_acc: 0.6736\n",
      "Epoch 2334/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6636 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2335/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6050 - acc: 0.6573 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2336/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6636 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2337/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6698 - val_loss: 15.8494 - val_acc: 0.6736\n",
      "Epoch 2338/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6594 - val_loss: 15.8493 - val_acc: 0.6736\n",
      "Epoch 2339/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6687 - val_loss: 15.8492 - val_acc: 0.6736\n",
      "Epoch 2340/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5951 - acc: 0.6625 - val_loss: 15.8492 - val_acc: 0.6736\n",
      "Epoch 2341/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6656 - val_loss: 15.8491 - val_acc: 0.6736\n",
      "Epoch 2342/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6584 - val_loss: 15.8490 - val_acc: 0.6736\n",
      "Epoch 2343/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6615 - val_loss: 15.8490 - val_acc: 0.6736\n",
      "Epoch 2344/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6718 - val_loss: 15.8489 - val_acc: 0.6736\n",
      "Epoch 2345/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6667 - val_loss: 15.8488 - val_acc: 0.6736\n",
      "Epoch 2346/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5965 - acc: 0.6656 - val_loss: 15.8489 - val_acc: 0.6736\n",
      "Epoch 2347/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6729 - val_loss: 15.8488 - val_acc: 0.6736\n",
      "Epoch 2348/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6594 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2349/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6687 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2350/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6781 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2351/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6687 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2352/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6698 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2353/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6636 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2354/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6142 - acc: 0.6677 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2355/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6636 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2356/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5979 - acc: 0.6646 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2357/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6708 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2358/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6605 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2359/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6625 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2360/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6698 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2361/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6563 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2362/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5986 - acc: 0.6605 - val_loss: 15.8489 - val_acc: 0.6777\n",
      "Epoch 2363/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6625 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2364/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6687 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2365/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6667 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2366/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6625 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2367/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5971 - acc: 0.6573 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2368/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6667 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2369/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5609 - acc: 0.6677 - val_loss: 15.8488 - val_acc: 0.6777\n",
      "Epoch 2370/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6698 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2371/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6667 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2372/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6625 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2373/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6594 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2374/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5985 - acc: 0.6677 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2375/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6028 - acc: 0.6636 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2376/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6563 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2377/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6729 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2378/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6698 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2379/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6698 - val_loss: 15.8487 - val_acc: 0.6777\n",
      "Epoch 2380/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5905 - acc: 0.6656 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2381/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6698 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2382/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6677 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2383/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6718 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2384/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6677 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2385/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6542 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2386/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6698 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2387/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6656 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2388/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6698 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2389/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6625 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2390/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5949 - acc: 0.6553 - val_loss: 15.8485 - val_acc: 0.6777\n",
      "Epoch 2391/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6625 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2392/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6023 - acc: 0.6667 - val_loss: 15.8486 - val_acc: 0.6777\n",
      "Epoch 2393/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6656 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2394/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6656 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2395/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6708 - val_loss: 15.8483 - val_acc: 0.6777\n",
      "Epoch 2396/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6667 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2397/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6708 - val_loss: 15.8483 - val_acc: 0.6777\n",
      "Epoch 2398/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6687 - val_loss: 15.8484 - val_acc: 0.6777\n",
      "Epoch 2399/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6667 - val_loss: 15.8483 - val_acc: 0.6777\n",
      "Epoch 2400/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6667 - val_loss: 15.8483 - val_acc: 0.6777\n",
      "Epoch 2401/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6625 - val_loss: 15.8483 - val_acc: 0.6777\n",
      "Epoch 2402/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6667 - val_loss: 15.8481 - val_acc: 0.6777\n",
      "Epoch 2403/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6708 - val_loss: 15.8481 - val_acc: 0.6777\n",
      "Epoch 2404/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6698 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2405/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6615 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2406/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6605 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2407/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5983 - acc: 0.6511 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2408/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6636 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2409/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6718 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2410/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6605 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2411/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6532 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2412/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6625 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2413/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6646 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2414/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6729 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2415/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6677 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2416/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6646 - val_loss: 15.8481 - val_acc: 0.6777\n",
      "Epoch 2417/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6698 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2418/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6625 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2419/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6698 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2420/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6739 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2421/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6656 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2422/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6625 - val_loss: 15.8480 - val_acc: 0.6777\n",
      "Epoch 2423/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6636 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2424/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6636 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2425/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6708 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2426/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6615 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2427/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6667 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2428/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6656 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2429/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5852 - acc: 0.6698 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2430/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6636 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2431/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5628 - acc: 0.6677 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2432/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6667 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2433/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6718 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2434/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6636 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2435/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6667 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2436/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6677 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2437/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6667 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2438/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6553 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2439/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6594 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2440/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6636 - val_loss: 15.8479 - val_acc: 0.6777\n",
      "Epoch 2441/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6594 - val_loss: 15.8478 - val_acc: 0.6777\n",
      "Epoch 2442/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6636 - val_loss: 15.8477 - val_acc: 0.6777\n",
      "Epoch 2443/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6667 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2444/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5945 - acc: 0.6687 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2445/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6646 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2446/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6646 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2447/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6656 - val_loss: 15.8477 - val_acc: 0.6777\n",
      "Epoch 2448/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6584 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2449/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6646 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2450/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6749 - val_loss: 15.8475 - val_acc: 0.6777\n",
      "Epoch 2451/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6615 - val_loss: 15.8476 - val_acc: 0.6777\n",
      "Epoch 2452/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6698 - val_loss: 15.8475 - val_acc: 0.6777\n",
      "Epoch 2453/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5586 - acc: 0.6656 - val_loss: 15.8475 - val_acc: 0.6777\n",
      "Epoch 2454/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5880 - acc: 0.6729 - val_loss: 15.8475 - val_acc: 0.6777\n",
      "Epoch 2455/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6656 - val_loss: 15.8474 - val_acc: 0.6777\n",
      "Epoch 2456/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5965 - acc: 0.6636 - val_loss: 15.8473 - val_acc: 0.6777\n",
      "Epoch 2457/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6729 - val_loss: 15.8473 - val_acc: 0.6777\n",
      "Epoch 2458/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6656 - val_loss: 15.8474 - val_acc: 0.6777\n",
      "Epoch 2459/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6625 - val_loss: 15.8474 - val_acc: 0.6777\n",
      "Epoch 2460/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6573 - val_loss: 15.8474 - val_acc: 0.6777\n",
      "Epoch 2461/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6014 - acc: 0.6667 - val_loss: 15.8473 - val_acc: 0.6777\n",
      "Epoch 2462/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6625 - val_loss: 15.8472 - val_acc: 0.6777\n",
      "Epoch 2463/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6667 - val_loss: 15.8472 - val_acc: 0.6777\n",
      "Epoch 2464/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6646 - val_loss: 15.8472 - val_acc: 0.6777\n",
      "Epoch 2465/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6584 - val_loss: 15.8470 - val_acc: 0.6777\n",
      "Epoch 2466/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6636 - val_loss: 15.8471 - val_acc: 0.6777\n",
      "Epoch 2467/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6625 - val_loss: 15.8470 - val_acc: 0.6777\n",
      "Epoch 2468/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6615 - val_loss: 15.8470 - val_acc: 0.6777\n",
      "Epoch 2469/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6594 - val_loss: 15.8471 - val_acc: 0.6777\n",
      "Epoch 2470/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6667 - val_loss: 15.8470 - val_acc: 0.6777\n",
      "Epoch 2471/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5852 - acc: 0.6667 - val_loss: 15.8469 - val_acc: 0.6777\n",
      "Epoch 2472/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5947 - acc: 0.6636 - val_loss: 15.8469 - val_acc: 0.6777\n",
      "Epoch 2473/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5989 - acc: 0.6656 - val_loss: 15.8468 - val_acc: 0.6777\n",
      "Epoch 2474/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6749 - val_loss: 15.8468 - val_acc: 0.6777\n",
      "Epoch 2475/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6615 - val_loss: 15.8468 - val_acc: 0.6777\n",
      "Epoch 2476/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6667 - val_loss: 15.8468 - val_acc: 0.6777\n",
      "Epoch 2477/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6636 - val_loss: 15.8468 - val_acc: 0.6777\n",
      "Epoch 2478/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6708 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2479/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6677 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2480/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6573 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2481/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5828 - acc: 0.6636 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2482/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6625 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2483/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6563 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2484/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6698 - val_loss: 15.8467 - val_acc: 0.6777\n",
      "Epoch 2485/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6584 - val_loss: 15.8466 - val_acc: 0.6777\n",
      "Epoch 2486/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6729 - val_loss: 15.8466 - val_acc: 0.6777\n",
      "Epoch 2487/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5901 - acc: 0.6729 - val_loss: 15.8466 - val_acc: 0.6777\n",
      "Epoch 2488/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6656 - val_loss: 15.8466 - val_acc: 0.6777\n",
      "Epoch 2489/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6605 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2490/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6119 - acc: 0.6553 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2491/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6615 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2492/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5588 - acc: 0.6656 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2493/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6605 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2494/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6698 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2495/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5858 - acc: 0.6625 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2496/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6646 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2497/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6646 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2498/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6687 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2499/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6584 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2500/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6636 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2501/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6625 - val_loss: 15.8465 - val_acc: 0.6777\n",
      "Epoch 2502/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6667 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2503/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6636 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2504/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6749 - val_loss: 15.8464 - val_acc: 0.6777\n",
      "Epoch 2505/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5908 - acc: 0.6636 - val_loss: 15.8463 - val_acc: 0.6777\n",
      "Epoch 2506/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6667 - val_loss: 15.8462 - val_acc: 0.6777\n",
      "Epoch 2507/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6553 - val_loss: 15.8463 - val_acc: 0.6777\n",
      "Epoch 2508/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6636 - val_loss: 15.8462 - val_acc: 0.6777\n",
      "Epoch 2509/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5985 - acc: 0.6605 - val_loss: 15.8462 - val_acc: 0.6777\n",
      "Epoch 2510/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6563 - val_loss: 15.8461 - val_acc: 0.6777\n",
      "Epoch 2511/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6594 - val_loss: 15.8462 - val_acc: 0.6777\n",
      "Epoch 2512/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6636 - val_loss: 15.8462 - val_acc: 0.6777\n",
      "Epoch 2513/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5674 - acc: 0.6677 - val_loss: 15.8461 - val_acc: 0.6777\n",
      "Epoch 2514/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5887 - acc: 0.6594 - val_loss: 15.8460 - val_acc: 0.6777\n",
      "Epoch 2515/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6646 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2516/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6584 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2517/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6698 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2518/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6718 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2519/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5585 - acc: 0.6739 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2520/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6625 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2521/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6594 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2522/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6718 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2523/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6636 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2524/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6667 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2525/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6646 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2526/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6656 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2527/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6708 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2528/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6708 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2529/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5909 - acc: 0.6687 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2530/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5841 - acc: 0.6677 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2531/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6646 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2532/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6605 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2533/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5905 - acc: 0.6687 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2534/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6729 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2535/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6708 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2536/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5598 - acc: 0.6573 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2537/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6605 - val_loss: 15.8459 - val_acc: 0.6777\n",
      "Epoch 2538/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6687 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2539/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5698 - acc: 0.6677 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2540/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6646 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2541/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5698 - acc: 0.6625 - val_loss: 15.8458 - val_acc: 0.6777\n",
      "Epoch 2542/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6667 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2543/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6760 - val_loss: 15.8456 - val_acc: 0.6777\n",
      "Epoch 2544/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6573 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2545/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6708 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2546/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6080 - acc: 0.6605 - val_loss: 15.8457 - val_acc: 0.6777\n",
      "Epoch 2547/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6594 - val_loss: 15.8456 - val_acc: 0.6777\n",
      "Epoch 2548/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6646 - val_loss: 15.8456 - val_acc: 0.6777\n",
      "Epoch 2549/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6646 - val_loss: 15.8455 - val_acc: 0.6777\n",
      "Epoch 2550/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6584 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2551/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6573 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2552/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6718 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2553/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5945 - acc: 0.6573 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2554/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6615 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2555/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6605 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2556/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5896 - acc: 0.6667 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2557/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5925 - acc: 0.6553 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2558/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6667 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2559/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5908 - acc: 0.6605 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2560/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5915 - acc: 0.6687 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2561/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6801 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2562/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6636 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2563/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5858 - acc: 0.6667 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2564/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6646 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2565/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6708 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2566/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6698 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2567/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6594 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2568/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6770 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2569/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6708 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2570/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6667 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2571/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6667 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2572/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6718 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2573/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6698 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2574/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6656 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2575/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6584 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2576/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6656 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2577/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6698 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2578/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6553 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2579/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6656 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2580/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6594 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2581/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6594 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2582/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6698 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2583/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6584 - val_loss: 15.8455 - val_acc: 0.6777\n",
      "Epoch 2584/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6625 - val_loss: 15.8455 - val_acc: 0.6777\n",
      "Epoch 2585/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6739 - val_loss: 15.8455 - val_acc: 0.6777\n",
      "Epoch 2586/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6687 - val_loss: 15.8455 - val_acc: 0.6777\n",
      "Epoch 2587/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6636 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2588/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5617 - acc: 0.6563 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2589/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6615 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2590/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6739 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2591/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6605 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2592/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5956 - acc: 0.6708 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2593/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6667 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2594/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5933 - acc: 0.6636 - val_loss: 15.8454 - val_acc: 0.6777\n",
      "Epoch 2595/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6687 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2596/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5878 - acc: 0.6615 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2597/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6781 - val_loss: 15.8452 - val_acc: 0.6777\n",
      "Epoch 2598/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6698 - val_loss: 15.8453 - val_acc: 0.6777\n",
      "Epoch 2599/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5839 - acc: 0.6594 - val_loss: 15.8452 - val_acc: 0.6777\n",
      "Epoch 2600/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6749 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2601/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5901 - acc: 0.6656 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2602/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6698 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2603/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5897 - acc: 0.6667 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2604/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6636 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2605/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6687 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2606/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6646 - val_loss: 15.8451 - val_acc: 0.6777\n",
      "Epoch 2607/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5944 - acc: 0.6636 - val_loss: 15.8450 - val_acc: 0.6777\n",
      "Epoch 2608/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6573 - val_loss: 15.8450 - val_acc: 0.6777\n",
      "Epoch 2609/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6646 - val_loss: 15.8450 - val_acc: 0.6777\n",
      "Epoch 2610/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5862 - acc: 0.6594 - val_loss: 15.8450 - val_acc: 0.6777\n",
      "Epoch 2611/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5909 - acc: 0.6636 - val_loss: 15.8450 - val_acc: 0.6777\n",
      "Epoch 2612/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6687 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2613/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6687 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2614/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6667 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2615/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6013 - acc: 0.6625 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2616/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6625 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2617/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6687 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2618/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6708 - val_loss: 15.8449 - val_acc: 0.6777\n",
      "Epoch 2619/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6625 - val_loss: 15.8448 - val_acc: 0.6777\n",
      "Epoch 2620/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6636 - val_loss: 15.8448 - val_acc: 0.6777\n",
      "Epoch 2621/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6605 - val_loss: 15.8448 - val_acc: 0.6777\n",
      "Epoch 2622/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6770 - val_loss: 15.8448 - val_acc: 0.6777\n",
      "Epoch 2623/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6563 - val_loss: 15.8447 - val_acc: 0.6777\n",
      "Epoch 2624/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6729 - val_loss: 15.8447 - val_acc: 0.6777\n",
      "Epoch 2625/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6667 - val_loss: 15.8446 - val_acc: 0.6777\n",
      "Epoch 2626/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5633 - acc: 0.6739 - val_loss: 15.8446 - val_acc: 0.6777\n",
      "Epoch 2627/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6749 - val_loss: 15.8445 - val_acc: 0.6777\n",
      "Epoch 2628/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6615 - val_loss: 15.8445 - val_acc: 0.6777\n",
      "Epoch 2629/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6656 - val_loss: 15.8444 - val_acc: 0.6777\n",
      "Epoch 2630/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6636 - val_loss: 15.8444 - val_acc: 0.6777\n",
      "Epoch 2631/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5839 - acc: 0.6656 - val_loss: 15.8444 - val_acc: 0.6777\n",
      "Epoch 2632/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6687 - val_loss: 15.8443 - val_acc: 0.6777\n",
      "Epoch 2633/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6646 - val_loss: 15.8443 - val_acc: 0.6777\n",
      "Epoch 2634/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6667 - val_loss: 15.8443 - val_acc: 0.6777\n",
      "Epoch 2635/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6043 - acc: 0.6718 - val_loss: 15.8443 - val_acc: 0.6777\n",
      "Epoch 2636/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6708 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2637/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6615 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2638/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6615 - val_loss: 15.8443 - val_acc: 0.6777\n",
      "Epoch 2639/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5989 - acc: 0.6687 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2640/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6605 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2641/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6739 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2642/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6667 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2643/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6615 - val_loss: 15.8442 - val_acc: 0.6777\n",
      "Epoch 2644/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6656 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2645/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6687 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2646/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6584 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2647/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6636 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2648/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6687 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2649/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6594 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2650/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6667 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2651/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6687 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2652/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6636 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2653/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6687 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2654/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6625 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2655/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6708 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2656/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6594 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2657/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6687 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2658/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6677 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2659/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6749 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2660/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6646 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2661/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6594 - val_loss: 15.8439 - val_acc: 0.6777\n",
      "Epoch 2662/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6698 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2663/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6677 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2664/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6646 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2665/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6770 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2666/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6605 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2667/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6677 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2668/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6677 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2669/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6636 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2670/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6646 - val_loss: 15.8441 - val_acc: 0.6777\n",
      "Epoch 2671/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5851 - acc: 0.6698 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2672/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6677 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2673/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5982 - acc: 0.6646 - val_loss: 15.8440 - val_acc: 0.6777\n",
      "Epoch 2674/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6009 - acc: 0.6656 - val_loss: 15.8439 - val_acc: 0.6777\n",
      "Epoch 2675/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6667 - val_loss: 15.8439 - val_acc: 0.6777\n",
      "Epoch 2676/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6667 - val_loss: 15.8439 - val_acc: 0.6777\n",
      "Epoch 2677/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6667 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2678/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6667 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2679/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5850 - acc: 0.6656 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2680/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6739 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2681/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6104 - acc: 0.6615 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2682/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6687 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2683/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6667 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2684/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6718 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2685/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6698 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2686/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6677 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2687/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6687 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2688/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5594 - acc: 0.6646 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2689/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6656 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2690/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6708 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2691/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6656 - val_loss: 15.8438 - val_acc: 0.6777\n",
      "Epoch 2692/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6656 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2693/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6656 - val_loss: 15.8437 - val_acc: 0.6777\n",
      "Epoch 2694/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6584 - val_loss: 15.8436 - val_acc: 0.6777\n",
      "Epoch 2695/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6615 - val_loss: 15.8436 - val_acc: 0.6777\n",
      "Epoch 2696/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6656 - val_loss: 15.8436 - val_acc: 0.6777\n",
      "Epoch 2697/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6770 - val_loss: 15.8435 - val_acc: 0.6777\n",
      "Epoch 2698/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5636 - acc: 0.6656 - val_loss: 15.8435 - val_acc: 0.6777\n",
      "Epoch 2699/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6708 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2700/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6677 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2701/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6584 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2702/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6656 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2703/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5716 - acc: 0.6646 - val_loss: 15.8435 - val_acc: 0.6777\n",
      "Epoch 2704/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6605 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2705/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6677 - val_loss: 15.8434 - val_acc: 0.6777\n",
      "Epoch 2706/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6708 - val_loss: 15.8433 - val_acc: 0.6777\n",
      "Epoch 2707/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6656 - val_loss: 15.8433 - val_acc: 0.6777\n",
      "Epoch 2708/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5982 - acc: 0.6625 - val_loss: 15.8432 - val_acc: 0.6777\n",
      "Epoch 2709/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6677 - val_loss: 15.8432 - val_acc: 0.6777\n",
      "Epoch 2710/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6687 - val_loss: 15.8432 - val_acc: 0.6777\n",
      "Epoch 2711/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6615 - val_loss: 15.8432 - val_acc: 0.6777\n",
      "Epoch 2712/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6718 - val_loss: 15.8431 - val_acc: 0.6777\n",
      "Epoch 2713/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6605 - val_loss: 15.8432 - val_acc: 0.6777\n",
      "Epoch 2714/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6563 - val_loss: 15.8431 - val_acc: 0.6777\n",
      "Epoch 2715/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6739 - val_loss: 15.8431 - val_acc: 0.6777\n",
      "Epoch 2716/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6625 - val_loss: 15.8431 - val_acc: 0.6777\n",
      "Epoch 2717/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6636 - val_loss: 15.8431 - val_acc: 0.6777\n",
      "Epoch 2718/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6739 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2719/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6708 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2720/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6677 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2721/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6656 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2722/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6708 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2723/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6687 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2724/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6718 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2725/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6656 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2726/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6708 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2727/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6625 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2728/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6677 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2729/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6718 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2730/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6708 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2731/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5938 - acc: 0.6646 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2732/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6677 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2733/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6636 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2734/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6656 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2735/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6615 - val_loss: 15.8430 - val_acc: 0.6777\n",
      "Epoch 2736/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5993 - acc: 0.6656 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2737/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6656 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2738/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5956 - acc: 0.6646 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2739/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6615 - val_loss: 15.8429 - val_acc: 0.6777\n",
      "Epoch 2740/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6698 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2741/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6615 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2742/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5996 - acc: 0.6749 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2743/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6708 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2744/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6553 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2745/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5959 - acc: 0.6636 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2746/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6708 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2747/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6739 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2748/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6677 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2749/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6605 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2750/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6480 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2751/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6698 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2752/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6605 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2753/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6687 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2754/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6656 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2755/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6636 - val_loss: 15.8428 - val_acc: 0.6777\n",
      "Epoch 2756/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6698 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2757/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6729 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2758/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6625 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2759/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6698 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2760/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6563 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2761/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6729 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2762/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5941 - acc: 0.6667 - val_loss: 15.8427 - val_acc: 0.6777\n",
      "Epoch 2763/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6708 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2764/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5767 - acc: 0.6718 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2765/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6636 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2766/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6698 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2767/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6646 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2768/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6052 - acc: 0.6636 - val_loss: 15.8426 - val_acc: 0.6777\n",
      "Epoch 2769/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5617 - acc: 0.6677 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2770/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6553 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2771/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6646 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2772/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6594 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2773/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6698 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2774/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6760 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2775/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6708 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2776/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6656 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2777/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6615 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2778/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6667 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2779/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6636 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2780/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6656 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2781/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6687 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2782/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6573 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2783/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6584 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2784/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6729 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2785/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6646 - val_loss: 15.8425 - val_acc: 0.6777\n",
      "Epoch 2786/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6718 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2787/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6656 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2788/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6594 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2789/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6636 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2790/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6553 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2791/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6646 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2792/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6667 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2793/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6687 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2794/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6677 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2795/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5652 - acc: 0.6594 - val_loss: 15.8424 - val_acc: 0.6777\n",
      "Epoch 2796/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6605 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2797/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6553 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2798/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6605 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2799/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6708 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2800/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6615 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2801/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6708 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2802/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6708 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2803/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6708 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2804/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6625 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2805/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6749 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2806/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6646 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2807/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6625 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2808/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6605 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2809/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6667 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2810/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6636 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2811/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6646 - val_loss: 15.8423 - val_acc: 0.6777\n",
      "Epoch 2812/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6739 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2813/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6646 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2814/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6739 - val_loss: 15.8422 - val_acc: 0.6777\n",
      "Epoch 2815/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6677 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2816/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6677 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2817/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5541 - acc: 0.6646 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2818/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6656 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2819/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6636 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2820/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5716 - acc: 0.6625 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2821/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6656 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2822/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6646 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2823/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6708 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2824/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6646 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2825/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6573 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2826/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6718 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2827/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6667 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2828/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5897 - acc: 0.6667 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2829/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5908 - acc: 0.6677 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2830/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6605 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2831/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6667 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2832/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6687 - val_loss: 15.8419 - val_acc: 0.6777\n",
      "Epoch 2833/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6729 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2834/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5922 - acc: 0.6698 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2835/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6615 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2836/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6656 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2837/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5857 - acc: 0.6605 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2838/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6615 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2839/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6656 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2840/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5541 - acc: 0.6625 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2841/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6656 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2842/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6677 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2843/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6698 - val_loss: 15.8421 - val_acc: 0.6777\n",
      "Epoch 2844/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6677 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2845/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5622 - acc: 0.6615 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2846/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6636 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2847/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6656 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2848/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6677 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2849/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5714 - acc: 0.6677 - val_loss: 15.8420 - val_acc: 0.6777\n",
      "Epoch 2850/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6615 - val_loss: 15.8419 - val_acc: 0.6777\n",
      "Epoch 2851/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6749 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2852/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6553 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2853/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6656 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2854/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6708 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2855/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5870 - acc: 0.6636 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2856/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6615 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2857/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6698 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2858/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6718 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2859/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6625 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2860/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6729 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2861/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6615 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2862/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6760 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2863/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6563 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2864/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6646 - val_loss: 15.8418 - val_acc: 0.6777\n",
      "Epoch 2865/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5920 - acc: 0.6687 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2866/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6729 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2867/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6646 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2868/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6718 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2869/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6656 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2870/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6667 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2871/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6667 - val_loss: 15.8417 - val_acc: 0.6777\n",
      "Epoch 2872/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6625 - val_loss: 15.8416 - val_acc: 0.6777\n",
      "Epoch 2873/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6698 - val_loss: 15.8416 - val_acc: 0.6777\n",
      "Epoch 2874/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6687 - val_loss: 15.8416 - val_acc: 0.6777\n",
      "Epoch 2875/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6718 - val_loss: 15.8416 - val_acc: 0.6777\n",
      "Epoch 2876/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6012 - acc: 0.6729 - val_loss: 15.8416 - val_acc: 0.6777\n",
      "Epoch 2877/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6729 - val_loss: 15.8415 - val_acc: 0.6777\n",
      "Epoch 2878/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5896 - acc: 0.6646 - val_loss: 15.8415 - val_acc: 0.6777\n",
      "Epoch 2879/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6718 - val_loss: 15.8414 - val_acc: 0.6777\n",
      "Epoch 2880/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6646 - val_loss: 15.8414 - val_acc: 0.6777\n",
      "Epoch 2881/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6667 - val_loss: 15.8414 - val_acc: 0.6777\n",
      "Epoch 2882/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6677 - val_loss: 15.8414 - val_acc: 0.6777\n",
      "Epoch 2883/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6687 - val_loss: 15.8414 - val_acc: 0.6777\n",
      "Epoch 2884/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6739 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2885/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6615 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2886/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6718 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2887/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6646 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2888/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6636 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2889/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6687 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2890/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6625 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2891/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6646 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2892/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6729 - val_loss: 15.8413 - val_acc: 0.6777\n",
      "Epoch 2893/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6636 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2894/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6667 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2895/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6615 - val_loss: 15.8411 - val_acc: 0.6777\n",
      "Epoch 2896/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6594 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2897/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6698 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2898/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5935 - acc: 0.6656 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2899/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6708 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2900/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6605 - val_loss: 15.8411 - val_acc: 0.6777\n",
      "Epoch 2901/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5652 - acc: 0.6636 - val_loss: 15.8412 - val_acc: 0.6777\n",
      "Epoch 2902/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6687 - val_loss: 15.8411 - val_acc: 0.6777\n",
      "Epoch 2903/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6687 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2904/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6625 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2905/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5578 - acc: 0.6667 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2906/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5607 - acc: 0.6687 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2907/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5887 - acc: 0.6791 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2908/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6667 - val_loss: 15.8411 - val_acc: 0.6777\n",
      "Epoch 2909/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6646 - val_loss: 15.8410 - val_acc: 0.6777\n",
      "Epoch 2910/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6625 - val_loss: 15.8409 - val_acc: 0.6777\n",
      "Epoch 2911/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6760 - val_loss: 15.8409 - val_acc: 0.6777\n",
      "Epoch 2912/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6749 - val_loss: 15.8408 - val_acc: 0.6777\n",
      "Epoch 2913/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6646 - val_loss: 15.8408 - val_acc: 0.6777\n",
      "Epoch 2914/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6636 - val_loss: 15.8408 - val_acc: 0.6777\n",
      "Epoch 2915/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6605 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2916/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6594 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2917/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5906 - acc: 0.6718 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2918/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6605 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2919/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6677 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2920/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6708 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2921/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6687 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2922/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6687 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2923/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6667 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2924/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6646 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2925/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6687 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2926/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6708 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2927/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6625 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2928/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6667 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2929/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6718 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2930/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5880 - acc: 0.6636 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2931/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6760 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2932/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5887 - acc: 0.6573 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2933/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6677 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2934/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6760 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2935/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6677 - val_loss: 15.8408 - val_acc: 0.6777\n",
      "Epoch 2936/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6687 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2937/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6708 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2938/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6698 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2939/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6636 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2940/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6646 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2941/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6677 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2942/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6605 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2943/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6656 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2944/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6573 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2945/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6553 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2946/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6656 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2947/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6729 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2948/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6656 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2949/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6749 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2950/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6698 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2951/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6646 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2952/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6760 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2953/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6636 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2954/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6718 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2955/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6687 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2956/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6687 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2957/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6615 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2958/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5828 - acc: 0.6718 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2959/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6718 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2960/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6708 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2961/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6749 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2962/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6594 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2963/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6677 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2964/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6677 - val_loss: 15.8407 - val_acc: 0.6777\n",
      "Epoch 2965/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6615 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2966/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6136 - acc: 0.6677 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2967/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6677 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2968/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5551 - acc: 0.6729 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2969/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6687 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2970/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6636 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2971/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6729 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2972/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6646 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2973/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6677 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2974/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6584 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2975/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6698 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2976/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6646 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2977/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6636 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2978/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6760 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2979/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6667 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2980/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6698 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2981/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5908 - acc: 0.6687 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2982/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6646 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2983/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6760 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2984/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6646 - val_loss: 15.8406 - val_acc: 0.6777\n",
      "Epoch 2985/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6687 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2986/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6781 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2987/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5881 - acc: 0.6646 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2988/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.5682 - acc: 0.66 - 0s - loss: 16.5633 - acc: 0.6615 - val_loss: 15.8405 - val_acc: 0.6777\n",
      "Epoch 2989/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6605 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2990/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5618 - acc: 0.6708 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2991/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6708 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2992/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6646 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2993/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6760 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2994/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6708 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2995/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6718 - val_loss: 15.8404 - val_acc: 0.6777\n",
      "Epoch 2996/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5961 - acc: 0.6625 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 2997/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6636 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 2998/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5958 - acc: 0.6677 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 2999/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6615 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3000/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5957 - acc: 0.6718 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 3001/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6687 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 3002/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6729 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3003/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6781 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3004/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6770 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3005/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6687 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 3006/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6656 - val_loss: 15.8403 - val_acc: 0.6777\n",
      "Epoch 3007/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6687 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3008/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5995 - acc: 0.6677 - val_loss: 15.8402 - val_acc: 0.6777\n",
      "Epoch 3009/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6677 - val_loss: 15.8401 - val_acc: 0.6777\n",
      "Epoch 3010/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6687 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3011/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5870 - acc: 0.6667 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3012/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6656 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3013/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6615 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3014/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6563 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3015/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6739 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3016/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5808 - acc: 0.6708 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3017/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6718 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3018/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6708 - val_loss: 15.8401 - val_acc: 0.6777\n",
      "Epoch 3019/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6646 - val_loss: 15.8401 - val_acc: 0.6777\n",
      "Epoch 3020/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5674 - acc: 0.6718 - val_loss: 15.8401 - val_acc: 0.6777\n",
      "Epoch 3021/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6729 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3022/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6594 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3023/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6656 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3024/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6636 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3025/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5927 - acc: 0.6698 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3026/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6605 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3027/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6718 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3028/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6770 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3029/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6770 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3030/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.6608 - acc: 0.66 - 0s - loss: 16.5801 - acc: 0.6625 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3031/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6667 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3032/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6739 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3033/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6605 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3034/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6636 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3035/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6047 - acc: 0.6605 - val_loss: 15.8400 - val_acc: 0.6777\n",
      "Epoch 3036/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6656 - val_loss: 15.8399 - val_acc: 0.6777\n",
      "Epoch 3037/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6636 - val_loss: 15.8398 - val_acc: 0.6777\n",
      "Epoch 3038/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6656 - val_loss: 15.8398 - val_acc: 0.6777\n",
      "Epoch 3039/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6646 - val_loss: 15.8398 - val_acc: 0.6777\n",
      "Epoch 3040/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5650 - acc: 0.6656 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3041/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6718 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3042/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6760 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3043/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6749 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3044/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6677 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3045/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6687 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3046/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6667 - val_loss: 15.8396 - val_acc: 0.6777\n",
      "Epoch 3047/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6573 - val_loss: 15.8397 - val_acc: 0.6777\n",
      "Epoch 3048/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6677 - val_loss: 15.8396 - val_acc: 0.6777\n",
      "Epoch 3049/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6594 - val_loss: 15.8396 - val_acc: 0.6777\n",
      "Epoch 3050/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6636 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3051/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5602 - acc: 0.6677 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3052/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6708 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3053/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6749 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3054/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6667 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3055/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6656 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3056/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6687 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3057/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6729 - val_loss: 15.8395 - val_acc: 0.6777\n",
      "Epoch 3058/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6687 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3059/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5828 - acc: 0.6718 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3060/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6646 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3061/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6656 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3062/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6677 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3063/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6646 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3064/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6625 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3065/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6770 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3066/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6625 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3067/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5534 - acc: 0.6677 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3068/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6594 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3069/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6718 - val_loss: 15.8394 - val_acc: 0.6777\n",
      "Epoch 3070/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5926 - acc: 0.6729 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3071/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6667 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3072/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6625 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3073/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6677 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3074/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6687 - val_loss: 15.8393 - val_acc: 0.6777\n",
      "Epoch 3075/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6687 - val_loss: 15.8392 - val_acc: 0.6777\n",
      "Epoch 3076/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6698 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3077/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6770 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3078/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6646 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3079/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6605 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3080/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6718 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3081/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6625 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3082/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6687 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3083/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6698 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3084/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5774 - acc: 0.6687 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3085/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6687 - val_loss: 15.8391 - val_acc: 0.6777\n",
      "Epoch 3086/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6822 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3087/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6656 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3088/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6677 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3089/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6656 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3090/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6656 - val_loss: 15.8389 - val_acc: 0.6777\n",
      "Epoch 3091/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6760 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3092/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6615 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3093/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6646 - val_loss: 15.8390 - val_acc: 0.6777\n",
      "Epoch 3094/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6615 - val_loss: 15.8389 - val_acc: 0.6777\n",
      "Epoch 3095/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5573 - acc: 0.6801 - val_loss: 15.8389 - val_acc: 0.6777\n",
      "Epoch 3096/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6563 - val_loss: 15.8389 - val_acc: 0.6777\n",
      "Epoch 3097/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6605 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3098/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6646 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3099/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5777 - acc: 0.6749 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3100/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5947 - acc: 0.6718 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3101/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5552 - acc: 0.6605 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3102/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6615 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3103/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6687 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3104/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6636 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3105/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6553 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3106/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6698 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3107/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6687 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3108/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6687 - val_loss: 15.8389 - val_acc: 0.6777\n",
      "Epoch 3109/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6615 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3110/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5624 - acc: 0.6677 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3111/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6770 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3112/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6615 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3113/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6646 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3114/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6677 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3115/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6584 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3116/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6698 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3117/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6677 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3118/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6605 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3119/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6739 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3120/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6718 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3121/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6656 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3122/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6708 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3123/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5564 - acc: 0.6625 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3124/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6656 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3125/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6656 - val_loss: 15.8388 - val_acc: 0.6777\n",
      "Epoch 3126/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6687 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3127/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6677 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3128/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6677 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3129/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6708 - val_loss: 15.8387 - val_acc: 0.6777\n",
      "Epoch 3130/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6760 - val_loss: 15.8386 - val_acc: 0.6777\n",
      "Epoch 3131/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6677 - val_loss: 15.8386 - val_acc: 0.6777\n",
      "Epoch 3132/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6708 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3133/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5650 - acc: 0.6656 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3134/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5714 - acc: 0.6615 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3135/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6646 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3136/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5939 - acc: 0.6625 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3137/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6636 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3138/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5777 - acc: 0.6677 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3139/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6584 - val_loss: 15.8385 - val_acc: 0.6777\n",
      "Epoch 3140/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6656 - val_loss: 15.8384 - val_acc: 0.6777\n",
      "Epoch 3141/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6729 - val_loss: 15.8384 - val_acc: 0.6777\n",
      "Epoch 3142/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6646 - val_loss: 15.8383 - val_acc: 0.6777\n",
      "Epoch 3143/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5936 - acc: 0.6677 - val_loss: 15.8383 - val_acc: 0.6777\n",
      "Epoch 3144/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6698 - val_loss: 15.8383 - val_acc: 0.6777\n",
      "Epoch 3145/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5668 - acc: 0.6677 - val_loss: 15.8383 - val_acc: 0.6777\n",
      "Epoch 3146/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6749 - val_loss: 15.8382 - val_acc: 0.6777\n",
      "Epoch 3147/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - ETA: 0s - loss: 16.6699 - acc: 0.65 - 0s - loss: 16.5806 - acc: 0.6677 - val_loss: 15.8382 - val_acc: 0.6777\n",
      "Epoch 3148/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6646 - val_loss: 15.8381 - val_acc: 0.6777\n",
      "Epoch 3149/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6656 - val_loss: 15.8381 - val_acc: 0.6777\n",
      "Epoch 3150/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6698 - val_loss: 15.8381 - val_acc: 0.6777\n",
      "Epoch 3151/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6625 - val_loss: 15.8382 - val_acc: 0.6777\n",
      "Epoch 3152/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6677 - val_loss: 15.8381 - val_acc: 0.6777\n",
      "Epoch 3153/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5850 - acc: 0.6677 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3154/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6646 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3155/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6605 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3156/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6698 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3157/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6636 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3158/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6739 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3159/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6718 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3160/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6656 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3161/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6677 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3162/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5943 - acc: 0.6677 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3163/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6708 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3164/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3165/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3166/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6677 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3167/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6739 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3168/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6605 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3169/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6677 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3170/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6667 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3171/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5473 - acc: 0.6687 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3172/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6542 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3173/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6687 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3174/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3175/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3176/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6656 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3177/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3178/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6656 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3179/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6625 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3180/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5841 - acc: 0.6708 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3181/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6038 - acc: 0.6646 - val_loss: 15.8380 - val_acc: 0.6777\n",
      "Epoch 3182/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6646 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3183/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6563 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3184/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5629 - acc: 0.6677 - val_loss: 15.8379 - val_acc: 0.6777\n",
      "Epoch 3185/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6760 - val_loss: 15.8378 - val_acc: 0.6777\n",
      "Epoch 3186/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6625 - val_loss: 15.8378 - val_acc: 0.6777\n",
      "Epoch 3187/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6677 - val_loss: 15.8378 - val_acc: 0.6777\n",
      "Epoch 3188/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5837 - acc: 0.6677 - val_loss: 15.8377 - val_acc: 0.6777\n",
      "Epoch 3189/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6667 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3190/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6708 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3191/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6677 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3192/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6667 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3193/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6729 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3194/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6677 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3195/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6677 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3196/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6646 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3197/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6770 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3198/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5598 - acc: 0.6636 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3199/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6708 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3200/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6677 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3201/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6646 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3202/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6615 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3203/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5928 - acc: 0.6718 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3204/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5648 - acc: 0.6646 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3205/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6646 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3206/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6656 - val_loss: 15.8377 - val_acc: 0.6777\n",
      "Epoch 3207/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6656 - val_loss: 15.8377 - val_acc: 0.6777\n",
      "Epoch 3208/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6729 - val_loss: 15.8377 - val_acc: 0.6777\n",
      "Epoch 3209/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6667 - val_loss: 15.8377 - val_acc: 0.6777\n",
      "Epoch 3210/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6739 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3211/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5602 - acc: 0.6646 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3212/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5964 - acc: 0.6646 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3213/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6698 - val_loss: 15.8376 - val_acc: 0.6777\n",
      "Epoch 3214/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6698 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3215/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6636 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3216/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6718 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3217/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6656 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3218/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6003 - acc: 0.6718 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3219/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6698 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3220/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6646 - val_loss: 15.8375 - val_acc: 0.6777\n",
      "Epoch 3221/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5977 - acc: 0.6781 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3222/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6698 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3223/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6667 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3224/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5608 - acc: 0.6708 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3225/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6646 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3226/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6739 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3227/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6667 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3228/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6646 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3229/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6042 - acc: 0.6698 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3230/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6718 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3231/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6542 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3232/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6594 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3233/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6749 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3234/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6656 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3235/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5533 - acc: 0.6677 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3236/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6636 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3237/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6594 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3238/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6791 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3239/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6656 - val_loss: 15.8374 - val_acc: 0.6777\n",
      "Epoch 3240/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6739 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3241/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6656 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3242/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6563 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3243/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6698 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3244/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6698 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3245/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6656 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3246/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6667 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3247/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6687 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3248/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6656 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3249/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6625 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3250/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6646 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3251/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6594 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3252/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6718 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3253/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6553 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3254/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6770 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3255/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5899 - acc: 0.6708 - val_loss: 15.8373 - val_acc: 0.6777\n",
      "Epoch 3256/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6584 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3257/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6718 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3258/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6687 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3259/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5603 - acc: 0.6656 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3260/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6646 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3261/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6708 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3262/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6594 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3263/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5823 - acc: 0.6584 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3264/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6739 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3265/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6594 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3266/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6029 - acc: 0.6594 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3267/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6677 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3268/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5532 - acc: 0.6646 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3269/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6677 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3270/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6625 - val_loss: 15.8372 - val_acc: 0.6777\n",
      "Epoch 3271/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5931 - acc: 0.6749 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3272/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6677 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3273/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6625 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3274/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6687 - val_loss: 15.8371 - val_acc: 0.6777\n",
      "Epoch 3275/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6791 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3276/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6646 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3277/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6636 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3278/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6646 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3279/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6563 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3280/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6677 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3281/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6656 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3282/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6656 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3283/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6605 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3284/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6667 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3285/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6656 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3286/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6687 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3287/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6636 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3288/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6708 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3289/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6636 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3290/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6027 - acc: 0.6677 - val_loss: 15.8370 - val_acc: 0.6777\n",
      "Epoch 3291/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6615 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3292/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6646 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3293/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6656 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3294/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6667 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3295/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6553 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3296/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6646 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3297/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6698 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3298/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6563 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3299/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5714 - acc: 0.6646 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3300/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6636 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3301/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6646 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3302/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6729 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3303/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5915 - acc: 0.6625 - val_loss: 15.8369 - val_acc: 0.6777\n",
      "Epoch 3304/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6698 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3305/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6636 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3306/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6708 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3307/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6553 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3308/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6656 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3309/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6698 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3310/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6749 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3311/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6749 - val_loss: 15.8368 - val_acc: 0.6777\n",
      "Epoch 3312/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6656 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3313/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6739 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3314/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6646 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3315/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6636 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3316/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6656 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3317/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6667 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3318/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6698 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3319/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6677 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3320/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5878 - acc: 0.6646 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3321/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6615 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3322/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6646 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3323/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6687 - val_loss: 15.8367 - val_acc: 0.6777\n",
      "Epoch 3324/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6667 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3325/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6667 - val_loss: 15.8366 - val_acc: 0.6777\n",
      "Epoch 3326/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6563 - val_loss: 15.8365 - val_acc: 0.6777\n",
      "Epoch 3327/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6625 - val_loss: 15.8365 - val_acc: 0.6777\n",
      "Epoch 3328/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6687 - val_loss: 15.8365 - val_acc: 0.6777\n",
      "Epoch 3329/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6625 - val_loss: 15.8364 - val_acc: 0.6777\n",
      "Epoch 3330/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6656 - val_loss: 15.8364 - val_acc: 0.6777\n",
      "Epoch 3331/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6698 - val_loss: 15.8364 - val_acc: 0.6777\n",
      "Epoch 3332/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6749 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3333/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6698 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3334/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6729 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3335/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6584 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3336/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5828 - acc: 0.6677 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3337/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6677 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3338/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6646 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3339/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6656 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3340/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5880 - acc: 0.6708 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3341/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6594 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3342/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6770 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3343/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5652 - acc: 0.6677 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3344/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5841 - acc: 0.6646 - val_loss: 15.8363 - val_acc: 0.6777\n",
      "Epoch 3345/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5600 - acc: 0.6770 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3346/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6636 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3347/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6646 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3348/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5547 - acc: 0.6667 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3349/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5860 - acc: 0.6687 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3350/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6667 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3351/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5874 - acc: 0.6594 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3352/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6605 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3353/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6718 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3354/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6760 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3355/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6698 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3356/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6677 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3357/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6770 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3358/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6053 - acc: 0.6687 - val_loss: 15.8361 - val_acc: 0.6777\n",
      "Epoch 3359/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6594 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3360/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6687 - val_loss: 15.8362 - val_acc: 0.6777\n",
      "Epoch 3361/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5642 - acc: 0.6687 - val_loss: 15.8361 - val_acc: 0.6777\n",
      "Epoch 3362/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6718 - val_loss: 15.8361 - val_acc: 0.6777\n",
      "Epoch 3363/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6656 - val_loss: 15.8361 - val_acc: 0.6777\n",
      "Epoch 3364/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5941 - acc: 0.6677 - val_loss: 15.8360 - val_acc: 0.6777\n",
      "Epoch 3365/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6698 - val_loss: 15.8359 - val_acc: 0.6777\n",
      "Epoch 3366/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6708 - val_loss: 15.8359 - val_acc: 0.6777\n",
      "Epoch 3367/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6749 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3368/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6615 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3369/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6718 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3370/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6708 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3371/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5962 - acc: 0.6718 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3372/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5567 - acc: 0.6760 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3373/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6615 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3374/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6791 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3375/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5971 - acc: 0.6718 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3376/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6729 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3377/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5575 - acc: 0.6718 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3378/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6646 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3379/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6563 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3380/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6605 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3381/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6708 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3382/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6646 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3383/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5439 - acc: 0.6708 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3384/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5875 - acc: 0.6594 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3385/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6625 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3386/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5458 - acc: 0.6718 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3387/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5629 - acc: 0.6615 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3388/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6667 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3389/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6646 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3390/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6667 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3391/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6646 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3392/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6605 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3393/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6708 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3394/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6698 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3395/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6698 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3396/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5591 - acc: 0.6718 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3397/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5893 - acc: 0.6594 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3398/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6718 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3399/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6698 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3400/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5566 - acc: 0.6698 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3401/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6708 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3402/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6698 - val_loss: 15.8358 - val_acc: 0.6777\n",
      "Epoch 3403/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6646 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3404/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6708 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3405/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6636 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3406/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6749 - val_loss: 15.8357 - val_acc: 0.6777\n",
      "Epoch 3407/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6708 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3408/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6781 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3409/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6687 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3410/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5905 - acc: 0.6718 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3411/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6615 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3412/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6698 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3413/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6698 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3414/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6708 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3415/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5603 - acc: 0.6698 - val_loss: 15.8356 - val_acc: 0.6777\n",
      "Epoch 3416/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6636 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3417/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5911 - acc: 0.6791 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3418/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6687 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3419/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5968 - acc: 0.6636 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3420/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5633 - acc: 0.6739 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3421/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6698 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3422/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5974 - acc: 0.6698 - val_loss: 15.8355 - val_acc: 0.6777\n",
      "Epoch 3423/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6625 - val_loss: 15.8354 - val_acc: 0.6777\n",
      "Epoch 3424/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6594 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3425/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6667 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3426/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6781 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3427/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6615 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3428/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6615 - val_loss: 15.8354 - val_acc: 0.6777\n",
      "Epoch 3429/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6615 - val_loss: 15.8354 - val_acc: 0.6777\n",
      "Epoch 3430/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6708 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3431/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6677 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3432/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6687 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3433/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6656 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3434/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6687 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3435/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6677 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3436/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5940 - acc: 0.6615 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3437/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6656 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3438/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6625 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3439/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6615 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3440/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6594 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3441/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6739 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3442/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6718 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3443/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6646 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3444/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6687 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3445/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6718 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3446/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6605 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3447/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6667 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3448/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6708 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3449/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6760 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3450/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6687 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3451/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6605 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3452/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6708 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3453/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6698 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3454/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6656 - val_loss: 15.8353 - val_acc: 0.6777\n",
      "Epoch 3455/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5611 - acc: 0.6708 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3456/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6667 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3457/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6698 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3458/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6646 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3459/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6667 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3460/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6687 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3461/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6708 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3462/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5698 - acc: 0.6708 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3463/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6739 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3464/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6573 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3465/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6677 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3466/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6687 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3467/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5593 - acc: 0.6698 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3468/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6677 - val_loss: 15.8352 - val_acc: 0.6777\n",
      "Epoch 3469/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6594 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3470/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5628 - acc: 0.6698 - val_loss: 15.8351 - val_acc: 0.6777\n",
      "Epoch 3471/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6584 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3472/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6594 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3473/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6770 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3474/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6729 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3475/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6729 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3476/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6542 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3477/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5923 - acc: 0.6739 - val_loss: 15.8350 - val_acc: 0.6777\n",
      "Epoch 3478/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6605 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3479/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6542 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3480/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6698 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3481/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6656 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3482/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6625 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3483/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6687 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3484/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5851 - acc: 0.6718 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3485/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5538 - acc: 0.6760 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3486/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6749 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3487/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6667 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3488/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5636 - acc: 0.6573 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3489/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5884 - acc: 0.6646 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3490/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6718 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3491/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6563 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3492/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6594 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3493/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6698 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3494/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6708 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3495/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6708 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3496/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6625 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3497/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5495 - acc: 0.6636 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3498/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6636 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3499/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6708 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3500/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6553 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3501/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6708 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3502/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6584 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3503/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6718 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3504/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5582 - acc: 0.6781 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3505/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6636 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3506/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6542 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3507/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6687 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3508/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6698 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3509/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6656 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3510/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6667 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3511/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6667 - val_loss: 15.8349 - val_acc: 0.6777\n",
      "Epoch 3512/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6677 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3513/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6739 - val_loss: 15.8348 - val_acc: 0.6777\n",
      "Epoch 3514/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6687 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3515/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6770 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3516/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5579 - acc: 0.6656 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3517/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6687 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3518/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6594 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3519/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6563 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3520/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5720 - acc: 0.6708 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3521/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6718 - val_loss: 15.8347 - val_acc: 0.6777\n",
      "Epoch 3522/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5876 - acc: 0.6718 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3523/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6646 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3524/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5966 - acc: 0.6781 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3525/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6656 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3526/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6646 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3527/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5642 - acc: 0.6615 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3528/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6605 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3529/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6605 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3530/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6646 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3531/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6656 - val_loss: 15.8346 - val_acc: 0.6777\n",
      "Epoch 3532/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6615 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3533/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6708 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3534/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6656 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3535/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5831 - acc: 0.6667 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3536/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6698 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3537/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6708 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3538/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6708 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3539/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5865 - acc: 0.6573 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3540/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3541/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6729 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3542/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6739 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3543/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6646 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3544/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6698 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3545/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6677 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3546/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6729 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3547/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6667 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3548/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5582 - acc: 0.6656 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3549/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6677 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3550/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5698 - acc: 0.6584 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3551/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5839 - acc: 0.6739 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3552/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6687 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3553/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6605 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3554/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3555/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6636 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3556/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5699 - acc: 0.6718 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3557/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6677 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3558/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3559/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6739 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3560/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6708 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3561/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5846 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3562/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6636 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3563/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6729 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3564/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5888 - acc: 0.6656 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3565/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6698 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3566/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6677 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3567/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6729 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3568/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6687 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3569/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6760 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3570/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5572 - acc: 0.6749 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3571/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6749 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3572/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6573 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3573/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6677 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3574/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5578 - acc: 0.6687 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3575/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6594 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3576/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6667 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3577/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6677 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3578/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5513 - acc: 0.6594 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3579/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5774 - acc: 0.6636 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3580/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6656 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3581/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6667 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3582/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6625 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3583/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6646 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3584/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5828 - acc: 0.6770 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3585/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5554 - acc: 0.6656 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3586/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6656 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3587/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5609 - acc: 0.6677 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3588/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6605 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3589/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3590/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6708 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3591/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3592/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3593/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6739 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3594/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5648 - acc: 0.6729 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3595/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3596/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5950 - acc: 0.6698 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3597/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6708 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3598/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6646 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3599/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6625 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3600/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6729 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3601/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6749 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3602/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6636 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3603/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6677 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3604/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6677 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3605/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5635 - acc: 0.6698 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3606/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6656 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3607/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5648 - acc: 0.6760 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3608/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6791 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3609/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6646 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3610/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6708 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3611/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6594 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3612/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6656 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3613/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3614/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6760 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3615/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6749 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3616/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3617/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6698 - val_loss: 15.8345 - val_acc: 0.6777\n",
      "Epoch 3618/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6718 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3619/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6739 - val_loss: 15.8344 - val_acc: 0.6777\n",
      "Epoch 3620/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6708 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3621/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6656 - val_loss: 15.8343 - val_acc: 0.6777\n",
      "Epoch 3622/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6687 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3623/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6698 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3624/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5903 - acc: 0.6687 - val_loss: 15.8341 - val_acc: 0.6777\n",
      "Epoch 3625/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6605 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3626/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5559 - acc: 0.6656 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3627/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6656 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3628/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6636 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3629/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6625 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3630/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6718 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3631/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5948 - acc: 0.6646 - val_loss: 15.8342 - val_acc: 0.6777\n",
      "Epoch 3632/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6698 - val_loss: 15.8341 - val_acc: 0.6777\n",
      "Epoch 3633/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5649 - acc: 0.6677 - val_loss: 15.8341 - val_acc: 0.6777\n",
      "Epoch 3634/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5699 - acc: 0.6781 - val_loss: 15.8341 - val_acc: 0.6777\n",
      "Epoch 3635/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6729 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3636/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6677 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3637/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6781 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3638/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6584 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3639/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6646 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3640/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6677 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3641/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6636 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3642/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6656 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3643/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5960 - acc: 0.6615 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3644/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6677 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3645/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6739 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3646/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6749 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3647/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6656 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3648/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6698 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3649/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6729 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3650/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6718 - val_loss: 15.8340 - val_acc: 0.6777\n",
      "Epoch 3651/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6791 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3652/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5600 - acc: 0.6698 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3653/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6708 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3654/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6636 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3655/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5453 - acc: 0.6698 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3656/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6698 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3657/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6708 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3658/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5456 - acc: 0.6584 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3659/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6636 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3660/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6687 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3661/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5835 - acc: 0.6656 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3662/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6749 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3663/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5648 - acc: 0.6625 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3664/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6718 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3665/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6605 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3666/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5699 - acc: 0.6605 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3667/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5802 - acc: 0.6646 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3668/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6646 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3669/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6687 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3670/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6687 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3671/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5474 - acc: 0.6698 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3672/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6584 - val_loss: 15.8339 - val_acc: 0.6777\n",
      "Epoch 3673/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6656 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3674/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6625 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3675/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6667 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3676/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6718 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3677/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6739 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3678/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6760 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3679/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6646 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3680/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5517 - acc: 0.6677 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3681/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6739 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3682/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6636 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3683/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6687 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3684/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6698 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3685/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5670 - acc: 0.6708 - val_loss: 15.8338 - val_acc: 0.6777\n",
      "Epoch 3686/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6636 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3687/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6698 - val_loss: 15.8337 - val_acc: 0.6777\n",
      "Epoch 3688/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6708 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3689/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6667 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3690/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6667 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3691/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6687 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3692/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6677 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3693/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6667 - val_loss: 15.8336 - val_acc: 0.6777\n",
      "Epoch 3694/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6687 - val_loss: 15.8335 - val_acc: 0.6777\n",
      "Epoch 3695/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6698 - val_loss: 15.8335 - val_acc: 0.6777\n",
      "Epoch 3696/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5921 - acc: 0.6718 - val_loss: 15.8335 - val_acc: 0.6777\n",
      "Epoch 3697/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6698 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3698/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6708 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3699/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6636 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3700/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6729 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3701/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6698 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3702/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6615 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3703/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6718 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3704/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6646 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3705/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5777 - acc: 0.6656 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3706/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6646 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3707/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6667 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3708/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6708 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3709/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6698 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3710/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6812 - val_loss: 15.8334 - val_acc: 0.6777\n",
      "Epoch 3711/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6646 - val_loss: 15.8333 - val_acc: 0.6777\n",
      "Epoch 3712/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6718 - val_loss: 15.8333 - val_acc: 0.6777\n",
      "Epoch 3713/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6656 - val_loss: 15.8333 - val_acc: 0.6777\n",
      "Epoch 3714/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6801 - val_loss: 15.8333 - val_acc: 0.6777\n",
      "Epoch 3715/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6646 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3716/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6625 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3717/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6708 - val_loss: 15.8333 - val_acc: 0.6777\n",
      "Epoch 3718/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6656 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3719/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6656 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3720/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5495 - acc: 0.6667 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3721/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3722/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6718 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3723/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6718 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3724/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6677 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3725/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5487 - acc: 0.6708 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3726/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5596 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3727/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5912 - acc: 0.6656 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3728/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6718 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3729/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6605 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3730/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5813 - acc: 0.6656 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3731/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6646 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3732/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3733/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3734/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6770 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3735/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6822 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3736/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6656 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3737/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6677 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3738/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5856 - acc: 0.6636 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3739/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6708 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3740/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5987 - acc: 0.6687 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3741/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5611 - acc: 0.6749 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3742/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5568 - acc: 0.6677 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3743/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6708 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3744/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6605 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3745/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6594 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3746/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6625 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3747/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5834 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3748/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6718 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3749/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6677 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3750/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6584 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3751/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6708 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3752/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5798 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3753/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6573 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3754/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6781 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3755/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5486 - acc: 0.6708 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3756/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6656 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3757/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6636 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3758/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5554 - acc: 0.6812 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3759/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6667 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3760/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6687 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3761/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5572 - acc: 0.6739 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3762/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5932 - acc: 0.6708 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3763/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6739 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3764/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6646 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3765/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5767 - acc: 0.6718 - val_loss: 15.8332 - val_acc: 0.6777\n",
      "Epoch 3766/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5866 - acc: 0.6677 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3767/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6656 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3768/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6687 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3769/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5864 - acc: 0.6646 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3770/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6625 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3771/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6646 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3772/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6646 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3773/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6677 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3774/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5913 - acc: 0.6646 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3775/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6584 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3776/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6698 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3777/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6718 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3778/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5801 - acc: 0.6687 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3779/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5495 - acc: 0.6687 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3780/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5919 - acc: 0.6667 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3781/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6729 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3782/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5954 - acc: 0.6636 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3783/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6532 - val_loss: 15.8331 - val_acc: 0.6777\n",
      "Epoch 3784/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6708 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3785/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6708 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3786/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5709 - acc: 0.6677 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3787/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6760 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3788/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6677 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3789/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6718 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3790/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6760 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3791/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5486 - acc: 0.6729 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3792/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6739 - val_loss: 15.8330 - val_acc: 0.6777\n",
      "Epoch 3793/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5818 - acc: 0.6718 - val_loss: 15.8329 - val_acc: 0.6777\n",
      "Epoch 3794/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6646 - val_loss: 15.8329 - val_acc: 0.6777\n",
      "Epoch 3795/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6718 - val_loss: 15.8329 - val_acc: 0.6777\n",
      "Epoch 3796/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6729 - val_loss: 15.8329 - val_acc: 0.6777\n",
      "Epoch 3797/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6698 - val_loss: 15.8329 - val_acc: 0.6777\n",
      "Epoch 3798/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6677 - val_loss: 15.8328 - val_acc: 0.6777\n",
      "Epoch 3799/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6698 - val_loss: 15.8328 - val_acc: 0.6777\n",
      "Epoch 3800/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6708 - val_loss: 15.8328 - val_acc: 0.6777\n",
      "Epoch 3801/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5872 - acc: 0.6708 - val_loss: 15.8328 - val_acc: 0.6777\n",
      "Epoch 3802/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6760 - val_loss: 15.8327 - val_acc: 0.6777\n",
      "Epoch 3803/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6687 - val_loss: 15.8327 - val_acc: 0.6777\n",
      "Epoch 3804/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5894 - acc: 0.6687 - val_loss: 15.8327 - val_acc: 0.6777\n",
      "Epoch 3805/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6584 - val_loss: 15.8327 - val_acc: 0.6777\n",
      "Epoch 3806/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6698 - val_loss: 15.8327 - val_acc: 0.6777\n",
      "Epoch 3807/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6708 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3808/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6729 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3809/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6739 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3810/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5674 - acc: 0.6646 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3811/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6677 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3812/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6636 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3813/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6749 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3814/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5668 - acc: 0.6667 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3815/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6687 - val_loss: 15.8326 - val_acc: 0.6777\n",
      "Epoch 3816/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6729 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3817/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5476 - acc: 0.6615 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3818/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5900 - acc: 0.6698 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3819/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6749 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3820/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5870 - acc: 0.6729 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3821/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6718 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3822/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6615 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3823/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6739 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3824/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6667 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3825/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6739 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3826/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6739 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3827/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6760 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3828/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6677 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3829/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5536 - acc: 0.6677 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3830/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6760 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3831/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6625 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3832/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5878 - acc: 0.6760 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3833/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6615 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3834/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5469 - acc: 0.6656 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3835/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6739 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3836/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6718 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3837/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5862 - acc: 0.6656 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3838/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6615 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3839/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5573 - acc: 0.6687 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3840/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6832 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3841/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5924 - acc: 0.6667 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3842/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6656 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3843/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6636 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3844/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6636 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3845/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6656 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3846/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5790 - acc: 0.6646 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3847/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5821 - acc: 0.6708 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3848/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6646 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3849/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6656 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3850/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6729 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3851/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5552 - acc: 0.6698 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3852/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6646 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3853/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6687 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3854/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6636 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3855/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6646 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3856/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6667 - val_loss: 15.8325 - val_acc: 0.6777\n",
      "Epoch 3857/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6656 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3858/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5784 - acc: 0.6677 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3859/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6656 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3860/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6739 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3861/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6625 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3862/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5902 - acc: 0.6698 - val_loss: 15.8324 - val_acc: 0.6777\n",
      "Epoch 3863/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6698 - val_loss: 15.8323 - val_acc: 0.6777\n",
      "Epoch 3864/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5506 - acc: 0.6781 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3865/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6656 - val_loss: 15.8323 - val_acc: 0.6777\n",
      "Epoch 3866/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5848 - acc: 0.6698 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3867/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6625 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3868/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6553 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3869/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6636 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3870/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5574 - acc: 0.6698 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3871/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6801 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3872/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5481 - acc: 0.6729 - val_loss: 15.8323 - val_acc: 0.6777\n",
      "Epoch 3873/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6625 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3874/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6656 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3875/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6698 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3876/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6677 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3877/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6791 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3878/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5431 - acc: 0.6760 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3879/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6615 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3880/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5648 - acc: 0.6698 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3881/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6739 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3882/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5540 - acc: 0.6667 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3883/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5486 - acc: 0.6594 - val_loss: 15.8322 - val_acc: 0.6777\n",
      "Epoch 3884/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5948 - acc: 0.6770 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3885/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6636 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3886/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6687 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3887/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6729 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3888/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6625 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3889/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6760 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3890/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6729 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3891/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6594 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3892/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6729 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3893/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5562 - acc: 0.6687 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3894/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6615 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3895/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6760 - val_loss: 15.8321 - val_acc: 0.6777\n",
      "Epoch 3896/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5642 - acc: 0.6718 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3897/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6739 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3898/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6677 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3899/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5867 - acc: 0.6687 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3900/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5891 - acc: 0.6760 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3901/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6677 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3902/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5822 - acc: 0.6687 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3903/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5591 - acc: 0.6646 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3904/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5579 - acc: 0.6698 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3905/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6605 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3906/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6667 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3907/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6677 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3908/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6729 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3909/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6739 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3910/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6718 - val_loss: 15.8320 - val_acc: 0.6777\n",
      "Epoch 3911/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6687 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3912/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5502 - acc: 0.6729 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3913/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6667 - val_loss: 15.8319 - val_acc: 0.6777\n",
      "Epoch 3914/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6636 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3915/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6656 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3916/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5608 - acc: 0.6739 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3917/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6687 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3918/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6667 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3919/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6718 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3920/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6667 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3921/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6729 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3922/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6667 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3923/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5409 - acc: 0.6687 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3924/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6667 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3925/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5521 - acc: 0.6739 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3926/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6708 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3927/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6636 - val_loss: 15.8318 - val_acc: 0.6777\n",
      "Epoch 3928/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6636 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3929/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6739 - val_loss: 15.8317 - val_acc: 0.6777\n",
      "Epoch 3930/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5421 - acc: 0.6770 - val_loss: 15.8316 - val_acc: 0.6777\n",
      "Epoch 3931/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6687 - val_loss: 15.8316 - val_acc: 0.6777\n",
      "Epoch 3932/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6677 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3933/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6770 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3934/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6729 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3935/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5453 - acc: 0.6729 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3936/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5624 - acc: 0.6729 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3937/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5528 - acc: 0.6667 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3938/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6749 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3939/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6677 - val_loss: 15.8315 - val_acc: 0.6777\n",
      "Epoch 3940/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6760 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3941/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5852 - acc: 0.6636 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3942/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6594 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3943/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6801 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3944/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5572 - acc: 0.6708 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3945/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6718 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3946/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6822 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3947/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6749 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3948/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5674 - acc: 0.6760 - val_loss: 15.8314 - val_acc: 0.6777\n",
      "Epoch 3949/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6708 - val_loss: 15.8313 - val_acc: 0.6777\n",
      "Epoch 3950/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6739 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3951/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6677 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3952/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6636 - val_loss: 15.8313 - val_acc: 0.6777\n",
      "Epoch 3953/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6760 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3954/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6656 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3955/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5863 - acc: 0.6781 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3956/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6687 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3957/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6636 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3958/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6791 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3959/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6698 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3960/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6718 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3961/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6636 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3962/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5375 - acc: 0.6760 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3963/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6656 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3964/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6636 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3965/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6677 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3966/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6677 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3967/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5843 - acc: 0.6615 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3968/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6636 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3969/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6646 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3970/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6791 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3971/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6687 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3972/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6677 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3973/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6687 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3974/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6708 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3975/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6770 - val_loss: 15.8312 - val_acc: 0.6777\n",
      "Epoch 3976/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6718 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3977/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6605 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3978/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5762 - acc: 0.6656 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3979/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6677 - val_loss: 15.8311 - val_acc: 0.6777\n",
      "Epoch 3980/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6718 - val_loss: 15.8310 - val_acc: 0.6777\n",
      "Epoch 3981/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5537 - acc: 0.6708 - val_loss: 15.8310 - val_acc: 0.6777\n",
      "Epoch 3982/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5594 - acc: 0.6677 - val_loss: 15.8310 - val_acc: 0.6777\n",
      "Epoch 3983/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5505 - acc: 0.6812 - val_loss: 15.8310 - val_acc: 0.6777\n",
      "Epoch 3984/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5699 - acc: 0.6718 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3985/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6677 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3986/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6708 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3987/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5598 - acc: 0.6667 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3988/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6677 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3989/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6749 - val_loss: 15.8309 - val_acc: 0.6777\n",
      "Epoch 3990/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6791 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3991/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6781 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3992/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6646 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3993/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5538 - acc: 0.6791 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3994/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6760 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3995/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6615 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3996/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5767 - acc: 0.6749 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3997/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5674 - acc: 0.6677 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3998/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6646 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 3999/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6739 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4000/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6636 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4001/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6760 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4002/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6677 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4003/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6667 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4004/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5575 - acc: 0.6718 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4005/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6698 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4006/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6646 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4007/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5942 - acc: 0.6708 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4008/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6760 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4009/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6729 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4010/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6749 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4011/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6667 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4012/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5793 - acc: 0.6677 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4013/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5890 - acc: 0.6646 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4014/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6625 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4015/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6667 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4016/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6677 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4017/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6749 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4018/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5532 - acc: 0.6687 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4019/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6594 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4020/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6636 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4021/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6708 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4022/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5579 - acc: 0.6698 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4023/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5530 - acc: 0.6729 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4024/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5483 - acc: 0.6677 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4025/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6656 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4026/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6739 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4027/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6791 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4028/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6646 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4029/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6615 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4030/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6677 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4031/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6729 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4032/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6749 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4033/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6656 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4034/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6656 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4035/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6667 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4036/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5642 - acc: 0.6646 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4037/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6656 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4038/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6625 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4039/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6708 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4040/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6698 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4041/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5539 - acc: 0.6656 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4042/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5519 - acc: 0.6781 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4043/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6749 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4044/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6760 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4045/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6760 - val_loss: 15.8308 - val_acc: 0.6777\n",
      "Epoch 4046/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6729 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4047/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5868 - acc: 0.6656 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4048/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6656 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4049/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6625 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4050/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5650 - acc: 0.6656 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4051/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5517 - acc: 0.6781 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4052/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6687 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4053/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6698 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4054/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6687 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4055/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6636 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4056/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6008 - acc: 0.6760 - val_loss: 15.8307 - val_acc: 0.6777\n",
      "Epoch 4057/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6739 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4058/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6677 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4059/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6708 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4060/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6656 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4061/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5540 - acc: 0.6667 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4062/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6656 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4063/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6781 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4064/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5809 - acc: 0.6667 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4065/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6801 - val_loss: 15.8306 - val_acc: 0.6777\n",
      "Epoch 4066/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6646 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4067/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6677 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4068/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6625 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4069/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5490 - acc: 0.6687 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4070/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5589 - acc: 0.6646 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4071/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6667 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4072/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6791 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4073/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6698 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4074/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6615 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4075/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6749 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4076/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5882 - acc: 0.6718 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4077/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6625 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4078/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6615 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4079/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6698 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4080/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5830 - acc: 0.6677 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4081/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5883 - acc: 0.6625 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4082/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6770 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4083/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6739 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4084/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5578 - acc: 0.6718 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4085/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6729 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4086/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6615 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4087/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6677 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4088/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6677 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4089/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6698 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4090/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6698 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4091/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5716 - acc: 0.6760 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4092/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6739 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4093/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6687 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4094/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5631 - acc: 0.6739 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4095/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6636 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4096/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6656 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4097/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6667 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4098/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6605 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4099/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6708 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4100/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6584 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4101/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6708 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4102/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6698 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4103/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6677 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4104/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6667 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4105/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6718 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4106/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6729 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4107/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6553 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4108/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6625 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4109/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6708 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4110/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6667 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4111/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6729 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4112/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6667 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4113/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6729 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4114/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5537 - acc: 0.6708 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4115/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6646 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4116/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6656 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4117/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6729 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4118/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6563 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4119/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6770 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4120/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6625 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4121/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5849 - acc: 0.6718 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4122/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6573 - val_loss: 15.8305 - val_acc: 0.6777\n",
      "Epoch 4123/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6615 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4124/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6708 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4125/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5503 - acc: 0.6687 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4126/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6605 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4127/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6708 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4128/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5624 - acc: 0.6636 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4129/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6584 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4130/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6687 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4131/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6698 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4132/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6656 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4133/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6636 - val_loss: 15.8304 - val_acc: 0.6777\n",
      "Epoch 4134/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5889 - acc: 0.6698 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4135/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6605 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4136/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5877 - acc: 0.6677 - val_loss: 15.8303 - val_acc: 0.6777.\n",
      "Epoch 4137/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6667 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4138/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5588 - acc: 0.6708 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4139/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5760 - acc: 0.6698 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4140/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6718 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4141/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6729 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4142/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5500 - acc: 0.6646 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4143/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6739 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4144/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6677 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4145/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5747 - acc: 0.6739 - val_loss: 15.8303 - val_acc: 0.6777\n",
      "Epoch 4146/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6698 - val_loss: 15.8302 - val_acc: 0.6777\n",
      "Epoch 4147/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6687 - val_loss: 15.8302 - val_acc: 0.6777\n",
      "Epoch 4148/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6687 - val_loss: 15.8302 - val_acc: 0.6777\n",
      "Epoch 4149/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6770 - val_loss: 15.8302 - val_acc: 0.6777\n",
      "Epoch 4150/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5916 - acc: 0.6656 - val_loss: 15.8301 - val_acc: 0.6777\n",
      "Epoch 4151/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5608 - acc: 0.6563 - val_loss: 15.8301 - val_acc: 0.6777\n",
      "Epoch 4152/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6646 - val_loss: 15.8301 - val_acc: 0.6777\n",
      "Epoch 4153/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6687 - val_loss: 15.8301 - val_acc: 0.6777\n",
      "Epoch 4154/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6718 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4155/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6698 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4156/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6667 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4157/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6553 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4158/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5839 - acc: 0.6615 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4159/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5670 - acc: 0.6791 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4160/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5907 - acc: 0.6605 - val_loss: 15.8300 - val_acc: 0.6777\n",
      "Epoch 4161/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6646 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4162/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6708 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4163/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6770 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4164/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6677 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4165/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5496 - acc: 0.6718 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4166/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6667 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4167/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6729 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4168/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6770 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4169/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5815 - acc: 0.6615 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4170/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6739 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4171/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6760 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4172/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6739 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4173/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6749 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4174/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6687 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4175/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5629 - acc: 0.6605 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4176/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5626 - acc: 0.6677 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4177/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6625 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4178/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6708 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4179/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6708 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4180/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6781 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4181/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6636 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4182/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5631 - acc: 0.6625 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4183/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5892 - acc: 0.6770 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4184/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6677 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4185/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5813 - acc: 0.6636 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4186/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6698 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4187/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5777 - acc: 0.6708 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4188/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6656 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4189/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6749 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4190/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6636 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4191/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5531 - acc: 0.6708 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4192/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6698 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4193/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6687 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4194/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6656 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4195/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6687 - val_loss: 15.8299 - val_acc: 0.6777\n",
      "Epoch 4196/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5501 - acc: 0.6791 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4197/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6770 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4198/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6739 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4199/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6749 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4200/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5608 - acc: 0.6667 - val_loss: 15.8298 - val_acc: 0.6777\n",
      "Epoch 4201/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5576 - acc: 0.6718 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4202/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5844 - acc: 0.6708 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4203/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5501 - acc: 0.6749 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4204/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6687 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4205/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6605 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4206/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5859 - acc: 0.6729 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4207/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6615 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4208/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6718 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4209/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6739 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4210/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6708 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4211/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6667 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4212/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6729 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4213/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5523 - acc: 0.6770 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4214/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5575 - acc: 0.6687 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4215/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5482 - acc: 0.6687 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4216/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6708 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4217/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6718 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4218/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6656 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4219/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6698 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4220/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6770 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4221/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6656 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4222/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6718 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4223/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6667 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4224/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6636 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4225/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5509 - acc: 0.6656 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4226/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6729 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4227/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6542 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4228/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5575 - acc: 0.6708 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4229/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6687 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4230/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6615 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4231/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5829 - acc: 0.6636 - val_loss: 15.8297 - val_acc: 0.6777\n",
      "Epoch 4232/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6729 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4233/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6749 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4234/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5548 - acc: 0.6573 - val_loss: 15.8296 - val_acc: 0.6777\n",
      "Epoch 4235/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6646 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4236/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6656 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4237/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6739 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4238/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6791 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4239/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6698 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4240/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6667 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4241/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6646 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4242/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5511 - acc: 0.6718 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4243/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5478 - acc: 0.6656 - val_loss: 15.8295 - val_acc: 0.6777\n",
      "Epoch 4244/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6760 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4245/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5579 - acc: 0.6770 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4246/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6718 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4247/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5602 - acc: 0.6636 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4248/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5975 - acc: 0.6625 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4249/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5824 - acc: 0.6636 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4250/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5501 - acc: 0.6760 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4251/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6646 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4252/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6636 - val_loss: 15.8294 - val_acc: 0.6777\n",
      "Epoch 4253/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6698 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4254/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6801 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4255/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6656 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4256/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5799 - acc: 0.6760 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4257/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5716 - acc: 0.6646 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4258/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6698 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4259/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5555 - acc: 0.6687 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4260/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6760 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4261/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6667 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4262/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6677 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4263/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5723 - acc: 0.6656 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4264/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6625 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4265/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6770 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4266/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6812 - val_loss: 15.8293 - val_acc: 0.6777\n",
      "Epoch 4267/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6698 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4268/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5895 - acc: 0.6687 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4269/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6791 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4270/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6677 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4271/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6656 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4272/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6563 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4273/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6729 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4274/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6718 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4275/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5542 - acc: 0.6677 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4276/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5558 - acc: 0.6646 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4277/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6687 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4278/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6718 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4279/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6615 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4280/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6656 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4281/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6718 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4282/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6667 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4283/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6729 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4284/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5517 - acc: 0.6615 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4285/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5498 - acc: 0.6656 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4286/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6687 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4287/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6718 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4288/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6625 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4289/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5676 - acc: 0.6687 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4290/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5782 - acc: 0.6594 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4291/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6636 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4292/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6708 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4293/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5628 - acc: 0.6791 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4294/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6615 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4295/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6708 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4296/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6687 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4297/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5581 - acc: 0.6687 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4298/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6553 - val_loss: 15.8292 - val_acc: 0.6777\n",
      "Epoch 4299/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6636 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4300/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6656 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4301/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6615 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4302/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6667 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4303/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6677 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4304/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6708 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4305/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5833 - acc: 0.6718 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4306/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6822 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4307/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6646 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4308/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5542 - acc: 0.6729 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4309/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6687 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4310/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6770 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4311/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6646 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4312/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6677 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4313/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5558 - acc: 0.6646 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4314/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6718 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4315/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.5633 - acc: 0.67 - 0s - loss: 16.5742 - acc: 0.6770 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4316/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6584 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4317/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6729 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4318/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5551 - acc: 0.6667 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4319/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5557 - acc: 0.6718 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4320/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6853 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4321/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5593 - acc: 0.6605 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4322/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6677 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4323/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6729 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4324/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6708 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4325/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5670 - acc: 0.6636 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4326/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6708 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4327/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6625 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4328/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5562 - acc: 0.6687 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4329/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6770 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4330/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5869 - acc: 0.6677 - val_loss: 15.8291 - val_acc: 0.6777\n",
      "Epoch 4331/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5847 - acc: 0.6667 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4332/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6791 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4333/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5549 - acc: 0.6770 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4334/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5458 - acc: 0.6698 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4335/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5757 - acc: 0.6739 - val_loss: 15.8290 - val_acc: 0.6777\n",
      "Epoch 4336/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6739 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4337/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6729 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4338/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6511 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4339/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6625 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4340/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6791 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4341/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6770 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4342/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6698 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4343/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5769 - acc: 0.6656 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4344/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6749 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4345/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5726 - acc: 0.6698 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4346/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6656 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4347/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6615 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4348/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5871 - acc: 0.6687 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4349/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6729 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4350/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5929 - acc: 0.6749 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4351/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6739 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4352/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5561 - acc: 0.6563 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4353/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6698 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4354/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5649 - acc: 0.6636 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4355/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6718 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4356/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5668 - acc: 0.6739 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4357/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6760 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4358/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5522 - acc: 0.6625 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4359/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5842 - acc: 0.6646 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4360/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5763 - acc: 0.6708 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4361/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6677 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4362/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5600 - acc: 0.6708 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4363/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6667 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4364/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5500 - acc: 0.6781 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4365/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5630 - acc: 0.6708 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4366/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6625 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4367/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6739 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4368/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6770 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4369/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5791 - acc: 0.6749 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4370/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5584 - acc: 0.6625 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4371/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6687 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4372/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5741 - acc: 0.6625 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4373/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6636 - val_loss: 15.8289 - val_acc: 0.6777\n",
      "Epoch 4374/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6822 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4375/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6708 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4376/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6718 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4377/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5633 - acc: 0.6605 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4378/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5585 - acc: 0.6687 - val_loss: 15.8288 - val_acc: 0.6777\n",
      "Epoch 4379/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5617 - acc: 0.6625 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4380/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6698 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4381/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5746 - acc: 0.6573 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4382/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6615 - val_loss: 15.8286 - val_acc: 0.6777\n",
      "Epoch 4383/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6646 - val_loss: 15.8287 - val_acc: 0.6777\n",
      "Epoch 4384/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6749 - val_loss: 15.8286 - val_acc: 0.6777\n",
      "Epoch 4385/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5862 - acc: 0.6698 - val_loss: 15.8286 - val_acc: 0.6777\n",
      "Epoch 4386/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5591 - acc: 0.6718 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4387/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5650 - acc: 0.6718 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4388/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6667 - val_loss: 15.8286 - val_acc: 0.6777\n",
      "Epoch 4389/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6605 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4390/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6718 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4391/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5486 - acc: 0.6708 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4392/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5910 - acc: 0.6791 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4393/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6667 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4394/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6656 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4395/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5525 - acc: 0.6687 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4396/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6605 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4397/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5991 - acc: 0.6739 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4398/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6718 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4399/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6625 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4400/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5589 - acc: 0.6677 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4401/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6739 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4402/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5814 - acc: 0.6677 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4403/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6656 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4404/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6687 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4405/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4406/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5541 - acc: 0.6760 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4407/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6739 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4408/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6708 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4409/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5586 - acc: 0.6739 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4410/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5624 - acc: 0.6781 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4411/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6687 - val_loss: 15.8285 - val_acc: 0.6777\n",
      "Epoch 4412/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5507 - acc: 0.6677 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4413/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5690 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4414/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6801 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4415/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5558 - acc: 0.6656 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4416/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6656 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4417/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6656 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4418/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6749 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4419/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5855 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4420/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5743 - acc: 0.6687 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4421/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5613 - acc: 0.6708 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4422/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6739 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4423/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5914 - acc: 0.6636 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4424/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6698 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4425/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5576 - acc: 0.6667 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4426/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6739 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4427/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6718 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4428/5000\n",
      "966/966 [==============================] - 0s - loss: 16.6035 - acc: 0.6698 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4429/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5422 - acc: 0.6636 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4430/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6781 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4431/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5590 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4432/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6708 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4433/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6646 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4434/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4435/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6677 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4436/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6718 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4437/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5486 - acc: 0.6698 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4438/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5770 - acc: 0.6749 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4439/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6739 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4440/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6739 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4441/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6698 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4442/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6801 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4443/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6708 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4444/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6636 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4445/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5577 - acc: 0.6667 - val_loss: 15.8284 - val_acc: 0.6777\n",
      "Epoch 4446/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5917 - acc: 0.6770 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4447/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5492 - acc: 0.6718 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4448/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5480 - acc: 0.6822 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4449/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5759 - acc: 0.6667 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4450/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6698 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4451/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6718 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4452/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5547 - acc: 0.6770 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4453/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6687 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4454/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6687 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4455/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6770 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4456/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6646 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4457/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6781 - val_loss: 15.8283 - val_acc: 0.6777\n",
      "Epoch 4458/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6646 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4459/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5861 - acc: 0.6615 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4460/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6625 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4461/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6656 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4462/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5510 - acc: 0.6812 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4463/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5584 - acc: 0.6636 - val_loss: 15.8282 - val_acc: 0.6777\n",
      "Epoch 4464/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6718 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4465/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5631 - acc: 0.6698 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4466/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5594 - acc: 0.6708 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4467/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6739 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4468/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6687 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4469/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6677 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4470/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6698 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4471/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5713 - acc: 0.6584 - val_loss: 15.8281 - val_acc: 0.6777\n",
      "Epoch 4472/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6749 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4473/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6749 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4474/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6687 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4475/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5562 - acc: 0.6646 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4476/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6749 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4477/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6677 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4478/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6656 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4479/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6687 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4480/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5622 - acc: 0.6605 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4481/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6708 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4482/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6667 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4483/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5734 - acc: 0.6749 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4484/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6822 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4485/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6667 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4486/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6667 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4487/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5716 - acc: 0.6718 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4488/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6677 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4489/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6832 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4490/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5535 - acc: 0.6708 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4491/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6781 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4492/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6677 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4493/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6677 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4494/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5854 - acc: 0.6646 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4495/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5538 - acc: 0.6563 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4496/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5545 - acc: 0.6791 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4497/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6791 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4498/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5466 - acc: 0.6739 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4499/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6729 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4500/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6636 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4501/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5582 - acc: 0.6770 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4502/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5643 - acc: 0.6718 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4503/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5552 - acc: 0.6760 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4504/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6718 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4505/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6656 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4506/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6687 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4507/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6646 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4508/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5697 - acc: 0.6791 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4509/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6667 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4510/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5810 - acc: 0.6760 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4511/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6729 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4512/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6687 - val_loss: 15.8280 - val_acc: 0.6777\n",
      "Epoch 4513/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6698 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4514/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6718 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4515/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5452 - acc: 0.6687 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4516/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6677 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4517/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5508 - acc: 0.6687 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4518/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5776 - acc: 0.6615 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4519/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6718 - val_loss: 15.8279 - val_acc: 0.6777\n",
      "Epoch 4520/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5542 - acc: 0.6677 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4521/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5813 - acc: 0.6791 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4522/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6760 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4523/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6708 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4524/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6584 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4525/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5526 - acc: 0.6729 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4526/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6739 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4527/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5806 - acc: 0.6718 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4528/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5573 - acc: 0.6760 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4529/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6677 - val_loss: 15.8278 - val_acc: 0.6777\n",
      "Epoch 4530/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6812 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4531/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5796 - acc: 0.6615 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4532/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6739 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4533/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6781 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4534/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6749 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4535/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5636 - acc: 0.6698 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4536/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6625 - val_loss: 15.8277 - val_acc: 0.6777\n",
      "Epoch 4537/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6708 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4538/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6801 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4539/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5679 - acc: 0.6770 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4540/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6625 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4541/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5700 - acc: 0.6749 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4542/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6781 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4543/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5708 - acc: 0.6698 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4544/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6801 - val_loss: 15.8276 - val_acc: 0.6777\n",
      "Epoch 4545/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5582 - acc: 0.6749 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4546/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5821 - acc: 0.6708 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4547/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6708 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4548/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5511 - acc: 0.6791 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4549/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5812 - acc: 0.6636 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4550/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5692 - acc: 0.6739 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4551/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5953 - acc: 0.6801 - val_loss: 15.8274 - val_acc: 0.6777\n",
      "Epoch 4552/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5434 - acc: 0.6708 - val_loss: 15.8274 - val_acc: 0.6777\n",
      "Epoch 4553/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6687 - val_loss: 15.8275 - val_acc: 0.6777\n",
      "Epoch 4554/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5885 - acc: 0.6646 - val_loss: 15.8274 - val_acc: 0.6777\n",
      "Epoch 4555/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6760 - val_loss: 15.8274 - val_acc: 0.6777\n",
      "Epoch 4556/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6708 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4557/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6636 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4558/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6625 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4559/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6739 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4560/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6667 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4561/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6770 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4562/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5845 - acc: 0.6667 - val_loss: 15.8274 - val_acc: 0.6777\n",
      "Epoch 4563/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6698 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4564/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5595 - acc: 0.6760 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4565/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6781 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4566/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6760 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4567/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6667 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4568/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5738 - acc: 0.6667 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4569/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6605 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4570/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6584 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4571/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5764 - acc: 0.6739 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4572/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5580 - acc: 0.6718 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4573/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6718 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4574/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5705 - acc: 0.6863 - val_loss: 15.8273 - val_acc: 0.6777\n",
      "Epoch 4575/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6656 - val_loss: 15.8272 - val_acc: 0.6777\n",
      "Epoch 4576/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6698 - val_loss: 15.8272 - val_acc: 0.6777\n",
      "Epoch 4577/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5617 - acc: 0.6729 - val_loss: 15.8272 - val_acc: 0.6777\n",
      "Epoch 4578/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5558 - acc: 0.6615 - val_loss: 15.8272 - val_acc: 0.6777\n",
      "Epoch 4579/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6739 - val_loss: 15.8272 - val_acc: 0.6777\n",
      "Epoch 4580/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6636 - val_loss: 15.8271 - val_acc: 0.6777\n",
      "Epoch 4581/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5642 - acc: 0.6791 - val_loss: 15.8271 - val_acc: 0.6777\n",
      "Epoch 4582/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5825 - acc: 0.6760 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4583/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6791 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4584/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5558 - acc: 0.6770 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4585/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5600 - acc: 0.6760 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4586/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5634 - acc: 0.6739 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4587/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6687 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4588/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5506 - acc: 0.6791 - val_loss: 15.8271 - val_acc: 0.6777\n",
      "Epoch 4589/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6760 - val_loss: 15.8271 - val_acc: 0.6777\n",
      "Epoch 4590/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5832 - acc: 0.6625 - val_loss: 15.8271 - val_acc: 0.6777\n",
      "Epoch 4591/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5611 - acc: 0.6656 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4592/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6667 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4593/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5767 - acc: 0.6511 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4594/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6615 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4595/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5666 - acc: 0.6801 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4596/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5545 - acc: 0.6615 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4597/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5816 - acc: 0.6698 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4598/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5602 - acc: 0.6667 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4599/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6698 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4600/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5717 - acc: 0.6760 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4601/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6677 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4602/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5633 - acc: 0.6729 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4603/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5751 - acc: 0.6749 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4604/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5544 - acc: 0.6739 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4605/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6739 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4606/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6791 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4607/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6615 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4608/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6677 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4609/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6708 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4610/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6760 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4611/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5772 - acc: 0.6698 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4612/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5569 - acc: 0.6708 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4613/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6708 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4614/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5774 - acc: 0.6636 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4615/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5560 - acc: 0.6729 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4616/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5794 - acc: 0.6708 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4617/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5628 - acc: 0.6667 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4618/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6739 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4619/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5548 - acc: 0.6760 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4620/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6718 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4621/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5750 - acc: 0.6698 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4622/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6656 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4623/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5501 - acc: 0.6718 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4624/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5725 - acc: 0.6677 - val_loss: 15.8270 - val_acc: 0.6777\n",
      "Epoch 4625/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5803 - acc: 0.6739 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4626/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6677 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4627/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5548 - acc: 0.6770 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4628/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6636 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4629/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6729 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4630/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6698 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4631/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5525 - acc: 0.6646 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4632/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6687 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4633/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5599 - acc: 0.6749 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4634/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5756 - acc: 0.6667 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4635/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6656 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4636/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5819 - acc: 0.6770 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4637/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6677 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4638/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5934 - acc: 0.6687 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4639/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5733 - acc: 0.6687 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4640/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5495 - acc: 0.6749 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4641/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5889 - acc: 0.6708 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4642/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5528 - acc: 0.6791 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4643/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5730 - acc: 0.6822 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4644/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6605 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4645/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6708 - val_loss: 15.8269 - val_acc: 0.6777\n",
      "Epoch 4646/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6698 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4647/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6760 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4648/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5918 - acc: 0.6718 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4649/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4650/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6625 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4651/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6739 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4652/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6677 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4653/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5765 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4654/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6801 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4655/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5680 - acc: 0.6760 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4656/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5539 - acc: 0.6687 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4657/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6646 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4658/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5811 - acc: 0.6739 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4659/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6708 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4660/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5781 - acc: 0.6749 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4661/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6677 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4662/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5534 - acc: 0.6718 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4663/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5660 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4664/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5603 - acc: 0.6646 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4665/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6625 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4666/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5515 - acc: 0.6698 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4667/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5688 - acc: 0.6843 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4668/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4669/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5538 - acc: 0.6687 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4670/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6718 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4671/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5561 - acc: 0.6687 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4672/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5609 - acc: 0.6698 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4673/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5598 - acc: 0.6739 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4674/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5635 - acc: 0.6739 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4675/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6708 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4676/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5836 - acc: 0.6656 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4677/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5618 - acc: 0.6646 - val_loss: 15.8268 - val_acc: 0.6777\n",
      "Epoch 4678/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5539 - acc: 0.6781 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4679/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5593 - acc: 0.6605 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4680/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5663 - acc: 0.6636 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4681/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5633 - acc: 0.6749 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4682/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6698 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4683/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5804 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4684/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5610 - acc: 0.6687 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4685/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5547 - acc: 0.6708 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4686/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6739 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4687/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6708 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4688/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6749 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4689/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6739 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4690/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5503 - acc: 0.6687 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4691/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6677 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4692/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5503 - acc: 0.6667 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4693/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6667 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4694/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5553 - acc: 0.6625 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4695/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5591 - acc: 0.6698 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4696/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5517 - acc: 0.6729 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4697/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5727 - acc: 0.6812 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4698/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6656 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4699/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6646 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4700/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6667 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4701/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6729 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4702/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5489 - acc: 0.6687 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4703/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6749 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4704/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6760 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4705/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6739 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4706/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5670 - acc: 0.6625 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4707/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5652 - acc: 0.6781 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4708/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5758 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4709/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5468 - acc: 0.6729 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4710/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6781 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4711/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5639 - acc: 0.6822 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4712/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5655 - acc: 0.6677 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4713/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5670 - acc: 0.6563 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4714/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5550 - acc: 0.6749 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4715/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6729 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4716/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6708 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4717/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6656 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4718/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5591 - acc: 0.6739 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4719/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4720/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4721/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5465 - acc: 0.6812 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4722/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4723/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5518 - acc: 0.6749 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4724/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5780 - acc: 0.6760 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4725/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6698 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4726/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6677 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4727/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6739 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4728/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6749 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4729/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4730/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5755 - acc: 0.6605 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4731/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5535 - acc: 0.6718 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4732/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5826 - acc: 0.6770 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4733/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5622 - acc: 0.6760 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4734/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6656 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4735/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6781 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4736/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6718 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4737/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6739 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4738/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6760 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4739/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5422 - acc: 0.6708 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4740/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5577 - acc: 0.6739 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4741/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5775 - acc: 0.6646 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4742/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5572 - acc: 0.6563 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4743/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5904 - acc: 0.6718 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4744/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6791 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4745/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5653 - acc: 0.6832 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4746/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5728 - acc: 0.6656 - val_loss: 15.8267 - val_acc: 0.6777\n",
      "Epoch 4747/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5703 - acc: 0.6812 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4748/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5712 - acc: 0.6749 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4749/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6625 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4750/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5548 - acc: 0.6615 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4751/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6791 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4752/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5786 - acc: 0.6718 - val_loss: 15.8266 - val_acc: 0.6777\n",
      "Epoch 4753/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6749 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4754/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5707 - acc: 0.6729 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4755/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6781 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4756/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6718 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4757/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5873 - acc: 0.6646 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4758/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5840 - acc: 0.6553 - val_loss: 15.8265 - val_acc: 0.6777\n",
      "Epoch 4759/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6822 - val_loss: 15.8264 - val_acc: 0.6777\n",
      "Epoch 4760/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6656 - val_loss: 15.8264 - val_acc: 0.6777\n",
      "Epoch 4761/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5652 - acc: 0.6729 - val_loss: 15.8264 - val_acc: 0.6777\n",
      "Epoch 4762/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5662 - acc: 0.6760 - val_loss: 15.8264 - val_acc: 0.6777\n",
      "Epoch 4763/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5513 - acc: 0.6615 - val_loss: 15.8264 - val_acc: 0.6777\n",
      "Epoch 4764/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6739 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4765/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5489 - acc: 0.6729 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4766/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5552 - acc: 0.6687 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4767/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5691 - acc: 0.6781 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4768/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5632 - acc: 0.6677 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4769/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6667 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4770/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5745 - acc: 0.6729 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4771/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6708 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4772/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5600 - acc: 0.6739 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4773/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5731 - acc: 0.6718 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4774/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5611 - acc: 0.6718 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4775/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5471 - acc: 0.6801 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4776/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6760 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4777/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5620 - acc: 0.6708 - val_loss: 15.8263 - val_acc: 0.6777\n",
      "Epoch 4778/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6677 - val_loss: 15.8262 - val_acc: 0.6777\n",
      "Epoch 4779/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5800 - acc: 0.6605 - val_loss: 15.8262 - val_acc: 0.6777\n",
      "Epoch 4780/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5742 - acc: 0.6739 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4781/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5931 - acc: 0.6687 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4782/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4783/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6625 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4784/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5853 - acc: 0.6708 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4785/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5640 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4786/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6718 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4787/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5696 - acc: 0.6625 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4788/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6739 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4789/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5589 - acc: 0.6718 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4790/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5551 - acc: 0.6708 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4791/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6646 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4792/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5514 - acc: 0.6760 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4793/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6749 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4794/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6636 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4795/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5685 - acc: 0.6646 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4796/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5647 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4797/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6646 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4798/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4799/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6749 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4800/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5615 - acc: 0.6760 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4801/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5621 - acc: 0.6677 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4802/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5554 - acc: 0.6698 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4803/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5768 - acc: 0.6801 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4804/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5714 - acc: 0.6770 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4805/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5507 - acc: 0.6739 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4806/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6698 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4807/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5715 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4808/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5671 - acc: 0.6646 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4809/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6698 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4810/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5678 - acc: 0.6646 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4811/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5778 - acc: 0.6739 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4812/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5736 - acc: 0.6636 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4813/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5748 - acc: 0.6667 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4814/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5469 - acc: 0.6781 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4815/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.5031 - acc: 0.67 - 0s - loss: 16.5751 - acc: 0.6667 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4816/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5838 - acc: 0.6791 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4817/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5547 - acc: 0.6729 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4818/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5684 - acc: 0.6708 - val_loss: 15.8261 - val_acc: 0.6777\n",
      "Epoch 4819/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6708 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4820/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5586 - acc: 0.6677 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4821/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5807 - acc: 0.6687 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4822/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5682 - acc: 0.6625 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4823/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5737 - acc: 0.6822 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4824/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6625 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4825/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5623 - acc: 0.6698 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4826/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5694 - acc: 0.6739 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4827/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5659 - acc: 0.6646 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4828/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6667 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4829/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6708 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4830/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5571 - acc: 0.6636 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4831/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5584 - acc: 0.6646 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4832/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5693 - acc: 0.6739 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4833/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6636 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4834/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6729 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4835/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6667 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4836/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5673 - acc: 0.6656 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4837/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5711 - acc: 0.6677 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4838/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5510 - acc: 0.6729 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4839/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5675 - acc: 0.6615 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4840/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5463 - acc: 0.6770 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4841/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5597 - acc: 0.6770 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4842/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5687 - acc: 0.6749 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4843/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6584 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4844/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5624 - acc: 0.6770 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4845/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5761 - acc: 0.6718 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4846/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5587 - acc: 0.6605 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4847/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6656 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4848/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6708 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4849/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5629 - acc: 0.6749 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4850/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5722 - acc: 0.6667 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4851/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6760 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4852/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5628 - acc: 0.6625 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4853/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5657 - acc: 0.6770 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4854/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5553 - acc: 0.6667 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4855/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5645 - acc: 0.6667 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4856/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6708 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4857/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5766 - acc: 0.6687 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4858/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5625 - acc: 0.6667 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4859/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5572 - acc: 0.6812 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4860/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6605 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4861/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5795 - acc: 0.6718 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4862/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5556 - acc: 0.6708 - val_loss: 15.8260 - val_acc: 0.6777\n",
      "Epoch 4863/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5665 - acc: 0.6822 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4864/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5546 - acc: 0.6646 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4865/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6739 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4866/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6770 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4867/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6636 - val_loss: 15.8259 - val_acc: 0.6777\n",
      "Epoch 4868/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5787 - acc: 0.6770 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4869/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5683 - acc: 0.6667 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4870/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6729 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4871/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5544 - acc: 0.6718 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4872/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5404 - acc: 0.6739 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4873/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5601 - acc: 0.6729 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4874/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5777 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4875/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5667 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4876/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5797 - acc: 0.6739 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4877/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5637 - acc: 0.6677 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4878/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5565 - acc: 0.6760 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4879/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6718 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4880/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5456 - acc: 0.6812 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4881/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6667 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4882/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5644 - acc: 0.6760 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4883/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4884/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4885/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5516 - acc: 0.6667 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4886/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5562 - acc: 0.6843 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4887/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5654 - acc: 0.6625 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4888/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5735 - acc: 0.6718 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4889/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5739 - acc: 0.6646 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4890/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5574 - acc: 0.6708 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4891/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5817 - acc: 0.6573 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4892/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5567 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4893/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5488 - acc: 0.6656 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4894/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5805 - acc: 0.6729 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4895/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4896/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5593 - acc: 0.6749 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4897/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6718 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4898/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5593 - acc: 0.6749 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4899/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6667 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4900/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5740 - acc: 0.6729 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4901/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5669 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4902/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5464 - acc: 0.6760 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4903/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6749 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4904/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6718 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4905/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5827 - acc: 0.6781 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4906/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5754 - acc: 0.6646 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4907/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5599 - acc: 0.6781 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4908/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5689 - acc: 0.6749 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4909/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5886 - acc: 0.6667 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4910/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6625 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4911/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5530 - acc: 0.6749 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4912/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6708 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4913/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5783 - acc: 0.6739 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4914/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5444 - acc: 0.6698 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4915/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5582 - acc: 0.6801 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4916/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5590 - acc: 0.6822 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4917/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5493 - acc: 0.6718 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4918/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5651 - acc: 0.6698 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4919/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5698 - acc: 0.6718 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4920/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5664 - acc: 0.6708 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4921/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5771 - acc: 0.6770 - val_loss: 15.8258 - val_acc: 0.6777\n",
      "Epoch 4922/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6687 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4923/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5719 - acc: 0.6760 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4924/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5789 - acc: 0.6698 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4925/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5390 - acc: 0.6687 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4926/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5616 - acc: 0.6801 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4927/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6729 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4928/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5718 - acc: 0.6760 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4929/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5686 - acc: 0.6718 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4930/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5729 - acc: 0.6739 - val_loss: 15.8257 - val_acc: 0.6777\n",
      "Epoch 4931/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6739 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4932/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5820 - acc: 0.6749 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4933/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5672 - acc: 0.6791 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4934/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6739 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4935/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6729 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4936/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5752 - acc: 0.6656 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4937/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5638 - acc: 0.6677 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4938/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5614 - acc: 0.6708 - val_loss: 15.8255 - val_acc: 0.6777\n",
      "Epoch 4939/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5706 - acc: 0.6656 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4940/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5531 - acc: 0.6739 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4941/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6687 - val_loss: 15.8256 - val_acc: 0.6777.\n",
      "Epoch 4942/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5607 - acc: 0.6770 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4943/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5681 - acc: 0.6667 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4944/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5592 - acc: 0.6781 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4945/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5612 - acc: 0.6739 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4946/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5545 - acc: 0.6677 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4947/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5597 - acc: 0.6729 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4948/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5547 - acc: 0.6718 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4949/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6760 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4950/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5402 - acc: 0.6687 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4951/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5496 - acc: 0.6801 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4952/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5605 - acc: 0.6770 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4953/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5603 - acc: 0.6698 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4954/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5570 - acc: 0.6687 - val_loss: 15.8256 - val_acc: 0.6777\n",
      "Epoch 4955/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5773 - acc: 0.6677 - val_loss: 15.8255 - val_acc: 0.6777\n",
      "Epoch 4956/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5850 - acc: 0.6749 - val_loss: 15.8255 - val_acc: 0.6777\n",
      "Epoch 4957/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5561 - acc: 0.6770 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4958/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5504 - acc: 0.6729 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4959/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5753 - acc: 0.6749 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4960/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5879 - acc: 0.6822 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4961/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5785 - acc: 0.6791 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4962/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5543 - acc: 0.6760 - val_loss: 15.8254 - val_acc: 0.6777\n",
      "Epoch 4963/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5583 - acc: 0.6687 - val_loss: 15.8253 - val_acc: 0.6777\n",
      "Epoch 4964/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6646 - val_loss: 15.8253 - val_acc: 0.6777\n",
      "Epoch 4965/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5695 - acc: 0.6832 - val_loss: 15.8253 - val_acc: 0.6777\n",
      "Epoch 4966/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5619 - acc: 0.6739 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4967/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5636 - acc: 0.6739 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4968/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5596 - acc: 0.6687 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4969/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5732 - acc: 0.6749 - val_loss: 15.8253 - val_acc: 0.6777\n",
      "Epoch 4970/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5658 - acc: 0.6687 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4971/5000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "966/966 [==============================] - 0s - loss: 16.5661 - acc: 0.6770 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4972/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5677 - acc: 0.6770 - val_loss: 15.8253 - val_acc: 0.6777\n",
      "Epoch 4973/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5496 - acc: 0.6729 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4974/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5641 - acc: 0.6749 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4975/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5744 - acc: 0.6791 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4976/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5409 - acc: 0.6739 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4977/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5481 - acc: 0.6739 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4978/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5656 - acc: 0.6708 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4979/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5724 - acc: 0.6822 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4980/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5521 - acc: 0.6739 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4981/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5545 - acc: 0.6739 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4982/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5702 - acc: 0.6718 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4983/5000\n",
      "966/966 [==============================] - ETA: 0s - loss: 16.5577 - acc: 0.67 - 0s - loss: 16.5635 - acc: 0.6770 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4984/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5704 - acc: 0.6739 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4985/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5788 - acc: 0.6760 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4986/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5779 - acc: 0.6677 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4987/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5579 - acc: 0.6677 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4988/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5617 - acc: 0.6677 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4989/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5606 - acc: 0.6739 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4990/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5721 - acc: 0.6718 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4991/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5576 - acc: 0.6739 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4992/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5701 - acc: 0.6708 - val_loss: 15.8252 - val_acc: 0.6777\n",
      "Epoch 4993/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5498 - acc: 0.6791 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4994/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5710 - acc: 0.6708 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4995/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5604 - acc: 0.6801 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4996/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5749 - acc: 0.6729 - val_loss: 15.8250 - val_acc: 0.6777\n",
      "Epoch 4997/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5627 - acc: 0.6791 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4998/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5792 - acc: 0.6729 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 4999/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5505 - acc: 0.6718 - val_loss: 15.8251 - val_acc: 0.6777\n",
      "Epoch 5000/5000\n",
      "966/966 [==============================] - 0s - loss: 16.5646 - acc: 0.6781 - val_loss: 15.8251 - val_acc: 0.6777\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c2c04d1cc0>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = create_model({'layer_1': 10, 'layer_2': 0}, 0.4)\n",
    "model.fit(x=x_train, y=x_train, batch_size=batch_size, epochs=epochs, validation_data=(x_test, x_test), callbacks=get_callbacks())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:tensorflow]",
   "language": "python",
   "name": "conda-env-tensorflow-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}