ml_option_pricing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Options Pricing Using Machine Learning."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inroduction\n",
    "\n",
    "In this project we are applying different machine learning methods to estimate Options contracts market prices based on various features. We compare our models'  errors to that of the Black-Scholes model as a baseline.\n",
    "\n",
    "We use 3 different ML models, namely, **LightGBM**, **Neural Networks**, and **Support Vector Machines**. \n",
    "\n",
    "For data, the historical data of 98 options contarcs for past 3 years in Iran's Options market were used."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Importing packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "\n",
    "\n",
    "import lightgbm as lgbm\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import RandomizedSearchCV,KFold, GridSearchCV\n",
    "\n",
    "from tensorflow.keras import layers, models\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "EPOCHS = 800\n",
    "BATCH_SIZE = 256\n",
    "VAL_SPLIT = 0.15\n",
    "TRAIN = False # set to True for training again\n",
    "\n",
    "SEED = 42\n",
    "np.random.seed(SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Loading and processing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['date_fa',\n",
       " 'symbol',\n",
       " 'exp_date',\n",
       " 'close_price',\n",
       " 'strike_price',\n",
       " 'equity_price',\n",
       " 'annual_volatility',\n",
       " 'riskfree_rate',\n",
       " 'time_until_exp',\n",
       " 'BS']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_excel(\"option_prices_data.xlsx\").dropna().reset_index().drop(columns=['index'])\n",
    "cols = list(df.columns)\n",
    "cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missing Values:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "date_fa              0\n",
       "symbol               0\n",
       "exp_date             0\n",
       "close_price          0\n",
       "strike_price         0\n",
       "equity_price         0\n",
       "annual_volatility    0\n",
       "riskfree_rate        0\n",
       "time_until_exp       0\n",
       "BS                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Missing Values:\")\n",
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>date_fa</th>\n",
       "      <th>symbol</th>\n",
       "      <th>exp_date</th>\n",
       "      <th>close_price</th>\n",
       "      <th>strike_price</th>\n",
       "      <th>equity_price</th>\n",
       "      <th>annual_volatility</th>\n",
       "      <th>riskfree_rate</th>\n",
       "      <th>time_until_exp</th>\n",
       "      <th>BS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1400/03/10</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>32</td>\n",
       "      <td>3000</td>\n",
       "      <td>1740</td>\n",
       "      <td>0.444299</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.5042</td>\n",
       "      <td>24.239068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1400/03/11</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>30</td>\n",
       "      <td>3000</td>\n",
       "      <td>1686</td>\n",
       "      <td>0.448107</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>19.314782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1400/03/12</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>37</td>\n",
       "      <td>3000</td>\n",
       "      <td>1738</td>\n",
       "      <td>0.462402</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.4958</td>\n",
       "      <td>27.025510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1400/03/17</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>48</td>\n",
       "      <td>3000</td>\n",
       "      <td>1808</td>\n",
       "      <td>0.479096</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.4750</td>\n",
       "      <td>36.700559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1400/03/18</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>53</td>\n",
       "      <td>3000</td>\n",
       "      <td>1894</td>\n",
       "      <td>0.498049</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.4708</td>\n",
       "      <td>55.503608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4863</th>\n",
       "      <td>1399/11/19</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1003</td>\n",
       "      <td>10000</td>\n",
       "      <td>10990</td>\n",
       "      <td>0.582477</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0292</td>\n",
       "      <td>1157.298735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4864</th>\n",
       "      <td>1399/11/20</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1313</td>\n",
       "      <td>10000</td>\n",
       "      <td>11510</td>\n",
       "      <td>0.598842</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0250</td>\n",
       "      <td>1609.719453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4865</th>\n",
       "      <td>1399/11/21</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1887</td>\n",
       "      <td>10000</td>\n",
       "      <td>12060</td>\n",
       "      <td>0.601912</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0208</td>\n",
       "      <td>2125.992878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4866</th>\n",
       "      <td>1399/11/25</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>2482</td>\n",
       "      <td>10000</td>\n",
       "      <td>12770</td>\n",
       "      <td>0.619110</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0042</td>\n",
       "      <td>2782.493790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4867</th>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>2117</td>\n",
       "      <td>10000</td>\n",
       "      <td>12730</td>\n",
       "      <td>0.614972</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1365.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4868 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         date_fa     symbol    exp_date  close_price  strike_price  \\\n",
       "0     1400/03/10  ضخود70771  1400/07/07           32          3000   \n",
       "1     1400/03/11  ضخود70771  1400/07/07           30          3000   \n",
       "2     1400/03/12  ضخود70771  1400/07/07           37          3000   \n",
       "3     1400/03/17  ضخود70771  1400/07/07           48          3000   \n",
       "4     1400/03/18  ضخود70771  1400/07/07           53          3000   \n",
       "...          ...        ...         ...          ...           ...   \n",
       "4863  1399/11/19  ضغدر11191  1399/11/26         1003         10000   \n",
       "4864  1399/11/20  ضغدر11191  1399/11/26         1313         10000   \n",
       "4865  1399/11/21  ضغدر11191  1399/11/26         1887         10000   \n",
       "4866  1399/11/25  ضغدر11191  1399/11/26         2482         10000   \n",
       "4867  1399/11/26  ضغدر11191  1399/11/26         2117         10000   \n",
       "\n",
       "      equity_price  annual_volatility  riskfree_rate  time_until_exp  \\\n",
       "0             1740           0.444299       0.196389          0.5042   \n",
       "1             1686           0.448107       0.196389          0.5000   \n",
       "2             1738           0.462402       0.196389          0.4958   \n",
       "3             1808           0.479096       0.196389          0.4750   \n",
       "4             1894           0.498049       0.196389          0.4708   \n",
       "...            ...                ...            ...             ...   \n",
       "4863         10990           0.582477       0.297657          0.0292   \n",
       "4864         11510           0.598842       0.297657          0.0250   \n",
       "4865         12060           0.601912       0.297657          0.0208   \n",
       "4866         12770           0.619110       0.297657          0.0042   \n",
       "4867         12730           0.614972       0.297657          0.0000   \n",
       "\n",
       "               BS  \n",
       "0       24.239068  \n",
       "1       19.314782  \n",
       "2       27.025510  \n",
       "3       36.700559  \n",
       "4       55.503608  \n",
       "...           ...  \n",
       "4863  1157.298735  \n",
       "4864  1609.719453  \n",
       "4865  2125.992878  \n",
       "4866  2782.493790  \n",
       "4867  1365.000000  \n",
       "\n",
       "[4868 rows x 10 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>close_price</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>1048.384347</td>\n",
       "      <td>1553.900115</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>137.000000</td>\n",
       "      <td>486.000000</td>\n",
       "      <td>1137.250000</td>\n",
       "      <td>11370.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>strike_price</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>5611.128595</td>\n",
       "      <td>6286.114667</td>\n",
       "      <td>500.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>3000.000000</td>\n",
       "      <td>7000.000000</td>\n",
       "      <td>39750.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>equity_price</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>5708.069638</td>\n",
       "      <td>5934.679427</td>\n",
       "      <td>488.000000</td>\n",
       "      <td>2174.000000</td>\n",
       "      <td>2890.000000</td>\n",
       "      <td>6925.000000</td>\n",
       "      <td>37120.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>annual_volatility</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>0.514253</td>\n",
       "      <td>0.222192</td>\n",
       "      <td>0.187354</td>\n",
       "      <td>0.390653</td>\n",
       "      <td>0.487253</td>\n",
       "      <td>0.587536</td>\n",
       "      <td>6.820169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>riskfree_rate</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>0.290325</td>\n",
       "      <td>0.106985</td>\n",
       "      <td>0.173617</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.294980</td>\n",
       "      <td>0.331172</td>\n",
       "      <td>0.486492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time_until_exp</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>0.206605</td>\n",
       "      <td>0.166233</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.087500</td>\n",
       "      <td>0.179200</td>\n",
       "      <td>0.291700</td>\n",
       "      <td>1.320800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BS</th>\n",
       "      <td>4868.0</td>\n",
       "      <td>1180.424673</td>\n",
       "      <td>1884.824974</td>\n",
       "      <td>-4480.000000</td>\n",
       "      <td>118.534691</td>\n",
       "      <td>496.495395</td>\n",
       "      <td>1236.836838</td>\n",
       "      <td>18273.157515</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    count         mean          std          min          25%  \\\n",
       "close_price        4868.0  1048.384347  1553.900115     1.000000   137.000000   \n",
       "strike_price       4868.0  5611.128595  6286.114667   500.000000  2000.000000   \n",
       "equity_price       4868.0  5708.069638  5934.679427   488.000000  2174.000000   \n",
       "annual_volatility  4868.0     0.514253     0.222192     0.187354     0.390653   \n",
       "riskfree_rate      4868.0     0.290325     0.106985     0.173617     0.196389   \n",
       "time_until_exp     4868.0     0.206605     0.166233     0.000000     0.087500   \n",
       "BS                 4868.0  1180.424673  1884.824974 -4480.000000   118.534691   \n",
       "\n",
       "                           50%          75%           max  \n",
       "close_price         486.000000  1137.250000  11370.000000  \n",
       "strike_price       3000.000000  7000.000000  39750.000000  \n",
       "equity_price       2890.000000  6925.000000  37120.000000  \n",
       "annual_volatility     0.487253     0.587536      6.820169  \n",
       "riskfree_rate         0.294980     0.331172      0.486492  \n",
       "time_until_exp        0.179200     0.291700      1.320800  \n",
       "BS                  496.495395  1236.836838  18273.157515  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "descriptive = df.describe().T\n",
    "# descriptive.to_csv(\"descriptive.csv\")\n",
    "descriptive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"moneyness\"] = (df[\"strike_price\"] - df[\"equity_price\"]).apply(\n",
    "    lambda x: \"ITM\" if x > 0 else (lambda x: \"OTM\" if x < 0 else \"ATM\")(x)\n",
    ")\n",
    "\n",
    "df[\"time_group\"] = df[\"time_until_exp\"].apply(\n",
    "    lambda x: \"short_term\" if x < 0.5 else \"long_term\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "symbols = df[\"symbol\"].unique()\n",
    "TRAIN_SIZE = 0.8\n",
    "\n",
    "train_df = pd.DataFrame()\n",
    "test_df = pd.DataFrame()\n",
    "\n",
    "for symbol in symbols:\n",
    "    section = df[df[\"symbol\"] == symbol].sample(frac=1).copy()\n",
    "    l = len(section)\n",
    "\n",
    "    train_df = pd.concat([train_df,section.iloc[:int(l*TRAIN_SIZE),:]])\n",
    "    test_df = pd.concat([test_df,section.iloc[int(l*TRAIN_SIZE):,:]])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "98"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(symbols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "all data:\n",
      "OTM    2951\n",
      "ITM    1909\n",
      "ATM       8\n",
      "Name: moneyness, dtype: int64\n",
      "\n",
      "============\n",
      "train data:\n",
      "OTM    2357\n",
      "ITM    1491\n",
      "ATM       6\n",
      "Name: moneyness, dtype: int64\n",
      "\n",
      "============\n",
      "all data:\n",
      "OTM    594\n",
      "ITM    418\n",
      "ATM      2\n",
      "Name: moneyness, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"all data:\")\n",
    "print(df[\"moneyness\"].value_counts())\n",
    "print('\\n============')\n",
    "print(\"train data:\")\n",
    "print(train_df[\"moneyness\"].value_counts())\n",
    "print('\\n============')\n",
    "print(\"all data:\")\n",
    "print(test_df[\"moneyness\"].value_counts())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OTM    594\n",
       "ITM    418\n",
       "ATM      2\n",
       "Name: moneyness, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df[\"moneyness\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>date_fa</th>\n",
       "      <th>symbol</th>\n",
       "      <th>exp_date</th>\n",
       "      <th>close_price</th>\n",
       "      <th>strike_price</th>\n",
       "      <th>equity_price</th>\n",
       "      <th>annual_volatility</th>\n",
       "      <th>riskfree_rate</th>\n",
       "      <th>time_until_exp</th>\n",
       "      <th>BS</th>\n",
       "      <th>moneyness</th>\n",
       "      <th>time_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1400/03/18</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>53</td>\n",
       "      <td>3000</td>\n",
       "      <td>1894</td>\n",
       "      <td>0.498049</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.4708</td>\n",
       "      <td>55.503608</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>1400/06/23</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>2</td>\n",
       "      <td>3000</td>\n",
       "      <td>2122</td>\n",
       "      <td>0.411762</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.0667</td>\n",
       "      <td>0.063147</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1400/04/07</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>56</td>\n",
       "      <td>3000</td>\n",
       "      <td>2163</td>\n",
       "      <td>0.520485</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.3875</td>\n",
       "      <td>99.071829</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1400/03/10</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>32</td>\n",
       "      <td>3000</td>\n",
       "      <td>1740</td>\n",
       "      <td>0.444299</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.5042</td>\n",
       "      <td>24.239068</td>\n",
       "      <td>ITM</td>\n",
       "      <td>long_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1400/04/21</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>81</td>\n",
       "      <td>3000</td>\n",
       "      <td>2095</td>\n",
       "      <td>0.421454</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.3375</td>\n",
       "      <td>33.434641</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4860</th>\n",
       "      <td>1399/11/14</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1353</td>\n",
       "      <td>10000</td>\n",
       "      <td>11490</td>\n",
       "      <td>0.545061</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0500</td>\n",
       "      <td>1701.657805</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4849</th>\n",
       "      <td>1399/10/29</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>560</td>\n",
       "      <td>10000</td>\n",
       "      <td>9380</td>\n",
       "      <td>0.476268</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.1125</td>\n",
       "      <td>472.065071</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4867</th>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>2117</td>\n",
       "      <td>10000</td>\n",
       "      <td>12730</td>\n",
       "      <td>0.614972</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4836</th>\n",
       "      <td>1399/10/09</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>3303</td>\n",
       "      <td>10000</td>\n",
       "      <td>12340</td>\n",
       "      <td>0.539483</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.1958</td>\n",
       "      <td>3073.564990</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4830</th>\n",
       "      <td>1399/10/01</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>3600</td>\n",
       "      <td>10000</td>\n",
       "      <td>12840</td>\n",
       "      <td>0.575815</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.2292</td>\n",
       "      <td>3684.182988</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3854 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         date_fa     symbol    exp_date  close_price  strike_price  \\\n",
       "4     1400/03/18  ضخود70771  1400/07/07           53          3000   \n",
       "62    1400/06/23  ضخود70771  1400/07/07            2          3000   \n",
       "18    1400/04/07  ضخود70771  1400/07/07           56          3000   \n",
       "0     1400/03/10  ضخود70771  1400/07/07           32          3000   \n",
       "28    1400/04/21  ضخود70771  1400/07/07           81          3000   \n",
       "...          ...        ...         ...          ...           ...   \n",
       "4860  1399/11/14  ضغدر11191  1399/11/26         1353         10000   \n",
       "4849  1399/10/29  ضغدر11191  1399/11/26          560         10000   \n",
       "4867  1399/11/26  ضغدر11191  1399/11/26         2117         10000   \n",
       "4836  1399/10/09  ضغدر11191  1399/11/26         3303         10000   \n",
       "4830  1399/10/01  ضغدر11191  1399/11/26         3600         10000   \n",
       "\n",
       "      equity_price  annual_volatility  riskfree_rate  time_until_exp  \\\n",
       "4             1894           0.498049       0.196389          0.4708   \n",
       "62            2122           0.411762       0.196389          0.0667   \n",
       "18            2163           0.520485       0.196389          0.3875   \n",
       "0             1740           0.444299       0.196389          0.5042   \n",
       "28            2095           0.421454       0.196389          0.3375   \n",
       "...            ...                ...            ...             ...   \n",
       "4860         11490           0.545061       0.297657          0.0500   \n",
       "4849          9380           0.476268       0.297657          0.1125   \n",
       "4867         12730           0.614972       0.297657          0.0000   \n",
       "4836         12340           0.539483       0.297657          0.1958   \n",
       "4830         12840           0.575815       0.297657          0.2292   \n",
       "\n",
       "               BS moneyness  time_group  \n",
       "4       55.503608       ITM  short_term  \n",
       "62       0.063147       ITM  short_term  \n",
       "18      99.071829       ITM  short_term  \n",
       "0       24.239068       ITM   long_term  \n",
       "28      33.434641       ITM  short_term  \n",
       "...           ...       ...         ...  \n",
       "4860  1701.657805       OTM  short_term  \n",
       "4849   472.065071       ITM  short_term  \n",
       "4867  1365.000000       OTM  short_term  \n",
       "4836  3073.564990       OTM  short_term  \n",
       "4830  3684.182988       OTM  short_term  \n",
       "\n",
       "[3854 rows x 12 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>date_fa</th>\n",
       "      <th>symbol</th>\n",
       "      <th>exp_date</th>\n",
       "      <th>close_price</th>\n",
       "      <th>strike_price</th>\n",
       "      <th>equity_price</th>\n",
       "      <th>annual_volatility</th>\n",
       "      <th>riskfree_rate</th>\n",
       "      <th>time_until_exp</th>\n",
       "      <th>BS</th>\n",
       "      <th>moneyness</th>\n",
       "      <th>time_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>1400/06/31</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>1</td>\n",
       "      <td>3000</td>\n",
       "      <td>1814</td>\n",
       "      <td>0.439361</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.0333</td>\n",
       "      <td>8.459631e-09</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>1400/06/24</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>1</td>\n",
       "      <td>3000</td>\n",
       "      <td>2103</td>\n",
       "      <td>0.411888</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>2.906379e-02</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>1400/05/13</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>44</td>\n",
       "      <td>3000</td>\n",
       "      <td>1971</td>\n",
       "      <td>0.371506</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.2458</td>\n",
       "      <td>3.519341e+00</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1400/04/22</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>79</td>\n",
       "      <td>3000</td>\n",
       "      <td>2060</td>\n",
       "      <td>0.421670</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>2.795615e+01</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>1400/05/23</td>\n",
       "      <td>ضخود70771</td>\n",
       "      <td>1400/07/07</td>\n",
       "      <td>34</td>\n",
       "      <td>3000</td>\n",
       "      <td>2139</td>\n",
       "      <td>0.336803</td>\n",
       "      <td>0.196389</td>\n",
       "      <td>0.2042</td>\n",
       "      <td>3.569365e+00</td>\n",
       "      <td>ITM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4842</th>\n",
       "      <td>1399/10/17</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1989</td>\n",
       "      <td>10000</td>\n",
       "      <td>11070</td>\n",
       "      <td>0.503151</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.1625</td>\n",
       "      <td>1.819499e+03</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4854</th>\n",
       "      <td>1399/11/06</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>1813</td>\n",
       "      <td>10000</td>\n",
       "      <td>11420</td>\n",
       "      <td>0.546440</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.0833</td>\n",
       "      <td>1.803542e+03</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4835</th>\n",
       "      <td>1399/10/08</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>3500</td>\n",
       "      <td>10000</td>\n",
       "      <td>12690</td>\n",
       "      <td>0.536448</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>3.402455e+03</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4827</th>\n",
       "      <td>1399/09/26</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>3492</td>\n",
       "      <td>10000</td>\n",
       "      <td>12360</td>\n",
       "      <td>0.602526</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.2500</td>\n",
       "      <td>3.370322e+03</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4839</th>\n",
       "      <td>1399/10/14</td>\n",
       "      <td>ضغدر11191</td>\n",
       "      <td>1399/11/26</td>\n",
       "      <td>2347</td>\n",
       "      <td>10000</td>\n",
       "      <td>11520</td>\n",
       "      <td>0.522472</td>\n",
       "      <td>0.297657</td>\n",
       "      <td>0.1750</td>\n",
       "      <td>2.262320e+03</td>\n",
       "      <td>OTM</td>\n",
       "      <td>short_term</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1014 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         date_fa     symbol    exp_date  close_price  strike_price  \\\n",
       "68    1400/06/31  ضخود70771  1400/07/07            1          3000   \n",
       "63    1400/06/24  ضخود70771  1400/07/07            1          3000   \n",
       "37    1400/05/13  ضخود70771  1400/07/07           44          3000   \n",
       "29    1400/04/22  ضخود70771  1400/07/07           79          3000   \n",
       "43    1400/05/23  ضخود70771  1400/07/07           34          3000   \n",
       "...          ...        ...         ...          ...           ...   \n",
       "4842  1399/10/17  ضغدر11191  1399/11/26         1989         10000   \n",
       "4854  1399/11/06  ضغدر11191  1399/11/26         1813         10000   \n",
       "4835  1399/10/08  ضغدر11191  1399/11/26         3500         10000   \n",
       "4827  1399/09/26  ضغدر11191  1399/11/26         3492         10000   \n",
       "4839  1399/10/14  ضغدر11191  1399/11/26         2347         10000   \n",
       "\n",
       "      equity_price  annual_volatility  riskfree_rate  time_until_exp  \\\n",
       "68            1814           0.439361       0.196389          0.0333   \n",
       "63            2103           0.411888       0.196389          0.0625   \n",
       "37            1971           0.371506       0.196389          0.2458   \n",
       "29            2060           0.421670       0.196389          0.3333   \n",
       "43            2139           0.336803       0.196389          0.2042   \n",
       "...            ...                ...            ...             ...   \n",
       "4842         11070           0.503151       0.297657          0.1625   \n",
       "4854         11420           0.546440       0.297657          0.0833   \n",
       "4835         12690           0.536448       0.297657          0.2000   \n",
       "4827         12360           0.602526       0.297657          0.2500   \n",
       "4839         11520           0.522472       0.297657          0.1750   \n",
       "\n",
       "                BS moneyness  time_group  \n",
       "68    8.459631e-09       ITM  short_term  \n",
       "63    2.906379e-02       ITM  short_term  \n",
       "37    3.519341e+00       ITM  short_term  \n",
       "29    2.795615e+01       ITM  short_term  \n",
       "43    3.569365e+00       ITM  short_term  \n",
       "...            ...       ...         ...  \n",
       "4842  1.819499e+03       OTM  short_term  \n",
       "4854  1.803542e+03       OTM  short_term  \n",
       "4835  3.402455e+03       OTM  short_term  \n",
       "4827  3.370322e+03       OTM  short_term  \n",
       "4839  2.262320e+03       OTM  short_term  \n",
       "\n",
       "[1014 rows x 12 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = train_df[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values\n",
    "train_labels = train_df[\"close_price\"].values\n",
    "\n",
    "test_data = test_df[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values\n",
    "test_labels = test_df[\"close_price\"].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Model training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LightGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = lgbm.LGBMRegressor(\n",
    "    # n_estimators=15000,\n",
    "    # objective= 'mse',\n",
    "    # learning_rate=0.002,\n",
    "    # max_depth=32,\n",
    "    # lambda_l1=30,\n",
    "    # num_leaves =,\n",
    "    # max_depth =\n",
    "    )\n",
    "\n",
    "param_grid = {\n",
    "    \"n_estimators\": list(range(13000,20000,1000)),\n",
    "    \"learning_rate\": np.arange(0.0001, 0.0021, 0.0002),\n",
    "    \"num_leaves\":  list(range(30, 80, 10)),\n",
    "    \"max_depth\": list(range(5, 30,4)),\n",
    "    # \"min_data_in_leaf\":  list(range(200, 10000, 100)),\n",
    "    # \"lambda_l1\": list(range( 0, 100, 5)),\n",
    "    \"reg_lambda\": list(range(50, 100, 10)),\n",
    "    # \"min_gain_to_split\": np.arange(0, 10,0.2)\n",
    "}\n",
    "\n",
    "lgb_search = RandomizedSearchCV(reg, param_grid,cv=3, n_iter=100, scoring='neg_mean_squared_error', verbose=3)\n",
    "# eval_set = [(test_data, test_labels)]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "if TRAIN:\n",
    "    # lgb_search.fit(\n",
    "    #     train_data,\n",
    "    #     train_labels\n",
    "    # )\n",
    "    reg = lgbm.LGBMRegressor(\n",
    "        n_estimators=16000,\n",
    "        objective= 'mse',\n",
    "        learning_rate=0.0008,\n",
    "        max_depth=8,\n",
    "        reg_lambda=80,\n",
    "        num_leaves =30\n",
    "        )\n",
    "    reg.fit(\n",
    "        train_data,\n",
    "        train_labels\n",
    "    )\n",
    "    best_lgb_model = reg\n",
    "    # print(lgb_search.best_params_)\n",
    "    \n",
    "    filename = 'best_lgb.sav'\n",
    "    pickle.dump(best_lgb_model, open(filename, 'wb'))\n",
    "else:\n",
    "    # load the model from disk\n",
    "    filename = 'best_lgb.sav'\n",
    "    best_lgb_model = pickle.load(open(filename, 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'boosting_type': 'gbdt', 'class_weight': None, 'colsample_bytree': 1.0, 'importance_type': 'split', 'learning_rate': 0.0008, 'max_depth': 8, 'min_child_samples': 20, 'min_child_weight': 0.001, 'min_split_gain': 0.0, 'n_estimators': 16000, 'n_jobs': -1, 'num_leaves': 30, 'objective': 'mse', 'random_state': None, 'reg_alpha': 0.0, 'reg_lambda': 80, 'silent': 'warn', 'subsample': 1.0, 'subsample_for_bin': 200000, 'subsample_freq': 0}\n"
     ]
    }
   ],
   "source": [
    "print(best_lgb_model.get_params())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgbm_pred = best_lgb_model.predict(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LightGBM test RMSE: 332.3933933819285\n",
      "Black-Sholes test RMSE: 771.9785736192837\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(\"LightGBM test RMSE:\", mean_squared_error(lgbm_pred, test_df[\"close_price\"].values)**0.5)\n",
    "print(\"Black-Sholes test RMSE:\", mean_squared_error(test_df[\"BS\"].values, test_df[\"close_price\"].values)**0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Multi-layer Perceptron (Neural Nets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "train_data_sc = scaler.fit_transform(train_data)\n",
    "test_data_sc = scaler.transform(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "nn_model = models.Sequential(\n",
    "    [\n",
    "        layers.Dense(32, activation=\"relu\", input_shape=[train_data.shape[1]]),\n",
    "        layers.BatchNormalization(),\n",
    "        layers.Dropout(0.15),\n",
    "        layers.Dense(16, activation=\"relu\"),\n",
    "        layers.BatchNormalization(),\n",
    "        layers.Dropout(0.15),\n",
    "        layers.Dense(8, activation=\"relu\"),\n",
    "        layers.BatchNormalization(),\n",
    "        layers.Dropout(0.15),\n",
    "        layers.Dense(1, activation=\"relu\"),\n",
    "    ]\n",
    ")\n",
    "\n",
    "nn_model.compile(\n",
    "    optimizer = \"adam\",\n",
    "    loss = \"mse\",\n",
    "    metrics = [tf.keras.metrics.RootMeanSquaredError()]\n",
    "  )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 32)                192       \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 32)                128       \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 16)                528       \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 16)                64        \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 16)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 8)                 136       \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 8)                 32        \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 8)                 0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 9         \n",
      "=================================================================\n",
      "Total params: 1,089\n",
      "Trainable params: 977\n",
      "Non-trainable params: 112\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "nn_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1600\n",
      "13/13 [==============================] - 1s 21ms/step - loss: 4007647.0000 - root_mean_squared_error: 2001.9109 - val_loss: 1170540.1250 - val_root_mean_squared_error: 1081.9150\n",
      "Epoch 2/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 4006667.2500 - root_mean_squared_error: 2001.6661 - val_loss: 1170480.1250 - val_root_mean_squared_error: 1081.8873\n",
      "Epoch 3/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 4004939.2500 - root_mean_squared_error: 2001.2344 - val_loss: 1170385.7500 - val_root_mean_squared_error: 1081.8436\n",
      "Epoch 4/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 4003352.7500 - root_mean_squared_error: 2000.8380 - val_loss: 1170241.5000 - val_root_mean_squared_error: 1081.7770\n",
      "Epoch 5/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 4001641.2500 - root_mean_squared_error: 2000.4103 - val_loss: 1170074.7500 - val_root_mean_squared_error: 1081.7000\n",
      "Epoch 6/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 4000099.5000 - root_mean_squared_error: 2000.0249 - val_loss: 1169835.8750 - val_root_mean_squared_error: 1081.5895\n",
      "Epoch 7/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3998648.0000 - root_mean_squared_error: 1999.6620 - val_loss: 1169536.1250 - val_root_mean_squared_error: 1081.4509\n",
      "Epoch 8/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3997498.0000 - root_mean_squared_error: 1999.3744 - val_loss: 1169350.5000 - val_root_mean_squared_error: 1081.3651\n",
      "Epoch 9/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3996005.5000 - root_mean_squared_error: 1999.0011 - val_loss: 1169271.6250 - val_root_mean_squared_error: 1081.3286\n",
      "Epoch 10/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3994851.7500 - root_mean_squared_error: 1998.7125 - val_loss: 1169198.0000 - val_root_mean_squared_error: 1081.2946\n",
      "Epoch 11/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3993378.2500 - root_mean_squared_error: 1998.3439 - val_loss: 1169065.8750 - val_root_mean_squared_error: 1081.2335\n",
      "Epoch 12/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3991976.0000 - root_mean_squared_error: 1997.9930 - val_loss: 1168957.3750 - val_root_mean_squared_error: 1081.1833\n",
      "Epoch 13/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3990755.7500 - root_mean_squared_error: 1997.6876 - val_loss: 1168860.6250 - val_root_mean_squared_error: 1081.1385\n",
      "Epoch 14/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3989895.2500 - root_mean_squared_error: 1997.4722 - val_loss: 1168705.5000 - val_root_mean_squared_error: 1081.0669\n",
      "Epoch 15/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3988295.0000 - root_mean_squared_error: 1997.0717 - val_loss: 1168591.7500 - val_root_mean_squared_error: 1081.0143\n",
      "Epoch 16/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 3987140.5000 - root_mean_squared_error: 1996.7826 - val_loss: 1168493.6250 - val_root_mean_squared_error: 1080.9689\n",
      "Epoch 17/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 3985819.7500 - root_mean_squared_error: 1996.4518 - val_loss: 1168404.0000 - val_root_mean_squared_error: 1080.9274\n",
      "Epoch 18/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3984522.0000 - root_mean_squared_error: 1996.1267 - val_loss: 1168367.2500 - val_root_mean_squared_error: 1080.9104\n",
      "Epoch 19/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3983224.0000 - root_mean_squared_error: 1995.8016 - val_loss: 1168413.2500 - val_root_mean_squared_error: 1080.9316\n",
      "Epoch 20/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3981668.2500 - root_mean_squared_error: 1995.4117 - val_loss: 1168407.3750 - val_root_mean_squared_error: 1080.9290\n",
      "Epoch 21/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3980127.5000 - root_mean_squared_error: 1995.0256 - val_loss: 1168414.7500 - val_root_mean_squared_error: 1080.9324\n",
      "Epoch 22/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3978914.7500 - root_mean_squared_error: 1994.7217 - val_loss: 1168226.8750 - val_root_mean_squared_error: 1080.8455\n",
      "Epoch 23/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3977742.2500 - root_mean_squared_error: 1994.4279 - val_loss: 1168388.8750 - val_root_mean_squared_error: 1080.9204\n",
      "Epoch 24/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3975998.2500 - root_mean_squared_error: 1993.9905 - val_loss: 1168257.5000 - val_root_mean_squared_error: 1080.8596\n",
      "Epoch 25/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3974550.7500 - root_mean_squared_error: 1993.6276 - val_loss: 1168043.3750 - val_root_mean_squared_error: 1080.7605\n",
      "Epoch 26/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3972485.2500 - root_mean_squared_error: 1993.1095 - val_loss: 1167553.1250 - val_root_mean_squared_error: 1080.5337\n",
      "Epoch 27/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3971701.5000 - root_mean_squared_error: 1992.9128 - val_loss: 1167480.8750 - val_root_mean_squared_error: 1080.5004\n",
      "Epoch 28/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3969555.5000 - root_mean_squared_error: 1992.3744 - val_loss: 1167799.0000 - val_root_mean_squared_error: 1080.6475\n",
      "Epoch 29/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3968125.0000 - root_mean_squared_error: 1992.0153 - val_loss: 1167385.7500 - val_root_mean_squared_error: 1080.4563\n",
      "Epoch 30/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3967149.0000 - root_mean_squared_error: 1991.7703 - val_loss: 1166888.2500 - val_root_mean_squared_error: 1080.2261\n",
      "Epoch 31/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3965356.7500 - root_mean_squared_error: 1991.3203 - val_loss: 1166029.3750 - val_root_mean_squared_error: 1079.8284\n",
      "Epoch 32/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3963796.7500 - root_mean_squared_error: 1990.9286 - val_loss: 1165384.1250 - val_root_mean_squared_error: 1079.5295\n",
      "Epoch 33/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3961168.7500 - root_mean_squared_error: 1990.2686 - val_loss: 1165276.3750 - val_root_mean_squared_error: 1079.4797\n",
      "Epoch 34/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3960191.7500 - root_mean_squared_error: 1990.0231 - val_loss: 1164738.0000 - val_root_mean_squared_error: 1079.2302\n",
      "Epoch 35/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3959038.0000 - root_mean_squared_error: 1989.7332 - val_loss: 1164296.1250 - val_root_mean_squared_error: 1079.0255\n",
      "Epoch 36/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3957208.7500 - root_mean_squared_error: 1989.2734 - val_loss: 1163665.2500 - val_root_mean_squared_error: 1078.7332\n",
      "Epoch 37/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3955413.7500 - root_mean_squared_error: 1988.8221 - val_loss: 1163350.5000 - val_root_mean_squared_error: 1078.5873\n",
      "Epoch 38/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3953799.0000 - root_mean_squared_error: 1988.4163 - val_loss: 1163455.6250 - val_root_mean_squared_error: 1078.6360\n",
      "Epoch 39/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3951939.0000 - root_mean_squared_error: 1987.9485 - val_loss: 1162853.2500 - val_root_mean_squared_error: 1078.3567\n",
      "Epoch 40/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3950035.5000 - root_mean_squared_error: 1987.4696 - val_loss: 1161827.8750 - val_root_mean_squared_error: 1077.8812\n",
      "Epoch 41/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3948173.7500 - root_mean_squared_error: 1987.0012 - val_loss: 1161178.8750 - val_root_mean_squared_error: 1077.5801\n",
      "Epoch 42/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3945822.2500 - root_mean_squared_error: 1986.4094 - val_loss: 1160352.7500 - val_root_mean_squared_error: 1077.1967\n",
      "Epoch 43/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3942582.5000 - root_mean_squared_error: 1985.5938 - val_loss: 1160262.3750 - val_root_mean_squared_error: 1077.1548\n",
      "Epoch 44/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3941773.7500 - root_mean_squared_error: 1985.3900 - val_loss: 1160087.5000 - val_root_mean_squared_error: 1077.0736\n",
      "Epoch 45/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3940586.2500 - root_mean_squared_error: 1985.0909 - val_loss: 1159700.8750 - val_root_mean_squared_error: 1076.8940\n",
      "Epoch 46/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3938338.0000 - root_mean_squared_error: 1984.5247 - val_loss: 1157971.8750 - val_root_mean_squared_error: 1076.0911\n",
      "Epoch 47/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3935560.2500 - root_mean_squared_error: 1983.8247 - val_loss: 1157046.6250 - val_root_mean_squared_error: 1075.6610\n",
      "Epoch 48/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3933098.2500 - root_mean_squared_error: 1983.2040 - val_loss: 1156466.1250 - val_root_mean_squared_error: 1075.3911\n",
      "Epoch 49/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3932207.5000 - root_mean_squared_error: 1982.9795 - val_loss: 1155891.7500 - val_root_mean_squared_error: 1075.1240\n",
      "Epoch 50/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3930695.2500 - root_mean_squared_error: 1982.5981 - val_loss: 1154843.2500 - val_root_mean_squared_error: 1074.6364\n",
      "Epoch 51/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3929366.5000 - root_mean_squared_error: 1982.2629 - val_loss: 1154533.5000 - val_root_mean_squared_error: 1074.4922\n",
      "Epoch 52/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3925434.0000 - root_mean_squared_error: 1981.2708 - val_loss: 1153259.6250 - val_root_mean_squared_error: 1073.8993\n",
      "Epoch 53/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3922812.0000 - root_mean_squared_error: 1980.6090 - val_loss: 1152839.0000 - val_root_mean_squared_error: 1073.7034\n",
      "Epoch 54/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3921868.0000 - root_mean_squared_error: 1980.3707 - val_loss: 1151357.7500 - val_root_mean_squared_error: 1073.0134\n",
      "Epoch 55/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3918488.5000 - root_mean_squared_error: 1979.5172 - val_loss: 1150617.8750 - val_root_mean_squared_error: 1072.6686\n",
      "Epoch 56/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3916865.7500 - root_mean_squared_error: 1979.1073 - val_loss: 1150673.3750 - val_root_mean_squared_error: 1072.6945\n",
      "Epoch 57/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3912771.0000 - root_mean_squared_error: 1978.0725 - val_loss: 1149558.2500 - val_root_mean_squared_error: 1072.1746\n",
      "Epoch 58/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3911416.7500 - root_mean_squared_error: 1977.7302 - val_loss: 1148297.7500 - val_root_mean_squared_error: 1071.5865\n",
      "Epoch 59/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3908819.0000 - root_mean_squared_error: 1977.0734 - val_loss: 1147904.5000 - val_root_mean_squared_error: 1071.4031\n",
      "Epoch 60/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3907413.0000 - root_mean_squared_error: 1976.7178 - val_loss: 1146191.7500 - val_root_mean_squared_error: 1070.6034\n",
      "Epoch 61/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3903835.0000 - root_mean_squared_error: 1975.8125 - val_loss: 1145352.6250 - val_root_mean_squared_error: 1070.2114\n",
      "Epoch 62/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3901446.0000 - root_mean_squared_error: 1975.2079 - val_loss: 1144946.2500 - val_root_mean_squared_error: 1070.0216\n",
      "Epoch 63/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3898967.5000 - root_mean_squared_error: 1974.5803 - val_loss: 1143825.7500 - val_root_mean_squared_error: 1069.4979\n",
      "Epoch 64/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3896664.2500 - root_mean_squared_error: 1973.9971 - val_loss: 1142285.8750 - val_root_mean_squared_error: 1068.7777\n",
      "Epoch 65/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3895290.0000 - root_mean_squared_error: 1973.6489 - val_loss: 1140999.5000 - val_root_mean_squared_error: 1068.1758\n",
      "Epoch 66/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3891533.2500 - root_mean_squared_error: 1972.6969 - val_loss: 1140127.3750 - val_root_mean_squared_error: 1067.7675\n",
      "Epoch 67/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3887084.2500 - root_mean_squared_error: 1971.5690 - val_loss: 1140066.3750 - val_root_mean_squared_error: 1067.7389\n",
      "Epoch 68/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3885971.0000 - root_mean_squared_error: 1971.2866 - val_loss: 1139595.8750 - val_root_mean_squared_error: 1067.5186\n",
      "Epoch 69/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3882390.7500 - root_mean_squared_error: 1970.3783 - val_loss: 1140007.1250 - val_root_mean_squared_error: 1067.7112\n",
      "Epoch 70/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3880799.2500 - root_mean_squared_error: 1969.9745 - val_loss: 1138961.3750 - val_root_mean_squared_error: 1067.2213\n",
      "Epoch 71/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3877049.0000 - root_mean_squared_error: 1969.0223 - val_loss: 1136783.7500 - val_root_mean_squared_error: 1066.2006\n",
      "Epoch 72/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3874970.2500 - root_mean_squared_error: 1968.4944 - val_loss: 1135811.6250 - val_root_mean_squared_error: 1065.7446\n",
      "Epoch 73/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3871996.0000 - root_mean_squared_error: 1967.7388 - val_loss: 1133822.0000 - val_root_mean_squared_error: 1064.8108\n",
      "Epoch 74/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3868769.5000 - root_mean_squared_error: 1966.9188 - val_loss: 1134525.2500 - val_root_mean_squared_error: 1065.1410\n",
      "Epoch 75/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3863715.5000 - root_mean_squared_error: 1965.6337 - val_loss: 1134464.5000 - val_root_mean_squared_error: 1065.1124\n",
      "Epoch 76/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3863219.0000 - root_mean_squared_error: 1965.5073 - val_loss: 1133043.6250 - val_root_mean_squared_error: 1064.4452\n",
      "Epoch 77/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3860265.0000 - root_mean_squared_error: 1964.7557 - val_loss: 1132032.1250 - val_root_mean_squared_error: 1063.9700\n",
      "Epoch 78/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3855187.5000 - root_mean_squared_error: 1963.4631 - val_loss: 1133360.0000 - val_root_mean_squared_error: 1064.5939\n",
      "Epoch 79/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3854293.2500 - root_mean_squared_error: 1963.2354 - val_loss: 1132632.2500 - val_root_mean_squared_error: 1064.2520\n",
      "Epoch 80/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3849521.5000 - root_mean_squared_error: 1962.0198 - val_loss: 1130769.2500 - val_root_mean_squared_error: 1063.3763\n",
      "Epoch 81/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3848648.5000 - root_mean_squared_error: 1961.7972 - val_loss: 1130263.3750 - val_root_mean_squared_error: 1063.1384\n",
      "Epoch 82/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3847412.5000 - root_mean_squared_error: 1961.4822 - val_loss: 1129192.6250 - val_root_mean_squared_error: 1062.6348\n",
      "Epoch 83/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3839413.7500 - root_mean_squared_error: 1959.4423 - val_loss: 1130029.5000 - val_root_mean_squared_error: 1063.0284\n",
      "Epoch 84/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3837051.5000 - root_mean_squared_error: 1958.8394 - val_loss: 1129522.8750 - val_root_mean_squared_error: 1062.7902\n",
      "Epoch 85/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3834653.2500 - root_mean_squared_error: 1958.2271 - val_loss: 1128810.5000 - val_root_mean_squared_error: 1062.4550\n",
      "Epoch 86/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3831177.7500 - root_mean_squared_error: 1957.3395 - val_loss: 1127250.2500 - val_root_mean_squared_error: 1061.7205\n",
      "Epoch 87/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3831022.2500 - root_mean_squared_error: 1957.2997 - val_loss: 1126599.5000 - val_root_mean_squared_error: 1061.4139\n",
      "Epoch 88/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3825465.5000 - root_mean_squared_error: 1955.8798 - val_loss: 1125471.6250 - val_root_mean_squared_error: 1060.8824\n",
      "Epoch 89/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3821577.7500 - root_mean_squared_error: 1954.8856 - val_loss: 1125123.0000 - val_root_mean_squared_error: 1060.7181\n",
      "Epoch 90/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3817616.7500 - root_mean_squared_error: 1953.8722 - val_loss: 1125157.6250 - val_root_mean_squared_error: 1060.7345\n",
      "Epoch 91/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3816006.5000 - root_mean_squared_error: 1953.4601 - val_loss: 1124136.6250 - val_root_mean_squared_error: 1060.2531\n",
      "Epoch 92/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3810945.0000 - root_mean_squared_error: 1952.1642 - val_loss: 1123690.0000 - val_root_mean_squared_error: 1060.0425\n",
      "Epoch 93/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 3809259.7500 - root_mean_squared_error: 1951.7325 - val_loss: 1123313.3750 - val_root_mean_squared_error: 1059.8647\n",
      "Epoch 94/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3805331.2500 - root_mean_squared_error: 1950.7258 - val_loss: 1123555.0000 - val_root_mean_squared_error: 1059.9788\n",
      "Epoch 95/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3797499.2500 - root_mean_squared_error: 1948.7173 - val_loss: 1122036.0000 - val_root_mean_squared_error: 1059.2620\n",
      "Epoch 96/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3797172.7500 - root_mean_squared_error: 1948.6335 - val_loss: 1121557.2500 - val_root_mean_squared_error: 1059.0360\n",
      "Epoch 97/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3795501.7500 - root_mean_squared_error: 1948.2047 - val_loss: 1119710.7500 - val_root_mean_squared_error: 1058.1638\n",
      "Epoch 98/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3793140.5000 - root_mean_squared_error: 1947.5986 - val_loss: 1119476.1250 - val_root_mean_squared_error: 1058.0530\n",
      "Epoch 99/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3789396.5000 - root_mean_squared_error: 1946.6372 - val_loss: 1117323.2500 - val_root_mean_squared_error: 1057.0352\n",
      "Epoch 100/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3782199.2500 - root_mean_squared_error: 1944.7877 - val_loss: 1117306.2500 - val_root_mean_squared_error: 1057.0271\n",
      "Epoch 101/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3780780.0000 - root_mean_squared_error: 1944.4229 - val_loss: 1117279.8750 - val_root_mean_squared_error: 1057.0145\n",
      "Epoch 102/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3778268.2500 - root_mean_squared_error: 1943.7769 - val_loss: 1114489.6250 - val_root_mean_squared_error: 1055.6938\n",
      "Epoch 103/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3777297.0000 - root_mean_squared_error: 1943.5270 - val_loss: 1114215.7500 - val_root_mean_squared_error: 1055.5642\n",
      "Epoch 104/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3769954.7500 - root_mean_squared_error: 1941.6371 - val_loss: 1112275.7500 - val_root_mean_squared_error: 1054.6449\n",
      "Epoch 105/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3768089.7500 - root_mean_squared_error: 1941.1569 - val_loss: 1113636.3750 - val_root_mean_squared_error: 1055.2897\n",
      "Epoch 106/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3765753.5000 - root_mean_squared_error: 1940.5549 - val_loss: 1112217.6250 - val_root_mean_squared_error: 1054.6173\n",
      "Epoch 107/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3759097.5000 - root_mean_squared_error: 1938.8392 - val_loss: 1109871.0000 - val_root_mean_squared_error: 1053.5042\n",
      "Epoch 108/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3758547.2500 - root_mean_squared_error: 1938.6973 - val_loss: 1108408.8750 - val_root_mean_squared_error: 1052.8099\n",
      "Epoch 109/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3753905.5000 - root_mean_squared_error: 1937.4998 - val_loss: 1108909.5000 - val_root_mean_squared_error: 1053.0477\n",
      "Epoch 110/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3747648.0000 - root_mean_squared_error: 1935.8843 - val_loss: 1111385.6250 - val_root_mean_squared_error: 1054.2228\n",
      "Epoch 111/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3743738.0000 - root_mean_squared_error: 1934.8741 - val_loss: 1109452.6250 - val_root_mean_squared_error: 1053.3055\n",
      "Epoch 112/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3741390.5000 - root_mean_squared_error: 1934.2675 - val_loss: 1107959.6250 - val_root_mean_squared_error: 1052.5966\n",
      "Epoch 113/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3737417.0000 - root_mean_squared_error: 1933.2400 - val_loss: 1106588.3750 - val_root_mean_squared_error: 1051.9451\n",
      "Epoch 114/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3736110.7500 - root_mean_squared_error: 1932.9021 - val_loss: 1107352.8750 - val_root_mean_squared_error: 1052.3083\n",
      "Epoch 115/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3730801.2500 - root_mean_squared_error: 1931.5282 - val_loss: 1106439.5000 - val_root_mean_squared_error: 1051.8743\n",
      "Epoch 116/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3725741.5000 - root_mean_squared_error: 1930.2180 - val_loss: 1103429.2500 - val_root_mean_squared_error: 1050.4424\n",
      "Epoch 117/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3722897.0000 - root_mean_squared_error: 1929.4811 - val_loss: 1104448.3750 - val_root_mean_squared_error: 1050.9274\n",
      "Epoch 118/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3719231.5000 - root_mean_squared_error: 1928.5309 - val_loss: 1101982.0000 - val_root_mean_squared_error: 1049.7533\n",
      "Epoch 119/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3711375.0000 - root_mean_squared_error: 1926.4929 - val_loss: 1102396.8750 - val_root_mean_squared_error: 1049.9509\n",
      "Epoch 120/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3711532.5000 - root_mean_squared_error: 1926.5338 - val_loss: 1100244.6250 - val_root_mean_squared_error: 1048.9254\n",
      "Epoch 121/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3706582.0000 - root_mean_squared_error: 1925.2485 - val_loss: 1101751.7500 - val_root_mean_squared_error: 1049.6437\n",
      "Epoch 122/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3700240.5000 - root_mean_squared_error: 1923.6010 - val_loss: 1097496.8750 - val_root_mean_squared_error: 1047.6149\n",
      "Epoch 123/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3697635.0000 - root_mean_squared_error: 1922.9236 - val_loss: 1096798.3750 - val_root_mean_squared_error: 1047.2814\n",
      "Epoch 124/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3691289.0000 - root_mean_squared_error: 1921.2727 - val_loss: 1091772.7500 - val_root_mean_squared_error: 1044.8793\n",
      "Epoch 125/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3692982.2500 - root_mean_squared_error: 1921.7134 - val_loss: 1091810.1250 - val_root_mean_squared_error: 1044.8972\n",
      "Epoch 126/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3686931.0000 - root_mean_squared_error: 1920.1383 - val_loss: 1089710.2500 - val_root_mean_squared_error: 1043.8918\n",
      "Epoch 127/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3684994.7500 - root_mean_squared_error: 1919.6340 - val_loss: 1090386.7500 - val_root_mean_squared_error: 1044.2158\n",
      "Epoch 128/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3679212.5000 - root_mean_squared_error: 1918.1273 - val_loss: 1091003.0000 - val_root_mean_squared_error: 1044.5109\n",
      "Epoch 129/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3672633.0000 - root_mean_squared_error: 1916.4115 - val_loss: 1091303.7500 - val_root_mean_squared_error: 1044.6549\n",
      "Epoch 130/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3669519.5000 - root_mean_squared_error: 1915.5990 - val_loss: 1088259.3750 - val_root_mean_squared_error: 1043.1967\n",
      "Epoch 131/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3665108.0000 - root_mean_squared_error: 1914.4471 - val_loss: 1085817.3750 - val_root_mean_squared_error: 1042.0256\n",
      "Epoch 132/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3664312.2500 - root_mean_squared_error: 1914.2394 - val_loss: 1083034.2500 - val_root_mean_squared_error: 1040.6893\n",
      "Epoch 133/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3658625.0000 - root_mean_squared_error: 1912.7533 - val_loss: 1082372.8750 - val_root_mean_squared_error: 1040.3715\n",
      "Epoch 134/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3655500.5000 - root_mean_squared_error: 1911.9363 - val_loss: 1082431.6250 - val_root_mean_squared_error: 1040.3998\n",
      "Epoch 135/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3650935.0000 - root_mean_squared_error: 1910.7419 - val_loss: 1080157.8750 - val_root_mean_squared_error: 1039.3064\n",
      "Epoch 136/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3648216.0000 - root_mean_squared_error: 1910.0304 - val_loss: 1079261.6250 - val_root_mean_squared_error: 1038.8751\n",
      "Epoch 137/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3642941.2500 - root_mean_squared_error: 1908.6490 - val_loss: 1075441.3750 - val_root_mean_squared_error: 1037.0349\n",
      "Epoch 138/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3641496.5000 - root_mean_squared_error: 1908.2705 - val_loss: 1076110.7500 - val_root_mean_squared_error: 1037.3575\n",
      "Epoch 139/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3631649.5000 - root_mean_squared_error: 1905.6887 - val_loss: 1074575.2500 - val_root_mean_squared_error: 1036.6172\n",
      "Epoch 140/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3631675.2500 - root_mean_squared_error: 1905.6954 - val_loss: 1073032.0000 - val_root_mean_squared_error: 1035.8726\n",
      "Epoch 141/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3621602.2500 - root_mean_squared_error: 1903.0508 - val_loss: 1070528.5000 - val_root_mean_squared_error: 1034.6635\n",
      "Epoch 142/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3624923.5000 - root_mean_squared_error: 1903.9232 - val_loss: 1068860.7500 - val_root_mean_squared_error: 1033.8572\n",
      "Epoch 143/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3611978.7500 - root_mean_squared_error: 1900.5206 - val_loss: 1073164.3750 - val_root_mean_squared_error: 1035.9365\n",
      "Epoch 144/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3613210.5000 - root_mean_squared_error: 1900.8447 - val_loss: 1075995.8750 - val_root_mean_squared_error: 1037.3022\n",
      "Epoch 145/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3609372.2500 - root_mean_squared_error: 1899.8348 - val_loss: 1067794.7500 - val_root_mean_squared_error: 1033.3416\n",
      "Epoch 146/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3602404.7500 - root_mean_squared_error: 1898.0002 - val_loss: 1066330.5000 - val_root_mean_squared_error: 1032.6328\n",
      "Epoch 147/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3597054.0000 - root_mean_squared_error: 1896.5901 - val_loss: 1064036.5000 - val_root_mean_squared_error: 1031.5215\n",
      "Epoch 148/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3590877.0000 - root_mean_squared_error: 1894.9609 - val_loss: 1066070.3750 - val_root_mean_squared_error: 1032.5068\n",
      "Epoch 149/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3587598.0000 - root_mean_squared_error: 1894.0956 - val_loss: 1066879.7500 - val_root_mean_squared_error: 1032.8987\n",
      "Epoch 150/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 3983078.0000 - root_mean_squared_error: 1995.76 - 0s 5ms/step - loss: 3590602.0000 - root_mean_squared_error: 1894.8884 - val_loss: 1066941.8750 - val_root_mean_squared_error: 1032.9288\n",
      "Epoch 151/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3585152.2500 - root_mean_squared_error: 1893.4498 - val_loss: 1055958.8750 - val_root_mean_squared_error: 1027.5986\n",
      "Epoch 152/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3573708.7500 - root_mean_squared_error: 1890.4255 - val_loss: 1058944.8750 - val_root_mean_squared_error: 1029.0505\n",
      "Epoch 153/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3569768.7500 - root_mean_squared_error: 1889.3832 - val_loss: 1059314.8750 - val_root_mean_squared_error: 1029.2302\n",
      "Epoch 154/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3563987.5000 - root_mean_squared_error: 1887.8527 - val_loss: 1066617.7500 - val_root_mean_squared_error: 1032.7719\n",
      "Epoch 155/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3560568.7500 - root_mean_squared_error: 1886.9469 - val_loss: 1059334.5000 - val_root_mean_squared_error: 1029.2397\n",
      "Epoch 156/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3561235.2500 - root_mean_squared_error: 1887.1235 - val_loss: 1057837.2500 - val_root_mean_squared_error: 1028.5122\n",
      "Epoch 157/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3553377.0000 - root_mean_squared_error: 1885.0403 - val_loss: 1053450.5000 - val_root_mean_squared_error: 1026.3773\n",
      "Epoch 158/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3549825.2500 - root_mean_squared_error: 1884.0980 - val_loss: 1057686.3750 - val_root_mean_squared_error: 1028.4388\n",
      "Epoch 159/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3542287.5000 - root_mean_squared_error: 1882.0966 - val_loss: 1050037.6250 - val_root_mean_squared_error: 1024.7134\n",
      "Epoch 160/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3532239.5000 - root_mean_squared_error: 1879.4253 - val_loss: 1044839.1250 - val_root_mean_squared_error: 1022.1737\n",
      "Epoch 161/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3531370.5000 - root_mean_squared_error: 1879.1941 - val_loss: 1043179.4375 - val_root_mean_squared_error: 1021.3616\n",
      "Epoch 162/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3525749.2500 - root_mean_squared_error: 1877.6979 - val_loss: 1046236.4375 - val_root_mean_squared_error: 1022.8569\n",
      "Epoch 163/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3523661.2500 - root_mean_squared_error: 1877.1417 - val_loss: 1047905.6250 - val_root_mean_squared_error: 1023.6726\n",
      "Epoch 164/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3520851.7500 - root_mean_squared_error: 1876.3933 - val_loss: 1046150.1875 - val_root_mean_squared_error: 1022.8148\n",
      "Epoch 165/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 3181977.7500 - root_mean_squared_error: 1783.80 - 0s 4ms/step - loss: 3522941.2500 - root_mean_squared_error: 1876.9500 - val_loss: 1037658.5000 - val_root_mean_squared_error: 1018.6552\n",
      "Epoch 166/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3506971.0000 - root_mean_squared_error: 1872.6908 - val_loss: 1041055.3125 - val_root_mean_squared_error: 1020.3212\n",
      "Epoch 167/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3509094.5000 - root_mean_squared_error: 1873.2577 - val_loss: 1034221.1875 - val_root_mean_squared_error: 1016.9667\n",
      "Epoch 168/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3505210.7500 - root_mean_squared_error: 1872.2208 - val_loss: 1035996.5000 - val_root_mean_squared_error: 1017.8391\n",
      "Epoch 169/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3496238.7500 - root_mean_squared_error: 1869.8232 - val_loss: 1034050.0000 - val_root_mean_squared_error: 1016.8825\n",
      "Epoch 170/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3494490.5000 - root_mean_squared_error: 1869.3556 - val_loss: 1032683.6875 - val_root_mean_squared_error: 1016.2104\n",
      "Epoch 171/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3493472.5000 - root_mean_squared_error: 1869.0834 - val_loss: 1034422.8125 - val_root_mean_squared_error: 1017.0658\n",
      "Epoch 172/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3475218.7500 - root_mean_squared_error: 1864.1938 - val_loss: 1023660.0000 - val_root_mean_squared_error: 1011.7609\n",
      "Epoch 173/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3477156.0000 - root_mean_squared_error: 1864.7134 - val_loss: 1026610.8750 - val_root_mean_squared_error: 1013.2181\n",
      "Epoch 174/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3471049.0000 - root_mean_squared_error: 1863.0752 - val_loss: 1028607.3125 - val_root_mean_squared_error: 1014.2028\n",
      "Epoch 175/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3470196.5000 - root_mean_squared_error: 1862.8463 - val_loss: 1024919.4375 - val_root_mean_squared_error: 1012.3831\n",
      "Epoch 176/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3462853.7500 - root_mean_squared_error: 1860.8745 - val_loss: 1021709.9375 - val_root_mean_squared_error: 1010.7967\n",
      "Epoch 177/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3448600.5000 - root_mean_squared_error: 1857.0408 - val_loss: 1018853.9375 - val_root_mean_squared_error: 1009.3829\n",
      "Epoch 178/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 3456042.2500 - root_mean_squared_error: 1859.0433 - val_loss: 1017458.7500 - val_root_mean_squared_error: 1008.6916\n",
      "Epoch 179/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3451497.2500 - root_mean_squared_error: 1857.8206 - val_loss: 1017625.5625 - val_root_mean_squared_error: 1008.7743\n",
      "Epoch 180/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3438768.2500 - root_mean_squared_error: 1854.3916 - val_loss: 1018515.1250 - val_root_mean_squared_error: 1009.2151\n",
      "Epoch 181/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3436912.2500 - root_mean_squared_error: 1853.8911 - val_loss: 1013586.5625 - val_root_mean_squared_error: 1006.7704\n",
      "Epoch 182/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3430982.5000 - root_mean_squared_error: 1852.2911 - val_loss: 1016079.5000 - val_root_mean_squared_error: 1008.0077\n",
      "Epoch 183/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3430937.2500 - root_mean_squared_error: 1852.2789 - val_loss: 1012984.1250 - val_root_mean_squared_error: 1006.4711\n",
      "Epoch 184/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3430124.7500 - root_mean_squared_error: 1852.0596 - val_loss: 1009221.8750 - val_root_mean_squared_error: 1004.6003\n",
      "Epoch 185/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3419994.5000 - root_mean_squared_error: 1849.3228 - val_loss: 1011816.1250 - val_root_mean_squared_error: 1005.8907\n",
      "Epoch 186/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3414889.2500 - root_mean_squared_error: 1847.9419 - val_loss: 1005260.8125 - val_root_mean_squared_error: 1002.6270\n",
      "Epoch 187/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3414579.0000 - root_mean_squared_error: 1847.8579 - val_loss: 999758.0625 - val_root_mean_squared_error: 999.8790\n",
      "Epoch 188/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3405729.5000 - root_mean_squared_error: 1845.4619 - val_loss: 996908.1875 - val_root_mean_squared_error: 998.4529\n",
      "Epoch 189/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3403637.5000 - root_mean_squared_error: 1844.8950 - val_loss: 992288.0625 - val_root_mean_squared_error: 996.1366\n",
      "Epoch 190/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3394643.2500 - root_mean_squared_error: 1842.4558 - val_loss: 996304.0000 - val_root_mean_squared_error: 998.1503\n",
      "Epoch 191/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3386626.2500 - root_mean_squared_error: 1840.2788 - val_loss: 996368.2500 - val_root_mean_squared_error: 998.1825\n",
      "Epoch 192/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3382584.0000 - root_mean_squared_error: 1839.1803 - val_loss: 990537.0625 - val_root_mean_squared_error: 995.2573\n",
      "Epoch 193/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3377749.2500 - root_mean_squared_error: 1837.8654 - val_loss: 990631.2500 - val_root_mean_squared_error: 995.3046\n",
      "Epoch 194/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3373966.5000 - root_mean_squared_error: 1836.8361 - val_loss: 988570.5000 - val_root_mean_squared_error: 994.2688\n",
      "Epoch 195/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3368744.0000 - root_mean_squared_error: 1835.4138 - val_loss: 991130.5000 - val_root_mean_squared_error: 995.5554\n",
      "Epoch 196/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3371822.7500 - root_mean_squared_error: 1836.2523 - val_loss: 986619.8125 - val_root_mean_squared_error: 993.2874\n",
      "Epoch 197/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3364321.7500 - root_mean_squared_error: 1834.2087 - val_loss: 988418.4375 - val_root_mean_squared_error: 994.1924\n",
      "Epoch 198/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3364023.0000 - root_mean_squared_error: 1834.1273 - val_loss: 991525.3750 - val_root_mean_squared_error: 995.7537\n",
      "Epoch 199/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3354488.5000 - root_mean_squared_error: 1831.5262 - val_loss: 989791.8125 - val_root_mean_squared_error: 994.8828\n",
      "Epoch 200/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3350341.7500 - root_mean_squared_error: 1830.3939 - val_loss: 978929.6250 - val_root_mean_squared_error: 989.4088\n",
      "Epoch 201/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3337972.7500 - root_mean_squared_error: 1827.0120 - val_loss: 977156.1875 - val_root_mean_squared_error: 988.5121\n",
      "Epoch 202/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3330787.7500 - root_mean_squared_error: 1825.0446 - val_loss: 979571.8125 - val_root_mean_squared_error: 989.7332\n",
      "Epoch 203/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3327605.5000 - root_mean_squared_error: 1824.1726 - val_loss: 977403.8125 - val_root_mean_squared_error: 988.6373\n",
      "Epoch 204/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3321525.5000 - root_mean_squared_error: 1822.5052 - val_loss: 982349.5000 - val_root_mean_squared_error: 991.1354\n",
      "Epoch 205/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 3319376.0000 - root_mean_squared_error: 1821.9155 - val_loss: 976083.8125 - val_root_mean_squared_error: 987.9695\n",
      "Epoch 206/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3312582.7500 - root_mean_squared_error: 1820.0502 - val_loss: 982791.3750 - val_root_mean_squared_error: 991.3583\n",
      "Epoch 207/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3306094.2500 - root_mean_squared_error: 1818.2668 - val_loss: 976321.0000 - val_root_mean_squared_error: 988.0896\n",
      "Epoch 208/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3304943.2500 - root_mean_squared_error: 1817.9503 - val_loss: 978945.4375 - val_root_mean_squared_error: 989.4167\n",
      "Epoch 209/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3292293.5000 - root_mean_squared_error: 1814.4678 - val_loss: 973920.6875 - val_root_mean_squared_error: 986.8742\n",
      "Epoch 210/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3288141.0000 - root_mean_squared_error: 1813.3232 - val_loss: 970375.5000 - val_root_mean_squared_error: 985.0764\n",
      "Epoch 211/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3278390.0000 - root_mean_squared_error: 1810.6324 - val_loss: 968684.7500 - val_root_mean_squared_error: 984.2178\n",
      "Epoch 212/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3280700.0000 - root_mean_squared_error: 1811.2703 - val_loss: 967322.3125 - val_root_mean_squared_error: 983.5255\n",
      "Epoch 213/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3278264.7500 - root_mean_squared_error: 1810.5979 - val_loss: 958711.0625 - val_root_mean_squared_error: 979.1379\n",
      "Epoch 214/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3274763.0000 - root_mean_squared_error: 1809.6306 - val_loss: 962282.1250 - val_root_mean_squared_error: 980.9598\n",
      "Epoch 215/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3264955.7500 - root_mean_squared_error: 1806.9188 - val_loss: 962537.6875 - val_root_mean_squared_error: 981.0900\n",
      "Epoch 216/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3263743.5000 - root_mean_squared_error: 1806.5834 - val_loss: 952186.3750 - val_root_mean_squared_error: 975.8004\n",
      "Epoch 217/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3248394.5000 - root_mean_squared_error: 1802.3303 - val_loss: 954496.6875 - val_root_mean_squared_error: 976.9835\n",
      "Epoch 218/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3259937.0000 - root_mean_squared_error: 1805.5295 - val_loss: 952170.0000 - val_root_mean_squared_error: 975.7920\n",
      "Epoch 219/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3245736.2500 - root_mean_squared_error: 1801.5927 - val_loss: 960951.5000 - val_root_mean_squared_error: 980.2813\n",
      "Epoch 220/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3242832.0000 - root_mean_squared_error: 1800.7865 - val_loss: 955311.5000 - val_root_mean_squared_error: 977.4004\n",
      "Epoch 221/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3226933.7500 - root_mean_squared_error: 1796.3668 - val_loss: 949329.1250 - val_root_mean_squared_error: 974.3352\n",
      "Epoch 222/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3223787.7500 - root_mean_squared_error: 1795.4910 - val_loss: 955521.3125 - val_root_mean_squared_error: 977.5077\n",
      "Epoch 223/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3236456.2500 - root_mean_squared_error: 1799.0154 - val_loss: 948403.9375 - val_root_mean_squared_error: 973.8604\n",
      "Epoch 224/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3213107.0000 - root_mean_squared_error: 1792.5142 - val_loss: 938656.5000 - val_root_mean_squared_error: 968.8429\n",
      "Epoch 225/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3211234.5000 - root_mean_squared_error: 1791.9918 - val_loss: 942251.2500 - val_root_mean_squared_error: 970.6963\n",
      "Epoch 226/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3196950.5000 - root_mean_squared_error: 1788.0018 - val_loss: 940677.1875 - val_root_mean_squared_error: 969.8851\n",
      "Epoch 227/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3185584.2500 - root_mean_squared_error: 1784.8206 - val_loss: 942656.6875 - val_root_mean_squared_error: 970.9051\n",
      "Epoch 228/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 3194794.0000 - root_mean_squared_error: 1787.3987 - val_loss: 945677.6875 - val_root_mean_squared_error: 972.4596\n",
      "Epoch 229/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 3193925.7500 - root_mean_squared_error: 1787.1558 - val_loss: 936891.9375 - val_root_mean_squared_error: 967.9318\n",
      "Epoch 230/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3186369.5000 - root_mean_squared_error: 1785.0405 - val_loss: 925047.1250 - val_root_mean_squared_error: 961.7937\n",
      "Epoch 231/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3176979.7500 - root_mean_squared_error: 1782.4084 - val_loss: 933177.9375 - val_root_mean_squared_error: 966.0114\n",
      "Epoch 232/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3179617.7500 - root_mean_squared_error: 1783.1483 - val_loss: 930685.6875 - val_root_mean_squared_error: 964.7205\n",
      "Epoch 233/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3165458.2500 - root_mean_squared_error: 1779.1735 - val_loss: 927410.0625 - val_root_mean_squared_error: 963.0213\n",
      "Epoch 234/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3166866.2500 - root_mean_squared_error: 1779.5691 - val_loss: 933000.8125 - val_root_mean_squared_error: 965.9197\n",
      "Epoch 235/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3161458.5000 - root_mean_squared_error: 1778.0491 - val_loss: 932417.1875 - val_root_mean_squared_error: 965.6175\n",
      "Epoch 236/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3146205.7500 - root_mean_squared_error: 1773.7548 - val_loss: 927188.2500 - val_root_mean_squared_error: 962.9061\n",
      "Epoch 237/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3139731.5000 - root_mean_squared_error: 1771.9287 - val_loss: 930045.2500 - val_root_mean_squared_error: 964.3885\n",
      "Epoch 238/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3131465.7500 - root_mean_squared_error: 1769.5948 - val_loss: 923936.6875 - val_root_mean_squared_error: 961.2162\n",
      "Epoch 239/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3147958.5000 - root_mean_squared_error: 1774.2487 - val_loss: 916500.3125 - val_root_mean_squared_error: 957.3402\n",
      "Epoch 240/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3122632.2500 - root_mean_squared_error: 1767.0972 - val_loss: 915182.7500 - val_root_mean_squared_error: 956.6519\n",
      "Epoch 241/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3127357.0000 - root_mean_squared_error: 1768.4335 - val_loss: 916846.6250 - val_root_mean_squared_error: 957.5211\n",
      "Epoch 242/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3116589.7500 - root_mean_squared_error: 1765.3866 - val_loss: 915941.1250 - val_root_mean_squared_error: 957.0482\n",
      "Epoch 243/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3103709.0000 - root_mean_squared_error: 1761.7346 - val_loss: 919755.6875 - val_root_mean_squared_error: 959.0389\n",
      "Epoch 244/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3106650.2500 - root_mean_squared_error: 1762.5692 - val_loss: 905122.3750 - val_root_mean_squared_error: 951.3792\n",
      "Epoch 245/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3114646.2500 - root_mean_squared_error: 1764.8361 - val_loss: 903315.0000 - val_root_mean_squared_error: 950.4288\n",
      "Epoch 246/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3088839.2500 - root_mean_squared_error: 1757.5094 - val_loss: 917131.2500 - val_root_mean_squared_error: 957.6697\n",
      "Epoch 247/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3094923.2500 - root_mean_squared_error: 1759.2394 - val_loss: 899595.8125 - val_root_mean_squared_error: 948.4703\n",
      "Epoch 248/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3090135.5000 - root_mean_squared_error: 1757.8782 - val_loss: 897267.5625 - val_root_mean_squared_error: 947.2421\n",
      "Epoch 249/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3078218.7500 - root_mean_squared_error: 1754.4854 - val_loss: 902583.6250 - val_root_mean_squared_error: 950.0440\n",
      "Epoch 250/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3082802.5000 - root_mean_squared_error: 1755.7911 - val_loss: 898214.6250 - val_root_mean_squared_error: 947.7419\n",
      "Epoch 251/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3066859.7500 - root_mean_squared_error: 1751.2452 - val_loss: 903273.1250 - val_root_mean_squared_error: 950.4068\n",
      "Epoch 252/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3072511.7500 - root_mean_squared_error: 1752.8582 - val_loss: 888637.8750 - val_root_mean_squared_error: 942.6759\n",
      "Epoch 253/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3054368.7500 - root_mean_squared_error: 1747.6753 - val_loss: 898650.7500 - val_root_mean_squared_error: 947.9719\n",
      "Epoch 254/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3054403.5000 - root_mean_squared_error: 1747.6852 - val_loss: 895534.3125 - val_root_mean_squared_error: 946.3267\n",
      "Epoch 255/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3037912.5000 - root_mean_squared_error: 1742.9608 - val_loss: 888206.1875 - val_root_mean_squared_error: 942.4469\n",
      "Epoch 256/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3035609.5000 - root_mean_squared_error: 1742.3000 - val_loss: 887626.0000 - val_root_mean_squared_error: 942.1390\n",
      "Epoch 257/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3033553.0000 - root_mean_squared_error: 1741.7098 - val_loss: 888897.6875 - val_root_mean_squared_error: 942.8137\n",
      "Epoch 258/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3023946.5000 - root_mean_squared_error: 1738.9498 - val_loss: 890413.0000 - val_root_mean_squared_error: 943.6170\n",
      "Epoch 259/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3024011.2500 - root_mean_squared_error: 1738.9684 - val_loss: 876664.3750 - val_root_mean_squared_error: 936.3036\n",
      "Epoch 260/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3006764.7500 - root_mean_squared_error: 1734.0026 - val_loss: 880396.6250 - val_root_mean_squared_error: 938.2946\n",
      "Epoch 261/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 3002892.0000 - root_mean_squared_error: 1732.8855 - val_loss: 881808.5625 - val_root_mean_squared_error: 939.0466\n",
      "Epoch 262/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3005184.7500 - root_mean_squared_error: 1733.5469 - val_loss: 874760.0000 - val_root_mean_squared_error: 935.2861\n",
      "Epoch 263/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3002877.0000 - root_mean_squared_error: 1732.8811 - val_loss: 881690.4375 - val_root_mean_squared_error: 938.9837\n",
      "Epoch 264/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3003758.2500 - root_mean_squared_error: 1733.1354 - val_loss: 857853.5625 - val_root_mean_squared_error: 926.2039\n",
      "Epoch 265/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 3002630.7500 - root_mean_squared_error: 1732.8101 - val_loss: 868433.6875 - val_root_mean_squared_error: 931.8979\n",
      "Epoch 266/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2975160.2500 - root_mean_squared_error: 1724.8652 - val_loss: 859037.3125 - val_root_mean_squared_error: 926.8427\n",
      "Epoch 267/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2971071.5000 - root_mean_squared_error: 1723.6797 - val_loss: 857636.5625 - val_root_mean_squared_error: 926.0867\n",
      "Epoch 268/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2969128.2500 - root_mean_squared_error: 1723.1158 - val_loss: 866335.3125 - val_root_mean_squared_error: 930.7714\n",
      "Epoch 269/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2958546.7500 - root_mean_squared_error: 1720.0426 - val_loss: 866131.1250 - val_root_mean_squared_error: 930.6617\n",
      "Epoch 270/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2953879.2500 - root_mean_squared_error: 1718.6853 - val_loss: 866599.1250 - val_root_mean_squared_error: 930.9131\n",
      "Epoch 271/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2957571.0000 - root_mean_squared_error: 1719.7590 - val_loss: 858210.4375 - val_root_mean_squared_error: 926.3965\n",
      "Epoch 272/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2948288.2500 - root_mean_squared_error: 1717.0580 - val_loss: 859306.5625 - val_root_mean_squared_error: 926.9879\n",
      "Epoch 273/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2949934.5000 - root_mean_squared_error: 1717.5374 - val_loss: 851874.0625 - val_root_mean_squared_error: 922.9702\n",
      "Epoch 274/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2959470.0000 - root_mean_squared_error: 1720.3110 - val_loss: 856027.7500 - val_root_mean_squared_error: 925.2177\n",
      "Epoch 275/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2929701.7500 - root_mean_squared_error: 1711.6372 - val_loss: 846327.1875 - val_root_mean_squared_error: 919.9604\n",
      "Epoch 276/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2926916.2500 - root_mean_squared_error: 1710.8232 - val_loss: 854894.9375 - val_root_mean_squared_error: 924.6053\n",
      "Epoch 277/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2912924.5000 - root_mean_squared_error: 1706.7291 - val_loss: 847262.2500 - val_root_mean_squared_error: 920.4685\n",
      "Epoch 278/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2922240.0000 - root_mean_squared_error: 1709.4561 - val_loss: 852450.3750 - val_root_mean_squared_error: 923.2824\n",
      "Epoch 279/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2883533.0000 - root_mean_squared_error: 1698.0969 - val_loss: 856385.5625 - val_root_mean_squared_error: 925.4110\n",
      "Epoch 280/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 2908763.2500 - root_mean_squared_error: 1705.5096 - val_loss: 845700.2500 - val_root_mean_squared_error: 919.6196\n",
      "Epoch 281/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2891480.7500 - root_mean_squared_error: 1700.4354 - val_loss: 859561.1875 - val_root_mean_squared_error: 927.1252\n",
      "Epoch 282/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2881373.5000 - root_mean_squared_error: 1697.4609 - val_loss: 854658.1250 - val_root_mean_squared_error: 924.4772\n",
      "Epoch 283/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2879591.2500 - root_mean_squared_error: 1696.9358 - val_loss: 845973.3125 - val_root_mean_squared_error: 919.7681\n",
      "Epoch 284/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2877626.2500 - root_mean_squared_error: 1696.3568 - val_loss: 849417.5000 - val_root_mean_squared_error: 921.6385\n",
      "Epoch 285/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2870011.5000 - root_mean_squared_error: 1694.1108 - val_loss: 834995.1875 - val_root_mean_squared_error: 913.7807\n",
      "Epoch 286/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2858554.0000 - root_mean_squared_error: 1690.7258 - val_loss: 839388.7500 - val_root_mean_squared_error: 916.1816\n",
      "Epoch 287/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2867234.5000 - root_mean_squared_error: 1693.2910 - val_loss: 843294.5000 - val_root_mean_squared_error: 918.3107\n",
      "Epoch 288/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2859070.0000 - root_mean_squared_error: 1690.8784 - val_loss: 845279.4375 - val_root_mean_squared_error: 919.3908\n",
      "Epoch 289/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2860461.0000 - root_mean_squared_error: 1691.2898 - val_loss: 839492.8750 - val_root_mean_squared_error: 916.2385\n",
      "Epoch 290/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2833712.5000 - root_mean_squared_error: 1683.3634 - val_loss: 838045.7500 - val_root_mean_squared_error: 915.4484\n",
      "Epoch 291/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2844574.2500 - root_mean_squared_error: 1686.5865 - val_loss: 821369.2500 - val_root_mean_squared_error: 906.2943\n",
      "Epoch 292/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2842977.0000 - root_mean_squared_error: 1686.1130 - val_loss: 817996.2500 - val_root_mean_squared_error: 904.4315\n",
      "Epoch 293/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2822260.0000 - root_mean_squared_error: 1679.9584 - val_loss: 819212.2500 - val_root_mean_squared_error: 905.1035\n",
      "Epoch 294/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2817816.5000 - root_mean_squared_error: 1678.6353 - val_loss: 830048.3125 - val_root_mean_squared_error: 911.0699\n",
      "Epoch 295/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2808870.5000 - root_mean_squared_error: 1675.9685 - val_loss: 810663.3125 - val_root_mean_squared_error: 900.3684\n",
      "Epoch 296/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2803631.0000 - root_mean_squared_error: 1674.4047 - val_loss: 816547.0625 - val_root_mean_squared_error: 903.6299\n",
      "Epoch 297/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2787798.2500 - root_mean_squared_error: 1669.6700 - val_loss: 813332.0000 - val_root_mean_squared_error: 901.8492\n",
      "Epoch 298/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2797949.7500 - root_mean_squared_error: 1672.7073 - val_loss: 814480.9375 - val_root_mean_squared_error: 902.4860\n",
      "Epoch 299/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2797620.0000 - root_mean_squared_error: 1672.6088 - val_loss: 804763.1875 - val_root_mean_squared_error: 897.0859\n",
      "Epoch 300/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2787051.2500 - root_mean_squared_error: 1669.4464 - val_loss: 812960.1250 - val_root_mean_squared_error: 901.6430\n",
      "Epoch 301/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2768933.2500 - root_mean_squared_error: 1664.0112 - val_loss: 812615.5625 - val_root_mean_squared_error: 901.4519\n",
      "Epoch 302/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2785979.2500 - root_mean_squared_error: 1669.1252 - val_loss: 797150.9375 - val_root_mean_squared_error: 892.8331\n",
      "Epoch 303/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2754983.2500 - root_mean_squared_error: 1659.8142 - val_loss: 815814.4375 - val_root_mean_squared_error: 903.2245\n",
      "Epoch 304/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2761580.7500 - root_mean_squared_error: 1661.8004 - val_loss: 802880.6875 - val_root_mean_squared_error: 896.0361\n",
      "Epoch 305/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 2769104.7500 - root_mean_squared_error: 1664.0627 - val_loss: 809308.4375 - val_root_mean_squared_error: 899.6157\n",
      "Epoch 306/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 2760748.7500 - root_mean_squared_error: 1661.5500 - val_loss: 809622.8125 - val_root_mean_squared_error: 899.7904\n",
      "Epoch 307/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2753067.2500 - root_mean_squared_error: 1659.2369 - val_loss: 794990.3125 - val_root_mean_squared_error: 891.6223\n",
      "Epoch 308/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2733525.2500 - root_mean_squared_error: 1653.3376 - val_loss: 804437.7500 - val_root_mean_squared_error: 896.9045\n",
      "Epoch 309/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2731254.0000 - root_mean_squared_error: 1652.6506 - val_loss: 791297.3750 - val_root_mean_squared_error: 889.5490\n",
      "Epoch 310/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2731167.2500 - root_mean_squared_error: 1652.6244 - val_loss: 796746.6250 - val_root_mean_squared_error: 892.6066\n",
      "Epoch 311/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2714363.7500 - root_mean_squared_error: 1647.5326 - val_loss: 792377.0625 - val_root_mean_squared_error: 890.1556\n",
      "Epoch 312/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2709369.5000 - root_mean_squared_error: 1646.0162 - val_loss: 785960.6875 - val_root_mean_squared_error: 886.5443\n",
      "Epoch 313/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2706403.5000 - root_mean_squared_error: 1645.1150 - val_loss: 789772.1250 - val_root_mean_squared_error: 888.6912\n",
      "Epoch 314/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2701774.2500 - root_mean_squared_error: 1643.7075 - val_loss: 785664.8750 - val_root_mean_squared_error: 886.3774\n",
      "Epoch 315/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2688588.2500 - root_mean_squared_error: 1639.6915 - val_loss: 786038.6875 - val_root_mean_squared_error: 886.5883\n",
      "Epoch 316/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2684905.2500 - root_mean_squared_error: 1638.5680 - val_loss: 771504.8750 - val_root_mean_squared_error: 878.3535\n",
      "Epoch 317/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2682120.7500 - root_mean_squared_error: 1637.7181 - val_loss: 779880.6875 - val_root_mean_squared_error: 883.1085\n",
      "Epoch 318/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2673765.5000 - root_mean_squared_error: 1635.1653 - val_loss: 775317.1250 - val_root_mean_squared_error: 880.5209\n",
      "Epoch 319/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2688514.0000 - root_mean_squared_error: 1639.6688 - val_loss: 769029.6250 - val_root_mean_squared_error: 876.9434\n",
      "Epoch 320/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2672680.0000 - root_mean_squared_error: 1634.8334 - val_loss: 773532.5625 - val_root_mean_squared_error: 879.5070\n",
      "Epoch 321/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2646669.7500 - root_mean_squared_error: 1626.8589 - val_loss: 766332.8125 - val_root_mean_squared_error: 875.4044\n",
      "Epoch 322/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2642849.0000 - root_mean_squared_error: 1625.6842 - val_loss: 773125.7500 - val_root_mean_squared_error: 879.2757\n",
      "Epoch 323/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 2637170.7500 - root_mean_squared_error: 1623.9368 - val_loss: 772808.1875 - val_root_mean_squared_error: 879.0951\n",
      "Epoch 324/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2631108.5000 - root_mean_squared_error: 1622.0692 - val_loss: 770516.5625 - val_root_mean_squared_error: 877.7907\n",
      "Epoch 325/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2635457.5000 - root_mean_squared_error: 1623.4092 - val_loss: 769377.1250 - val_root_mean_squared_error: 877.1415\n",
      "Epoch 326/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2627780.0000 - root_mean_squared_error: 1621.0428 - val_loss: 764750.5625 - val_root_mean_squared_error: 874.5002\n",
      "Epoch 327/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2632283.5000 - root_mean_squared_error: 1622.4314 - val_loss: 751283.5000 - val_root_mean_squared_error: 866.7661\n",
      "Epoch 328/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2630507.7500 - root_mean_squared_error: 1621.8840 - val_loss: 766348.1875 - val_root_mean_squared_error: 875.4131\n",
      "Epoch 329/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 2542014.5000 - root_mean_squared_error: 1594.36 - 0s 5ms/step - loss: 2622575.5000 - root_mean_squared_error: 1619.4368 - val_loss: 759995.9375 - val_root_mean_squared_error: 871.7775\n",
      "Epoch 330/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2613387.5000 - root_mean_squared_error: 1616.5975 - val_loss: 751984.1875 - val_root_mean_squared_error: 867.1702\n",
      "Epoch 331/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2598753.0000 - root_mean_squared_error: 1612.0648 - val_loss: 745835.8750 - val_root_mean_squared_error: 863.6179\n",
      "Epoch 332/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2586867.2500 - root_mean_squared_error: 1608.3741 - val_loss: 751040.0000 - val_root_mean_squared_error: 866.6256\n",
      "Epoch 333/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2585889.0000 - root_mean_squared_error: 1608.0699 - val_loss: 735867.4375 - val_root_mean_squared_error: 857.8271\n",
      "Epoch 334/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2575055.5000 - root_mean_squared_error: 1604.6980 - val_loss: 754008.8750 - val_root_mean_squared_error: 868.3369\n",
      "Epoch 335/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2577204.7500 - root_mean_squared_error: 1605.3674 - val_loss: 749383.5000 - val_root_mean_squared_error: 865.6694\n",
      "Epoch 336/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2577471.2500 - root_mean_squared_error: 1605.4504 - val_loss: 739933.1875 - val_root_mean_squared_error: 860.1937\n",
      "Epoch 337/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2555696.5000 - root_mean_squared_error: 1598.6545 - val_loss: 752888.8125 - val_root_mean_squared_error: 867.6917\n",
      "Epoch 338/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2568478.7500 - root_mean_squared_error: 1602.6475 - val_loss: 746292.3125 - val_root_mean_squared_error: 863.8821\n",
      "Epoch 339/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2554179.7500 - root_mean_squared_error: 1598.1802 - val_loss: 735059.5625 - val_root_mean_squared_error: 857.3561\n",
      "Epoch 340/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2538695.0000 - root_mean_squared_error: 1593.3282 - val_loss: 730002.1250 - val_root_mean_squared_error: 854.4016\n",
      "Epoch 341/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 2527589.0000 - root_mean_squared_error: 1589.8394 - val_loss: 731284.7500 - val_root_mean_squared_error: 855.1519\n",
      "Epoch 342/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2535656.5000 - root_mean_squared_error: 1592.3745 - val_loss: 736998.5625 - val_root_mean_squared_error: 858.4862\n",
      "Epoch 343/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2509326.5000 - root_mean_squared_error: 1584.0853 - val_loss: 718704.6875 - val_root_mean_squared_error: 847.7645\n",
      "Epoch 344/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2517043.7500 - root_mean_squared_error: 1586.5194 - val_loss: 734216.7500 - val_root_mean_squared_error: 856.8645\n",
      "Epoch 345/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2518172.0000 - root_mean_squared_error: 1586.8749 - val_loss: 736800.5625 - val_root_mean_squared_error: 858.3708\n",
      "Epoch 346/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2487588.0000 - root_mean_squared_error: 1577.2090 - val_loss: 715624.3750 - val_root_mean_squared_error: 845.9459\n",
      "Epoch 347/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2478235.5000 - root_mean_squared_error: 1574.2412 - val_loss: 713607.3750 - val_root_mean_squared_error: 844.7529\n",
      "Epoch 348/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2500022.7500 - root_mean_squared_error: 1581.1460 - val_loss: 716119.1875 - val_root_mean_squared_error: 846.2383\n",
      "Epoch 349/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2477667.2500 - root_mean_squared_error: 1574.0608 - val_loss: 719349.3750 - val_root_mean_squared_error: 848.1447\n",
      "Epoch 350/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2483541.5000 - root_mean_squared_error: 1575.9257 - val_loss: 710194.8750 - val_root_mean_squared_error: 842.7306\n",
      "Epoch 351/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2487253.7500 - root_mean_squared_error: 1577.1029 - val_loss: 704415.7500 - val_root_mean_squared_error: 839.2948\n",
      "Epoch 352/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2474348.0000 - root_mean_squared_error: 1573.0060 - val_loss: 716216.6250 - val_root_mean_squared_error: 846.2958\n",
      "Epoch 353/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2474522.7500 - root_mean_squared_error: 1573.0616 - val_loss: 702613.5625 - val_root_mean_squared_error: 838.2205\n",
      "Epoch 354/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2458344.7500 - root_mean_squared_error: 1567.9110 - val_loss: 698207.3750 - val_root_mean_squared_error: 835.5880\n",
      "Epoch 355/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2446193.2500 - root_mean_squared_error: 1564.0311 - val_loss: 694849.1250 - val_root_mean_squared_error: 833.5761\n",
      "Epoch 356/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2451460.7500 - root_mean_squared_error: 1565.7141 - val_loss: 692333.7500 - val_root_mean_squared_error: 832.0660\n",
      "Epoch 357/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2438503.7500 - root_mean_squared_error: 1561.5709 - val_loss: 703554.1250 - val_root_mean_squared_error: 838.7813\n",
      "Epoch 358/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2448317.5000 - root_mean_squared_error: 1564.7101 - val_loss: 694169.5625 - val_root_mean_squared_error: 833.1684\n",
      "Epoch 359/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2418748.5000 - root_mean_squared_error: 1555.2327 - val_loss: 699044.0000 - val_root_mean_squared_error: 836.0885\n",
      "Epoch 360/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2433291.7500 - root_mean_squared_error: 1559.9012 - val_loss: 700821.4375 - val_root_mean_squared_error: 837.1508\n",
      "Epoch 361/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2423211.0000 - root_mean_squared_error: 1556.6666 - val_loss: 688648.1250 - val_root_mean_squared_error: 829.8483\n",
      "Epoch 362/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2424732.0000 - root_mean_squared_error: 1557.1552 - val_loss: 685679.0000 - val_root_mean_squared_error: 828.0574\n",
      "Epoch 363/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2413071.5000 - root_mean_squared_error: 1553.4064 - val_loss: 697627.6875 - val_root_mean_squared_error: 835.2411\n",
      "Epoch 364/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2389606.5000 - root_mean_squared_error: 1545.8352 - val_loss: 693790.9375 - val_root_mean_squared_error: 832.9412\n",
      "Epoch 365/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2391297.7500 - root_mean_squared_error: 1546.3822 - val_loss: 687641.8750 - val_root_mean_squared_error: 829.2418\n",
      "Epoch 366/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2397884.5000 - root_mean_squared_error: 1548.5104 - val_loss: 683195.7500 - val_root_mean_squared_error: 826.5566\n",
      "Epoch 367/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2377855.0000 - root_mean_squared_error: 1542.0295 - val_loss: 682238.7500 - val_root_mean_squared_error: 825.9775\n",
      "Epoch 368/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2342578.0000 - root_mean_squared_error: 1530.5482 - val_loss: 679744.7500 - val_root_mean_squared_error: 824.4664\n",
      "Epoch 369/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2368605.2500 - root_mean_squared_error: 1539.0273 - val_loss: 692765.0625 - val_root_mean_squared_error: 832.3251\n",
      "Epoch 370/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2363087.0000 - root_mean_squared_error: 1537.2335 - val_loss: 664090.0625 - val_root_mean_squared_error: 814.9172\n",
      "Epoch 371/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 2359753.2500 - root_mean_squared_error: 1536.1488 - val_loss: 685612.4375 - val_root_mean_squared_error: 828.0172\n",
      "Epoch 372/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2351440.5000 - root_mean_squared_error: 1533.4408 - val_loss: 656179.2500 - val_root_mean_squared_error: 810.0489\n",
      "Epoch 373/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2323106.0000 - root_mean_squared_error: 1524.1738 - val_loss: 672477.7500 - val_root_mean_squared_error: 820.0474\n",
      "Epoch 374/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2319363.7500 - root_mean_squared_error: 1522.9458 - val_loss: 664448.7500 - val_root_mean_squared_error: 815.1373\n",
      "Epoch 375/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2334575.0000 - root_mean_squared_error: 1527.9316 - val_loss: 660145.8125 - val_root_mean_squared_error: 812.4936\n",
      "Epoch 376/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2325788.2500 - root_mean_squared_error: 1525.0535 - val_loss: 648829.5000 - val_root_mean_squared_error: 805.4995\n",
      "Epoch 377/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2330444.5000 - root_mean_squared_error: 1526.5793 - val_loss: 660324.5625 - val_root_mean_squared_error: 812.6036\n",
      "Epoch 378/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2316435.0000 - root_mean_squared_error: 1521.9839 - val_loss: 652521.9375 - val_root_mean_squared_error: 807.7883\n",
      "Epoch 379/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2290904.7500 - root_mean_squared_error: 1513.5735 - val_loss: 662480.8125 - val_root_mean_squared_error: 813.9293\n",
      "Epoch 380/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2311616.0000 - root_mean_squared_error: 1520.3999 - val_loss: 667547.8750 - val_root_mean_squared_error: 817.0360\n",
      "Epoch 381/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2294965.0000 - root_mean_squared_error: 1514.9142 - val_loss: 663656.0625 - val_root_mean_squared_error: 814.6509\n",
      "Epoch 382/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2276949.0000 - root_mean_squared_error: 1508.9563 - val_loss: 672935.8125 - val_root_mean_squared_error: 820.3267\n",
      "Epoch 383/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2276258.7500 - root_mean_squared_error: 1508.7275 - val_loss: 661539.7500 - val_root_mean_squared_error: 813.3510\n",
      "Epoch 384/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2269056.5000 - root_mean_squared_error: 1506.3387 - val_loss: 659916.6875 - val_root_mean_squared_error: 812.3525\n",
      "Epoch 385/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2270829.5000 - root_mean_squared_error: 1506.9271 - val_loss: 653132.5000 - val_root_mean_squared_error: 808.1661\n",
      "Epoch 386/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2268742.0000 - root_mean_squared_error: 1506.2344 - val_loss: 641523.1250 - val_root_mean_squared_error: 800.9514\n",
      "Epoch 387/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2251228.5000 - root_mean_squared_error: 1500.4094 - val_loss: 652279.8750 - val_root_mean_squared_error: 807.6384\n",
      "Epoch 388/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2248817.0000 - root_mean_squared_error: 1499.6056 - val_loss: 634039.6250 - val_root_mean_squared_error: 796.2661\n",
      "Epoch 389/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2253325.7500 - root_mean_squared_error: 1501.1082 - val_loss: 644367.8125 - val_root_mean_squared_error: 802.7252\n",
      "Epoch 390/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2257365.0000 - root_mean_squared_error: 1502.4530 - val_loss: 640724.4375 - val_root_mean_squared_error: 800.4526\n",
      "Epoch 391/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2237377.2500 - root_mean_squared_error: 1495.7865 - val_loss: 646056.5625 - val_root_mean_squared_error: 803.7764\n",
      "Epoch 392/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 2238476.0000 - root_mean_squared_error: 1496.1537 - val_loss: 644578.5000 - val_root_mean_squared_error: 802.8564\n",
      "Epoch 393/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2207478.7500 - root_mean_squared_error: 1485.7587 - val_loss: 628413.5000 - val_root_mean_squared_error: 792.7253\n",
      "Epoch 394/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2210792.5000 - root_mean_squared_error: 1486.8734 - val_loss: 631362.1250 - val_root_mean_squared_error: 794.5830\n",
      "Epoch 395/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2187253.7500 - root_mean_squared_error: 1478.9366 - val_loss: 631043.6250 - val_root_mean_squared_error: 794.3826\n",
      "Epoch 396/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2203956.7500 - root_mean_squared_error: 1484.5729 - val_loss: 618320.6875 - val_root_mean_squared_error: 786.3337\n",
      "Epoch 397/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2213349.5000 - root_mean_squared_error: 1487.7330 - val_loss: 620456.4375 - val_root_mean_squared_error: 787.6906\n",
      "Epoch 398/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2197729.0000 - root_mean_squared_error: 1482.4740 - val_loss: 608192.3125 - val_root_mean_squared_error: 779.8669\n",
      "Epoch 399/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2177753.0000 - root_mean_squared_error: 1475.7212 - val_loss: 610468.5000 - val_root_mean_squared_error: 781.3248\n",
      "Epoch 400/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2201128.7500 - root_mean_squared_error: 1483.6201 - val_loss: 625158.1875 - val_root_mean_squared_error: 790.6694\n",
      "Epoch 401/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2177509.0000 - root_mean_squared_error: 1475.6385 - val_loss: 605754.0000 - val_root_mean_squared_error: 778.3020\n",
      "Epoch 402/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2174456.7500 - root_mean_squared_error: 1474.6039 - val_loss: 619822.1875 - val_root_mean_squared_error: 787.2878\n",
      "Epoch 403/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2161273.2500 - root_mean_squared_error: 1470.1270 - val_loss: 605333.1250 - val_root_mean_squared_error: 778.0316\n",
      "Epoch 404/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2145100.0000 - root_mean_squared_error: 1464.6160 - val_loss: 601694.8750 - val_root_mean_squared_error: 775.6899\n",
      "Epoch 405/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2180533.5000 - root_mean_squared_error: 1476.6630 - val_loss: 602046.5625 - val_root_mean_squared_error: 775.9166\n",
      "Epoch 406/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2144323.0000 - root_mean_squared_error: 1464.3507 - val_loss: 612399.2500 - val_root_mean_squared_error: 782.5594\n",
      "Epoch 407/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2154374.7500 - root_mean_squared_error: 1467.7788 - val_loss: 599695.6875 - val_root_mean_squared_error: 774.4002\n",
      "Epoch 408/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2140782.7500 - root_mean_squared_error: 1463.1414 - val_loss: 594624.6875 - val_root_mean_squared_error: 771.1191\n",
      "Epoch 409/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2121846.2500 - root_mean_squared_error: 1456.6559 - val_loss: 598473.7500 - val_root_mean_squared_error: 773.6108\n",
      "Epoch 410/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2112846.5000 - root_mean_squared_error: 1453.5634 - val_loss: 602704.0000 - val_root_mean_squared_error: 776.3401\n",
      "Epoch 411/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2110935.7500 - root_mean_squared_error: 1452.9060 - val_loss: 598116.3750 - val_root_mean_squared_error: 773.3798\n",
      "Epoch 412/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 2099130.2500 - root_mean_squared_error: 1448.8375 - val_loss: 591617.8750 - val_root_mean_squared_error: 769.1670\n",
      "Epoch 413/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2110857.2500 - root_mean_squared_error: 1452.8789 - val_loss: 605553.4375 - val_root_mean_squared_error: 778.1732\n",
      "Epoch 414/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2104772.5000 - root_mean_squared_error: 1450.7834 - val_loss: 596841.6875 - val_root_mean_squared_error: 772.5553\n",
      "Epoch 415/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2091211.3750 - root_mean_squared_error: 1446.1022 - val_loss: 598194.6875 - val_root_mean_squared_error: 773.4305\n",
      "Epoch 416/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2069454.5000 - root_mean_squared_error: 1438.5598 - val_loss: 595772.6875 - val_root_mean_squared_error: 771.8631\n",
      "Epoch 417/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2078209.7500 - root_mean_squared_error: 1441.5997 - val_loss: 592654.3750 - val_root_mean_squared_error: 769.8405\n",
      "Epoch 418/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2072147.5000 - root_mean_squared_error: 1439.4956 - val_loss: 582219.2500 - val_root_mean_squared_error: 763.0330\n",
      "Epoch 419/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2077246.2500 - root_mean_squared_error: 1441.2655 - val_loss: 597384.9375 - val_root_mean_squared_error: 772.9068\n",
      "Epoch 420/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2044289.5000 - root_mean_squared_error: 1429.7865 - val_loss: 575562.5000 - val_root_mean_squared_error: 758.6583\n",
      "Epoch 421/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2043877.8750 - root_mean_squared_error: 1429.6426 - val_loss: 590614.1250 - val_root_mean_squared_error: 768.5142\n",
      "Epoch 422/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2054720.2500 - root_mean_squared_error: 1433.4296 - val_loss: 589829.6875 - val_root_mean_squared_error: 768.0037\n",
      "Epoch 423/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2056362.3750 - root_mean_squared_error: 1434.0022 - val_loss: 570060.2500 - val_root_mean_squared_error: 755.0233\n",
      "Epoch 424/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2031733.6250 - root_mean_squared_error: 1425.3889 - val_loss: 570735.6250 - val_root_mean_squared_error: 755.4705\n",
      "Epoch 425/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2032242.1250 - root_mean_squared_error: 1425.5673 - val_loss: 578860.1875 - val_root_mean_squared_error: 760.8286\n",
      "Epoch 426/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2036404.7500 - root_mean_squared_error: 1427.0265 - val_loss: 571416.6250 - val_root_mean_squared_error: 755.9210\n",
      "Epoch 427/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 2020028.7500 - root_mean_squared_error: 1421.2771 - val_loss: 572957.7500 - val_root_mean_squared_error: 756.9397\n",
      "Epoch 428/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 2008413.3750 - root_mean_squared_error: 1417.1851 - val_loss: 577406.7500 - val_root_mean_squared_error: 759.8729\n",
      "Epoch 429/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1995021.8750 - root_mean_squared_error: 1412.4524 - val_loss: 598908.3750 - val_root_mean_squared_error: 773.8917\n",
      "Epoch 430/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 2007371.8750 - root_mean_squared_error: 1416.8175 - val_loss: 567547.5000 - val_root_mean_squared_error: 753.3575\n",
      "Epoch 431/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1989516.0000 - root_mean_squared_error: 1410.5021 - val_loss: 570049.8750 - val_root_mean_squared_error: 755.0165\n",
      "Epoch 432/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1994905.0000 - root_mean_squared_error: 1412.4110 - val_loss: 568106.1875 - val_root_mean_squared_error: 753.7282\n",
      "Epoch 433/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1943083.7500 - root_mean_squared_error: 1393.9454 - val_loss: 565160.8750 - val_root_mean_squared_error: 751.7719\n",
      "Epoch 434/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1958645.1250 - root_mean_squared_error: 1399.5160 - val_loss: 581868.5625 - val_root_mean_squared_error: 762.8031\n",
      "Epoch 435/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1987582.2500 - root_mean_squared_error: 1409.8164 - val_loss: 567111.9375 - val_root_mean_squared_error: 753.0683\n",
      "Epoch 436/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1945841.3750 - root_mean_squared_error: 1394.9342 - val_loss: 568968.3750 - val_root_mean_squared_error: 754.2999\n",
      "Epoch 437/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1970452.7500 - root_mean_squared_error: 1403.7281 - val_loss: 559455.3750 - val_root_mean_squared_error: 747.9675\n",
      "Epoch 438/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1960844.6250 - root_mean_squared_error: 1400.3016 - val_loss: 563808.0000 - val_root_mean_squared_error: 750.8715\n",
      "Epoch 439/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1951854.6250 - root_mean_squared_error: 1397.0879 - val_loss: 554987.5000 - val_root_mean_squared_error: 744.9749\n",
      "Epoch 440/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1938868.7500 - root_mean_squared_error: 1392.4326 - val_loss: 555303.3750 - val_root_mean_squared_error: 745.1868\n",
      "Epoch 441/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1934407.6250 - root_mean_squared_error: 1390.8298 - val_loss: 546627.3750 - val_root_mean_squared_error: 739.3425\n",
      "Epoch 442/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1941460.3750 - root_mean_squared_error: 1393.3629 - val_loss: 560103.1250 - val_root_mean_squared_error: 748.4004\n",
      "Epoch 443/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1914857.7500 - root_mean_squared_error: 1383.7838 - val_loss: 547731.5625 - val_root_mean_squared_error: 740.0889\n",
      "Epoch 444/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 1884012.2500 - root_mean_squared_error: 1372.5933 - val_loss: 553706.3125 - val_root_mean_squared_error: 744.1144\n",
      "Epoch 445/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1907127.7500 - root_mean_squared_error: 1380.9879 - val_loss: 578357.3750 - val_root_mean_squared_error: 760.4981\n",
      "Epoch 446/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1915352.2500 - root_mean_squared_error: 1383.9625 - val_loss: 558570.3125 - val_root_mean_squared_error: 747.3756\n",
      "Epoch 447/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1889011.0000 - root_mean_squared_error: 1374.4130 - val_loss: 544267.6875 - val_root_mean_squared_error: 737.7450\n",
      "Epoch 448/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1900487.8750 - root_mean_squared_error: 1378.5818 - val_loss: 557127.8750 - val_root_mean_squared_error: 746.4100\n",
      "Epoch 449/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1903127.6250 - root_mean_squared_error: 1379.5389 - val_loss: 550756.3750 - val_root_mean_squared_error: 742.1296\n",
      "Epoch 450/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 2236388.5000 - root_mean_squared_error: 1495.45 - 0s 4ms/step - loss: 1898406.6250 - root_mean_squared_error: 1377.8268 - val_loss: 552922.3125 - val_root_mean_squared_error: 743.5875\n",
      "Epoch 451/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1872490.6250 - root_mean_squared_error: 1368.3898 - val_loss: 542882.6250 - val_root_mean_squared_error: 736.8057\n",
      "Epoch 452/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1865597.0000 - root_mean_squared_error: 1365.8685 - val_loss: 548432.5000 - val_root_mean_squared_error: 740.5623\n",
      "Epoch 453/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1856602.2500 - root_mean_squared_error: 1362.5719 - val_loss: 536231.1875 - val_root_mean_squared_error: 732.2781\n",
      "Epoch 454/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1833038.0000 - root_mean_squared_error: 1353.8973 - val_loss: 527998.5625 - val_root_mean_squared_error: 726.6351\n",
      "Epoch 455/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1856791.6250 - root_mean_squared_error: 1362.6414 - val_loss: 525336.6875 - val_root_mean_squared_error: 724.8011\n",
      "Epoch 456/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1847046.3750 - root_mean_squared_error: 1359.0608 - val_loss: 527400.3125 - val_root_mean_squared_error: 726.2233\n",
      "Epoch 457/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1850365.0000 - root_mean_squared_error: 1360.2812 - val_loss: 524096.5000 - val_root_mean_squared_error: 723.9451\n",
      "Epoch 458/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 1851301.5000 - root_mean_squared_error: 1360.6254 - val_loss: 522682.3750 - val_root_mean_squared_error: 722.9678\n",
      "Epoch 459/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1834966.8750 - root_mean_squared_error: 1354.6095 - val_loss: 530898.8750 - val_root_mean_squared_error: 728.6281\n",
      "Epoch 460/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1832312.6250 - root_mean_squared_error: 1353.6294 - val_loss: 540066.9375 - val_root_mean_squared_error: 734.8925\n",
      "Epoch 461/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1800055.3750 - root_mean_squared_error: 1341.6614 - val_loss: 521186.7188 - val_root_mean_squared_error: 721.9326\n",
      "Epoch 462/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1800348.3750 - root_mean_squared_error: 1341.7706 - val_loss: 525600.8125 - val_root_mean_squared_error: 724.9833\n",
      "Epoch 463/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1810216.8750 - root_mean_squared_error: 1345.4430 - val_loss: 530497.4375 - val_root_mean_squared_error: 728.3525\n",
      "Epoch 464/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1794038.7500 - root_mean_squared_error: 1339.4174 - val_loss: 525617.3750 - val_root_mean_squared_error: 724.9948\n",
      "Epoch 465/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1801402.8750 - root_mean_squared_error: 1342.1635 - val_loss: 503112.2812 - val_root_mean_squared_error: 709.3041\n",
      "Epoch 466/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1783298.5000 - root_mean_squared_error: 1335.4020 - val_loss: 513899.0625 - val_root_mean_squared_error: 716.8676\n",
      "Epoch 467/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1781178.7500 - root_mean_squared_error: 1334.6080 - val_loss: 523529.3438 - val_root_mean_squared_error: 723.5533\n",
      "Epoch 468/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1797498.0000 - root_mean_squared_error: 1340.7080 - val_loss: 512041.7812 - val_root_mean_squared_error: 715.5709\n",
      "Epoch 469/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1769416.3750 - root_mean_squared_error: 1330.1941 - val_loss: 503371.3750 - val_root_mean_squared_error: 709.4867\n",
      "Epoch 470/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1752542.8750 - root_mean_squared_error: 1323.8364 - val_loss: 506016.5938 - val_root_mean_squared_error: 711.3484\n",
      "Epoch 471/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1759158.8750 - root_mean_squared_error: 1326.3329 - val_loss: 500154.4688 - val_root_mean_squared_error: 707.2160\n",
      "Epoch 472/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1751070.5000 - root_mean_squared_error: 1323.2802 - val_loss: 482612.8438 - val_root_mean_squared_error: 694.7034\n",
      "Epoch 473/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1754496.6250 - root_mean_squared_error: 1324.5741 - val_loss: 493129.0625 - val_root_mean_squared_error: 702.2315\n",
      "Epoch 474/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1738938.2500 - root_mean_squared_error: 1318.6881 - val_loss: 503959.0938 - val_root_mean_squared_error: 709.9008\n",
      "Epoch 475/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1719514.8750 - root_mean_squared_error: 1311.3027 - val_loss: 494375.4688 - val_root_mean_squared_error: 703.1184\n",
      "Epoch 476/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1733872.1250 - root_mean_squared_error: 1316.7657 - val_loss: 497570.5938 - val_root_mean_squared_error: 705.3868\n",
      "Epoch 477/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1721127.0000 - root_mean_squared_error: 1311.9174 - val_loss: 503415.7188 - val_root_mean_squared_error: 709.5179\n",
      "Epoch 478/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1705292.2500 - root_mean_squared_error: 1305.8684 - val_loss: 507777.0000 - val_root_mean_squared_error: 712.5847\n",
      "Epoch 479/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1715642.8750 - root_mean_squared_error: 1309.8256 - val_loss: 502999.8750 - val_root_mean_squared_error: 709.2249\n",
      "Epoch 480/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1702720.3750 - root_mean_squared_error: 1304.8833 - val_loss: 492771.0312 - val_root_mean_squared_error: 701.9765\n",
      "Epoch 481/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1709061.5000 - root_mean_squared_error: 1307.3108 - val_loss: 520815.8750 - val_root_mean_squared_error: 721.6757\n",
      "Epoch 482/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1667215.2500 - root_mean_squared_error: 1291.2069 - val_loss: 502765.1562 - val_root_mean_squared_error: 709.0593\n",
      "Epoch 483/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1667076.2500 - root_mean_squared_error: 1291.1531 - val_loss: 498579.5625 - val_root_mean_squared_error: 706.1017\n",
      "Epoch 484/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1719297.7500 - root_mean_squared_error: 1311.2200 - val_loss: 478406.8125 - val_root_mean_squared_error: 691.6696\n",
      "Epoch 485/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1685360.7500 - root_mean_squared_error: 1298.2145 - val_loss: 490469.8125 - val_root_mean_squared_error: 700.3355\n",
      "Epoch 486/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1673764.8750 - root_mean_squared_error: 1293.7406 - val_loss: 499220.6562 - val_root_mean_squared_error: 706.5555\n",
      "Epoch 487/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1647006.1250 - root_mean_squared_error: 1283.3574 - val_loss: 492483.9688 - val_root_mean_squared_error: 701.7720\n",
      "Epoch 488/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1659250.8750 - root_mean_squared_error: 1288.1191 - val_loss: 496852.4375 - val_root_mean_squared_error: 704.8776\n",
      "Epoch 489/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1669967.7500 - root_mean_squared_error: 1292.2723 - val_loss: 496795.7500 - val_root_mean_squared_error: 704.8374\n",
      "Epoch 490/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1670754.1250 - root_mean_squared_error: 1292.5765 - val_loss: 486069.5625 - val_root_mean_squared_error: 697.1869\n",
      "Epoch 491/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1658264.3750 - root_mean_squared_error: 1287.7361 - val_loss: 480125.8438 - val_root_mean_squared_error: 692.9111\n",
      "Epoch 492/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1653368.7500 - root_mean_squared_error: 1285.8339 - val_loss: 470527.5625 - val_root_mean_squared_error: 685.9501\n",
      "Epoch 493/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1630491.7500 - root_mean_squared_error: 1276.9071 - val_loss: 461290.1562 - val_root_mean_squared_error: 679.1835\n",
      "Epoch 494/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1628999.3750 - root_mean_squared_error: 1276.3226 - val_loss: 479166.8438 - val_root_mean_squared_error: 692.2188\n",
      "Epoch 495/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1653240.6250 - root_mean_squared_error: 1285.7841 - val_loss: 462254.4688 - val_root_mean_squared_error: 679.8930\n",
      "Epoch 496/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1610772.0000 - root_mean_squared_error: 1269.1620 - val_loss: 473952.7188 - val_root_mean_squared_error: 688.4423\n",
      "Epoch 497/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 1619917.6250 - root_mean_squared_error: 1272.7599 - val_loss: 477115.4688 - val_root_mean_squared_error: 690.7355\n",
      "Epoch 498/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1608995.3750 - root_mean_squared_error: 1268.4618 - val_loss: 471044.1875 - val_root_mean_squared_error: 686.3266\n",
      "Epoch 499/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1619253.5000 - root_mean_squared_error: 1272.4989 - val_loss: 477236.9375 - val_root_mean_squared_error: 690.8234\n",
      "Epoch 500/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1593921.5000 - root_mean_squared_error: 1262.5060 - val_loss: 467211.5000 - val_root_mean_squared_error: 683.5287\n",
      "Epoch 501/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1579558.2500 - root_mean_squared_error: 1256.8048 - val_loss: 476421.1875 - val_root_mean_squared_error: 690.2327\n",
      "Epoch 502/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1622839.3750 - root_mean_squared_error: 1273.9071 - val_loss: 461558.3438 - val_root_mean_squared_error: 679.3809\n",
      "Epoch 503/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1564843.6250 - root_mean_squared_error: 1250.9371 - val_loss: 482875.5938 - val_root_mean_squared_error: 694.8925\n",
      "Epoch 504/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1590445.1250 - root_mean_squared_error: 1261.1285 - val_loss: 462609.8125 - val_root_mean_squared_error: 680.1542\n",
      "Epoch 505/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1571274.8750 - root_mean_squared_error: 1253.5050 - val_loss: 473349.1875 - val_root_mean_squared_error: 688.0038\n",
      "Epoch 506/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1578782.8750 - root_mean_squared_error: 1256.4962 - val_loss: 454058.7812 - val_root_mean_squared_error: 673.8389\n",
      "Epoch 507/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1569298.8750 - root_mean_squared_error: 1252.7166 - val_loss: 459522.8750 - val_root_mean_squared_error: 677.8812\n",
      "Epoch 508/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1543908.2500 - root_mean_squared_error: 1242.5410 - val_loss: 465148.0312 - val_root_mean_squared_error: 682.0176\n",
      "Epoch 509/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1532826.7500 - root_mean_squared_error: 1238.0739 - val_loss: 453875.2812 - val_root_mean_squared_error: 673.7026\n",
      "Epoch 510/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1545718.1250 - root_mean_squared_error: 1243.2692 - val_loss: 445714.7188 - val_root_mean_squared_error: 667.6187\n",
      "Epoch 511/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1581190.0000 - root_mean_squared_error: 1257.4537 - val_loss: 449752.2500 - val_root_mean_squared_error: 670.6357\n",
      "Epoch 512/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1564279.5000 - root_mean_squared_error: 1250.7115 - val_loss: 452004.8750 - val_root_mean_squared_error: 672.3130\n",
      "Epoch 513/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1539756.5000 - root_mean_squared_error: 1240.8693 - val_loss: 448904.1250 - val_root_mean_squared_error: 670.0031\n",
      "Epoch 514/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1492669.5000 - root_mean_squared_error: 1221.7485 - val_loss: 453165.0625 - val_root_mean_squared_error: 673.1754\n",
      "Epoch 515/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1504026.7500 - root_mean_squared_error: 1226.3877 - val_loss: 448783.6875 - val_root_mean_squared_error: 669.9132\n",
      "Epoch 516/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1476470.3750 - root_mean_squared_error: 1215.1010 - val_loss: 444656.4688 - val_root_mean_squared_error: 666.8257\n",
      "Epoch 517/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1491006.8750 - root_mean_squared_error: 1221.0679 - val_loss: 444722.8750 - val_root_mean_squared_error: 666.8754\n",
      "Epoch 518/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1507283.2500 - root_mean_squared_error: 1227.7146 - val_loss: 448917.1562 - val_root_mean_squared_error: 670.0128\n",
      "Epoch 519/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1516115.3750 - root_mean_squared_error: 1231.3064 - val_loss: 440122.3125 - val_root_mean_squared_error: 663.4172\n",
      "Epoch 520/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1492980.8750 - root_mean_squared_error: 1221.8760 - val_loss: 456225.5625 - val_root_mean_squared_error: 675.4447\n",
      "Epoch 521/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1475528.8750 - root_mean_squared_error: 1214.7135 - val_loss: 456652.0625 - val_root_mean_squared_error: 675.7604\n",
      "Epoch 522/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1459676.3750 - root_mean_squared_error: 1208.1707 - val_loss: 460396.5312 - val_root_mean_squared_error: 678.5253\n",
      "Epoch 523/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1473040.1250 - root_mean_squared_error: 1213.6886 - val_loss: 445610.5625 - val_root_mean_squared_error: 667.5407\n",
      "Epoch 524/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1441282.6250 - root_mean_squared_error: 1200.5343 - val_loss: 448426.9375 - val_root_mean_squared_error: 669.6469\n",
      "Epoch 525/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1440404.6250 - root_mean_squared_error: 1200.1686 - val_loss: 456474.9375 - val_root_mean_squared_error: 675.6293\n",
      "Epoch 526/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1433230.3750 - root_mean_squared_error: 1197.1760 - val_loss: 436878.3125 - val_root_mean_squared_error: 660.9677\n",
      "Epoch 527/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1453608.5000 - root_mean_squared_error: 1205.6569 - val_loss: 430002.2812 - val_root_mean_squared_error: 655.7456\n",
      "Epoch 528/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1450085.2500 - root_mean_squared_error: 1204.1948 - val_loss: 447503.9062 - val_root_mean_squared_error: 668.9573\n",
      "Epoch 529/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1417161.2500 - root_mean_squared_error: 1190.4458 - val_loss: 460918.1562 - val_root_mean_squared_error: 678.9095\n",
      "Epoch 530/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1401438.1250 - root_mean_squared_error: 1183.8235 - val_loss: 451800.3125 - val_root_mean_squared_error: 672.1609\n",
      "Epoch 531/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 1432875.1250 - root_mean_squared_error: 1197.0276 - val_loss: 439002.3438 - val_root_mean_squared_error: 662.5725\n",
      "Epoch 532/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1427444.3750 - root_mean_squared_error: 1194.7571 - val_loss: 424125.9062 - val_root_mean_squared_error: 651.2495\n",
      "Epoch 533/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1430409.6250 - root_mean_squared_error: 1195.9973 - val_loss: 426683.5312 - val_root_mean_squared_error: 653.2101\n",
      "Epoch 534/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1428415.0000 - root_mean_squared_error: 1195.1632 - val_loss: 426849.8125 - val_root_mean_squared_error: 653.3375\n",
      "Epoch 535/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1399233.1250 - root_mean_squared_error: 1182.8918 - val_loss: 426828.5000 - val_root_mean_squared_error: 653.3211\n",
      "Epoch 536/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1410778.3750 - root_mean_squared_error: 1187.7620 - val_loss: 409820.6875 - val_root_mean_squared_error: 640.1724\n",
      "Epoch 537/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1402640.6250 - root_mean_squared_error: 1184.3313 - val_loss: 441480.5312 - val_root_mean_squared_error: 664.4400\n",
      "Epoch 538/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1403490.1250 - root_mean_squared_error: 1184.6899 - val_loss: 428954.6250 - val_root_mean_squared_error: 654.9463\n",
      "Epoch 539/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1383386.7500 - root_mean_squared_error: 1176.1747 - val_loss: 414816.3438 - val_root_mean_squared_error: 644.0624\n",
      "Epoch 540/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1368241.5000 - root_mean_squared_error: 1169.7185 - val_loss: 429459.1250 - val_root_mean_squared_error: 655.3313\n",
      "Epoch 541/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1377016.8750 - root_mean_squared_error: 1173.4636 - val_loss: 421811.0625 - val_root_mean_squared_error: 649.4698\n",
      "Epoch 542/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1356648.8750 - root_mean_squared_error: 1164.7527 - val_loss: 418895.5938 - val_root_mean_squared_error: 647.2214\n",
      "Epoch 543/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1363287.6250 - root_mean_squared_error: 1167.5991 - val_loss: 421027.2188 - val_root_mean_squared_error: 648.8661\n",
      "Epoch 544/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1358001.7500 - root_mean_squared_error: 1165.3334 - val_loss: 412769.8750 - val_root_mean_squared_error: 642.4717\n",
      "Epoch 545/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1387778.8750 - root_mean_squared_error: 1178.0403 - val_loss: 426797.0000 - val_root_mean_squared_error: 653.2970\n",
      "Epoch 546/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1352646.1250 - root_mean_squared_error: 1163.0332 - val_loss: 409242.6875 - val_root_mean_squared_error: 639.7208\n",
      "Epoch 547/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1376832.6250 - root_mean_squared_error: 1173.3851 - val_loss: 415394.4062 - val_root_mean_squared_error: 644.5110\n",
      "Epoch 548/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1334597.0000 - root_mean_squared_error: 1155.2476 - val_loss: 403394.0000 - val_root_mean_squared_error: 635.1331\n",
      "Epoch 549/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1336356.0000 - root_mean_squared_error: 1156.0087 - val_loss: 405838.6875 - val_root_mean_squared_error: 637.0547\n",
      "Epoch 550/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1329889.8750 - root_mean_squared_error: 1153.2085 - val_loss: 417443.1562 - val_root_mean_squared_error: 646.0984\n",
      "Epoch 551/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1343794.8750 - root_mean_squared_error: 1159.2217 - val_loss: 418943.8438 - val_root_mean_squared_error: 647.2587\n",
      "Epoch 552/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1325440.6250 - root_mean_squared_error: 1151.2778 - val_loss: 401099.0000 - val_root_mean_squared_error: 633.3238\n",
      "Epoch 553/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 1294060.2500 - root_mean_squared_error: 1137.5677 - val_loss: 405390.5312 - val_root_mean_squared_error: 636.7029\n",
      "Epoch 554/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1302715.8750 - root_mean_squared_error: 1141.3658 - val_loss: 397733.1875 - val_root_mean_squared_error: 630.6609\n",
      "Epoch 555/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1321611.8750 - root_mean_squared_error: 1149.6138 - val_loss: 389593.1875 - val_root_mean_squared_error: 624.1740\n",
      "Epoch 556/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1309691.5000 - root_mean_squared_error: 1144.4175 - val_loss: 394652.7812 - val_root_mean_squared_error: 628.2140\n",
      "Epoch 557/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1272619.2500 - root_mean_squared_error: 1128.1042 - val_loss: 394773.6250 - val_root_mean_squared_error: 628.3101\n",
      "Epoch 558/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1277814.0000 - root_mean_squared_error: 1130.4043 - val_loss: 376147.6875 - val_root_mean_squared_error: 613.3088\n",
      "Epoch 559/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 1181871.2500 - root_mean_squared_error: 1087.13 - 0s 5ms/step - loss: 1307596.7500 - root_mean_squared_error: 1143.5020 - val_loss: 376437.5625 - val_root_mean_squared_error: 613.5451\n",
      "Epoch 560/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1312418.3750 - root_mean_squared_error: 1145.6083 - val_loss: 384950.0625 - val_root_mean_squared_error: 620.4434\n",
      "Epoch 561/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1251085.2500 - root_mean_squared_error: 1118.5192 - val_loss: 377603.0000 - val_root_mean_squared_error: 614.4941\n",
      "Epoch 562/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1298384.6250 - root_mean_squared_error: 1139.4668 - val_loss: 371956.0312 - val_root_mean_squared_error: 609.8820\n",
      "Epoch 563/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1237418.7500 - root_mean_squared_error: 1112.3933 - val_loss: 387735.6562 - val_root_mean_squared_error: 622.6843\n",
      "Epoch 564/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1256520.5000 - root_mean_squared_error: 1120.9463 - val_loss: 381035.3438 - val_root_mean_squared_error: 617.2806\n",
      "Epoch 565/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1248399.2500 - root_mean_squared_error: 1117.3179 - val_loss: 388122.1562 - val_root_mean_squared_error: 622.9945\n",
      "Epoch 566/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1241660.5000 - root_mean_squared_error: 1114.2982 - val_loss: 384186.9062 - val_root_mean_squared_error: 619.8281\n",
      "Epoch 567/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1249952.3750 - root_mean_squared_error: 1118.0127 - val_loss: 385547.1250 - val_root_mean_squared_error: 620.9244\n",
      "Epoch 568/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1227763.1250 - root_mean_squared_error: 1108.0447 - val_loss: 381641.4375 - val_root_mean_squared_error: 617.7714\n",
      "Epoch 569/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1245062.6250 - root_mean_squared_error: 1115.8237 - val_loss: 375510.2812 - val_root_mean_squared_error: 612.7889\n",
      "Epoch 570/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1212630.0000 - root_mean_squared_error: 1101.1948 - val_loss: 377340.6250 - val_root_mean_squared_error: 614.2806\n",
      "Epoch 571/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1217028.7500 - root_mean_squared_error: 1103.1903 - val_loss: 380252.7500 - val_root_mean_squared_error: 616.6464\n",
      "Epoch 572/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1233079.7500 - root_mean_squared_error: 1110.4413 - val_loss: 393271.7188 - val_root_mean_squared_error: 627.1138\n",
      "Epoch 573/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1214528.6250 - root_mean_squared_error: 1102.0565 - val_loss: 378873.0938 - val_root_mean_squared_error: 615.5267\n",
      "Epoch 574/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1207111.2500 - root_mean_squared_error: 1098.6862 - val_loss: 385824.0625 - val_root_mean_squared_error: 621.1474\n",
      "Epoch 575/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1221191.8750 - root_mean_squared_error: 1105.0756 - val_loss: 376531.6250 - val_root_mean_squared_error: 613.6217\n",
      "Epoch 576/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1198162.2500 - root_mean_squared_error: 1094.6060 - val_loss: 367625.5000 - val_root_mean_squared_error: 606.3213\n",
      "Epoch 577/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1169011.2500 - root_mean_squared_error: 1081.2083 - val_loss: 370067.9062 - val_root_mean_squared_error: 608.3321\n",
      "Epoch 578/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1159092.1250 - root_mean_squared_error: 1076.6115 - val_loss: 374392.5312 - val_root_mean_squared_error: 611.8762\n",
      "Epoch 579/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1170698.1250 - root_mean_squared_error: 1081.9880 - val_loss: 383291.6875 - val_root_mean_squared_error: 619.1055\n",
      "Epoch 580/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1183426.7500 - root_mean_squared_error: 1087.8542 - val_loss: 389959.5000 - val_root_mean_squared_error: 624.4673\n",
      "Epoch 581/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1184360.5000 - root_mean_squared_error: 1088.2833 - val_loss: 377039.4062 - val_root_mean_squared_error: 614.0353\n",
      "Epoch 582/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1180305.2500 - root_mean_squared_error: 1086.4186 - val_loss: 363037.5312 - val_root_mean_squared_error: 602.5259\n",
      "Epoch 583/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1173348.7500 - root_mean_squared_error: 1083.2123 - val_loss: 369975.0000 - val_root_mean_squared_error: 608.2557\n",
      "Epoch 584/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1189634.8750 - root_mean_squared_error: 1090.7039 - val_loss: 362680.7500 - val_root_mean_squared_error: 602.2299\n",
      "Epoch 585/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1144701.8750 - root_mean_squared_error: 1069.9075 - val_loss: 368152.3438 - val_root_mean_squared_error: 606.7556\n",
      "Epoch 586/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1135671.6250 - root_mean_squared_error: 1065.6790 - val_loss: 366112.8750 - val_root_mean_squared_error: 605.0726\n",
      "Epoch 587/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1116813.0000 - root_mean_squared_error: 1056.7937 - val_loss: 362960.9062 - val_root_mean_squared_error: 602.4623\n",
      "Epoch 588/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1127078.7500 - root_mean_squared_error: 1061.6396 - val_loss: 369846.3438 - val_root_mean_squared_error: 608.1500\n",
      "Epoch 589/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1162555.1250 - root_mean_squared_error: 1078.2185 - val_loss: 372171.1562 - val_root_mean_squared_error: 610.0583\n",
      "Epoch 590/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1111185.5000 - root_mean_squared_error: 1054.1278 - val_loss: 371674.8750 - val_root_mean_squared_error: 609.6514\n",
      "Epoch 591/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1148806.0000 - root_mean_squared_error: 1071.8237 - val_loss: 364302.8125 - val_root_mean_squared_error: 603.5750\n",
      "Epoch 592/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1122831.0000 - root_mean_squared_error: 1059.6372 - val_loss: 380460.3438 - val_root_mean_squared_error: 616.8147\n",
      "Epoch 593/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1086230.6250 - root_mean_squared_error: 1042.2239 - val_loss: 373548.7812 - val_root_mean_squared_error: 611.1864\n",
      "Epoch 594/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1095033.3750 - root_mean_squared_error: 1046.4385 - val_loss: 358992.6875 - val_root_mean_squared_error: 599.1600\n",
      "Epoch 595/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1130823.0000 - root_mean_squared_error: 1063.4016 - val_loss: 359737.5938 - val_root_mean_squared_error: 599.7813\n",
      "Epoch 596/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1102757.7500 - root_mean_squared_error: 1050.1227 - val_loss: 343176.3750 - val_root_mean_squared_error: 585.8126\n",
      "Epoch 597/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1109429.6250 - root_mean_squared_error: 1053.2947 - val_loss: 333911.2812 - val_root_mean_squared_error: 577.8506\n",
      "Epoch 598/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1090369.3750 - root_mean_squared_error: 1044.2075 - val_loss: 341274.1875 - val_root_mean_squared_error: 584.1868\n",
      "Epoch 599/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1080413.6250 - root_mean_squared_error: 1039.4294 - val_loss: 339976.2812 - val_root_mean_squared_error: 583.0748\n",
      "Epoch 600/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1089197.8750 - root_mean_squared_error: 1043.6465 - val_loss: 324737.3750 - val_root_mean_squared_error: 569.8573\n",
      "Epoch 601/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1131823.2500 - root_mean_squared_error: 1063.8718 - val_loss: 325852.6875 - val_root_mean_squared_error: 570.8351\n",
      "Epoch 602/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 1082792.5000 - root_mean_squared_error: 1040.5731 - val_loss: 354722.1562 - val_root_mean_squared_error: 595.5856\n",
      "Epoch 603/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1055452.3750 - root_mean_squared_error: 1027.3522 - val_loss: 351229.9062 - val_root_mean_squared_error: 592.6465\n",
      "Epoch 604/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 1081465.7500 - root_mean_squared_error: 1039.9354 - val_loss: 355843.8125 - val_root_mean_squared_error: 596.5264\n",
      "Epoch 605/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1065641.7500 - root_mean_squared_error: 1032.2992 - val_loss: 353205.4375 - val_root_mean_squared_error: 594.3109\n",
      "Epoch 606/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1067009.3750 - root_mean_squared_error: 1032.9614 - val_loss: 359271.4688 - val_root_mean_squared_error: 599.3926\n",
      "Epoch 607/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1087051.7500 - root_mean_squared_error: 1042.6178 - val_loss: 342587.4062 - val_root_mean_squared_error: 585.3097\n",
      "Epoch 608/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1015307.1875 - root_mean_squared_error: 1007.6245 - val_loss: 341496.5312 - val_root_mean_squared_error: 584.3771\n",
      "Epoch 609/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1054752.2500 - root_mean_squared_error: 1027.0114 - val_loss: 340429.7188 - val_root_mean_squared_error: 583.4636\n",
      "Epoch 610/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1041630.4375 - root_mean_squared_error: 1020.6030 - val_loss: 337141.8750 - val_root_mean_squared_error: 580.6392\n",
      "Epoch 611/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1025920.6875 - root_mean_squared_error: 1012.8774 - val_loss: 329986.7500 - val_root_mean_squared_error: 574.4447\n",
      "Epoch 612/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 1031140.2500 - root_mean_squared_error: 1015.4507 - val_loss: 329018.8125 - val_root_mean_squared_error: 573.6016\n",
      "Epoch 613/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1006672.1250 - root_mean_squared_error: 1003.3305 - val_loss: 334637.5000 - val_root_mean_squared_error: 578.4786\n",
      "Epoch 614/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1006691.5625 - root_mean_squared_error: 1003.3402 - val_loss: 343275.3125 - val_root_mean_squared_error: 585.8970\n",
      "Epoch 615/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1022229.5000 - root_mean_squared_error: 1011.0536 - val_loss: 331599.1875 - val_root_mean_squared_error: 575.8465\n",
      "Epoch 616/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1024579.2500 - root_mean_squared_error: 1012.2150 - val_loss: 331945.6250 - val_root_mean_squared_error: 576.1472\n",
      "Epoch 617/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1020582.3125 - root_mean_squared_error: 1010.2388 - val_loss: 328812.0000 - val_root_mean_squared_error: 573.4213\n",
      "Epoch 618/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1005228.5000 - root_mean_squared_error: 1002.6108 - val_loss: 324932.7500 - val_root_mean_squared_error: 570.0287\n",
      "Epoch 619/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 1004073.8750 - root_mean_squared_error: 1002.0349 - val_loss: 325693.3438 - val_root_mean_squared_error: 570.6955\n",
      "Epoch 620/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 1018807.7500 - root_mean_squared_error: 1009.3600 - val_loss: 325626.3750 - val_root_mean_squared_error: 570.6368\n",
      "Epoch 621/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 985745.1875 - root_mean_squared_error: 992.8470 - val_loss: 325410.0312 - val_root_mean_squared_error: 570.4472\n",
      "Epoch 622/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 987106.4375 - root_mean_squared_error: 993.5323 - val_loss: 341150.5625 - val_root_mean_squared_error: 584.0809\n",
      "Epoch 623/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 973815.3750 - root_mean_squared_error: 986.8209 - val_loss: 337945.0312 - val_root_mean_squared_error: 581.3304\n",
      "Epoch 624/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 998151.2500 - root_mean_squared_error: 999.0752 - val_loss: 335069.3438 - val_root_mean_squared_error: 578.8517\n",
      "Epoch 625/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 979965.2500 - root_mean_squared_error: 989.9319 - val_loss: 335585.2812 - val_root_mean_squared_error: 579.2972\n",
      "Epoch 626/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 967688.1250 - root_mean_squared_error: 983.7114 - val_loss: 335808.8750 - val_root_mean_squared_error: 579.4902\n",
      "Epoch 627/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 970657.1250 - root_mean_squared_error: 985.2193 - val_loss: 333163.3438 - val_root_mean_squared_error: 577.2031\n",
      "Epoch 628/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 961440.0000 - root_mean_squared_error: 980.5305 - val_loss: 320735.3438 - val_root_mean_squared_error: 566.3350\n",
      "Epoch 629/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 986122.5625 - root_mean_squared_error: 993.0370 - val_loss: 324659.5000 - val_root_mean_squared_error: 569.7890\n",
      "Epoch 630/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 963203.2500 - root_mean_squared_error: 981.4292 - val_loss: 325654.9688 - val_root_mean_squared_error: 570.6619\n",
      "Epoch 631/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 963067.8750 - root_mean_squared_error: 981.3602 - val_loss: 320230.7500 - val_root_mean_squared_error: 565.8893\n",
      "Epoch 632/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 950220.9375 - root_mean_squared_error: 974.7928 - val_loss: 320468.1875 - val_root_mean_squared_error: 566.0991\n",
      "Epoch 633/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 925000.6250 - root_mean_squared_error: 961.7695 - val_loss: 313302.9062 - val_root_mean_squared_error: 559.7347\n",
      "Epoch 634/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 943839.2500 - root_mean_squared_error: 971.5139 - val_loss: 308608.7188 - val_root_mean_squared_error: 555.5256\n",
      "Epoch 635/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 948870.8125 - root_mean_squared_error: 974.1000 - val_loss: 311971.9688 - val_root_mean_squared_error: 558.5445\n",
      "Epoch 636/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 895925.1875 - root_mean_squared_error: 946.5333 - val_loss: 312571.2500 - val_root_mean_squared_error: 559.0807\n",
      "Epoch 637/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 901284.9375 - root_mean_squared_error: 949.3603 - val_loss: 301637.1250 - val_root_mean_squared_error: 549.2150\n",
      "Epoch 638/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 927732.7500 - root_mean_squared_error: 963.1888 - val_loss: 300494.2188 - val_root_mean_squared_error: 548.1735\n",
      "Epoch 639/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 916881.3125 - root_mean_squared_error: 957.5392 - val_loss: 294694.0938 - val_root_mean_squared_error: 542.8574\n",
      "Epoch 640/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 882324.4375 - root_mean_squared_error: 939.3213 - val_loss: 301268.1250 - val_root_mean_squared_error: 548.8790\n",
      "Epoch 641/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 909393.0625 - root_mean_squared_error: 953.6210 - val_loss: 312975.1875 - val_root_mean_squared_error: 559.4418\n",
      "Epoch 642/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 905964.6875 - root_mean_squared_error: 951.8218 - val_loss: 317719.7500 - val_root_mean_squared_error: 563.6663\n",
      "Epoch 643/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 907509.6250 - root_mean_squared_error: 952.6330 - val_loss: 307112.7500 - val_root_mean_squared_error: 554.1776\n",
      "Epoch 644/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 897508.6250 - root_mean_squared_error: 947.3693 - val_loss: 314747.7188 - val_root_mean_squared_error: 561.0238\n",
      "Epoch 645/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 876319.1250 - root_mean_squared_error: 936.1192 - val_loss: 316069.5625 - val_root_mean_squared_error: 562.2006\n",
      "Epoch 646/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 887241.1875 - root_mean_squared_error: 941.9348 - val_loss: 315311.1875 - val_root_mean_squared_error: 561.5258\n",
      "Epoch 647/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 870454.6250 - root_mean_squared_error: 932.9816 - val_loss: 319905.4375 - val_root_mean_squared_error: 565.6018\n",
      "Epoch 648/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 887217.6875 - root_mean_squared_error: 941.9224 - val_loss: 324408.0312 - val_root_mean_squared_error: 569.5683\n",
      "Epoch 649/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 846036.8750 - root_mean_squared_error: 919.8026 - val_loss: 323591.6250 - val_root_mean_squared_error: 568.8511\n",
      "Epoch 650/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 896695.6250 - root_mean_squared_error: 946.9401 - val_loss: 318735.6250 - val_root_mean_squared_error: 564.5668\n",
      "Epoch 651/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 854532.0000 - root_mean_squared_error: 924.4090 - val_loss: 313282.0000 - val_root_mean_squared_error: 559.7160\n",
      "Epoch 652/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 490560.3750 - root_mean_squared_error: 700.40 - 0s 5ms/step - loss: 877011.1875 - root_mean_squared_error: 936.4888 - val_loss: 303218.9688 - val_root_mean_squared_error: 550.6532\n",
      "Epoch 653/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 874899.2500 - root_mean_squared_error: 935.3605 - val_loss: 308373.8750 - val_root_mean_squared_error: 555.3142\n",
      "Epoch 654/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 874668.5000 - root_mean_squared_error: 935.2371 - val_loss: 299869.8750 - val_root_mean_squared_error: 547.6038\n",
      "Epoch 655/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 839880.8125 - root_mean_squared_error: 916.4501 - val_loss: 291909.5938 - val_root_mean_squared_error: 540.2866\n",
      "Epoch 656/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 839631.1250 - root_mean_squared_error: 916.3139 - val_loss: 285201.9062 - val_root_mean_squared_error: 534.0430\n",
      "Epoch 657/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 823937.0625 - root_mean_squared_error: 907.7098 - val_loss: 291129.6250 - val_root_mean_squared_error: 539.5643\n",
      "Epoch 658/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 858907.2500 - root_mean_squared_error: 926.7725 - val_loss: 287019.5000 - val_root_mean_squared_error: 535.7420\n",
      "Epoch 659/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 822795.8125 - root_mean_squared_error: 907.0809 - val_loss: 290860.0625 - val_root_mean_squared_error: 539.3145\n",
      "Epoch 660/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 854738.8750 - root_mean_squared_error: 924.5209 - val_loss: 296296.5625 - val_root_mean_squared_error: 544.3313\n",
      "Epoch 661/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 862165.0625 - root_mean_squared_error: 928.5284 - val_loss: 279844.6562 - val_root_mean_squared_error: 529.0035\n",
      "Epoch 662/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 818876.0625 - root_mean_squared_error: 904.9177 - val_loss: 278379.3438 - val_root_mean_squared_error: 527.6166\n",
      "Epoch 663/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 831523.8125 - root_mean_squared_error: 911.8793 - val_loss: 285177.7188 - val_root_mean_squared_error: 534.0203\n",
      "Epoch 664/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 797171.6250 - root_mean_squared_error: 892.8447 - val_loss: 278920.4375 - val_root_mean_squared_error: 528.1292\n",
      "Epoch 665/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 838458.1875 - root_mean_squared_error: 915.6736 - val_loss: 279090.2812 - val_root_mean_squared_error: 528.2900\n",
      "Epoch 666/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 823385.7500 - root_mean_squared_error: 907.4061 - val_loss: 276362.1875 - val_root_mean_squared_error: 525.7016\n",
      "Epoch 667/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 808460.8125 - root_mean_squared_error: 899.1445 - val_loss: 287147.8125 - val_root_mean_squared_error: 535.8618\n",
      "Epoch 668/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 796280.8750 - root_mean_squared_error: 892.3457 - val_loss: 283454.4688 - val_root_mean_squared_error: 532.4044\n",
      "Epoch 669/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 797215.0625 - root_mean_squared_error: 892.8690 - val_loss: 270534.0000 - val_root_mean_squared_error: 520.1288\n",
      "Epoch 670/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 783247.0000 - root_mean_squared_error: 885.0125 - val_loss: 282652.3750 - val_root_mean_squared_error: 531.6506\n",
      "Epoch 671/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 799089.8125 - root_mean_squared_error: 893.9182 - val_loss: 274884.2500 - val_root_mean_squared_error: 524.2941\n",
      "Epoch 672/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 793177.0625 - root_mean_squared_error: 890.6049 - val_loss: 273184.5938 - val_root_mean_squared_error: 522.6707\n",
      "Epoch 673/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 798227.6875 - root_mean_squared_error: 893.4359 - val_loss: 276081.1875 - val_root_mean_squared_error: 525.4343\n",
      "Epoch 674/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 793076.5000 - root_mean_squared_error: 890.5484 - val_loss: 264115.0312 - val_root_mean_squared_error: 513.9212\n",
      "Epoch 675/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 776256.9375 - root_mean_squared_error: 881.0544 - val_loss: 262220.5312 - val_root_mean_squared_error: 512.0747\n",
      "Epoch 676/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 773986.3125 - root_mean_squared_error: 879.7649 - val_loss: 266285.3438 - val_root_mean_squared_error: 516.0284\n",
      "Epoch 677/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 777608.8750 - root_mean_squared_error: 881.8214 - val_loss: 273560.7812 - val_root_mean_squared_error: 523.0304\n",
      "Epoch 678/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 793021.0625 - root_mean_squared_error: 890.5173 - val_loss: 275009.3125 - val_root_mean_squared_error: 524.4133\n",
      "Epoch 679/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 746148.0625 - root_mean_squared_error: 863.7986 - val_loss: 280157.9062 - val_root_mean_squared_error: 529.2994\n",
      "Epoch 680/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 774606.9375 - root_mean_squared_error: 880.1176 - val_loss: 279147.9688 - val_root_mean_squared_error: 528.3445\n",
      "Epoch 681/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 741472.5625 - root_mean_squared_error: 861.0880 - val_loss: 277340.3125 - val_root_mean_squared_error: 526.6311\n",
      "Epoch 682/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 768930.1875 - root_mean_squared_error: 876.8867 - val_loss: 277846.2812 - val_root_mean_squared_error: 527.1113\n",
      "Epoch 683/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 742012.1250 - root_mean_squared_error: 861.4012 - val_loss: 280011.5938 - val_root_mean_squared_error: 529.1612\n",
      "Epoch 684/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 728250.8125 - root_mean_squared_error: 853.3761 - val_loss: 274126.0938 - val_root_mean_squared_error: 523.5705\n",
      "Epoch 685/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 734372.7500 - root_mean_squared_error: 856.9555 - val_loss: 280157.5000 - val_root_mean_squared_error: 529.2991\n",
      "Epoch 686/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 751997.0000 - root_mean_squared_error: 867.1776 - val_loss: 280530.7812 - val_root_mean_squared_error: 529.6516\n",
      "Epoch 687/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 738494.4375 - root_mean_squared_error: 859.3570 - val_loss: 270920.4688 - val_root_mean_squared_error: 520.5002\n",
      "Epoch 688/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 743954.7500 - root_mean_squared_error: 862.5281 - val_loss: 281527.3750 - val_root_mean_squared_error: 530.5916\n",
      "Epoch 689/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 715860.1250 - root_mean_squared_error: 846.0852 - val_loss: 284417.4375 - val_root_mean_squared_error: 533.3080\n",
      "Epoch 690/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 727935.7500 - root_mean_squared_error: 853.1915 - val_loss: 280624.4688 - val_root_mean_squared_error: 529.7400\n",
      "Epoch 691/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 726788.6250 - root_mean_squared_error: 852.5190 - val_loss: 271293.8438 - val_root_mean_squared_error: 520.8588\n",
      "Epoch 692/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 729434.4375 - root_mean_squared_error: 854.0693 - val_loss: 280509.6875 - val_root_mean_squared_error: 529.6317\n",
      "Epoch 693/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 722827.5000 - root_mean_squared_error: 850.1926 - val_loss: 271970.0312 - val_root_mean_squared_error: 521.5074\n",
      "Epoch 694/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 693036.8125 - root_mean_squared_error: 832.4883 - val_loss: 274462.8438 - val_root_mean_squared_error: 523.8920\n",
      "Epoch 695/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 670249.1875 - root_mean_squared_error: 818.6875 - val_loss: 275378.2812 - val_root_mean_squared_error: 524.7650\n",
      "Epoch 696/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 701979.2500 - root_mean_squared_error: 837.8420 - val_loss: 273032.5312 - val_root_mean_squared_error: 522.5251\n",
      "Epoch 697/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 711008.6250 - root_mean_squared_error: 843.2133 - val_loss: 263413.2812 - val_root_mean_squared_error: 513.2380\n",
      "Epoch 698/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 737881.8750 - root_mean_squared_error: 859.0005 - val_loss: 260266.7812 - val_root_mean_squared_error: 510.1635\n",
      "Epoch 699/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 694998.2500 - root_mean_squared_error: 833.6655 - val_loss: 255769.5312 - val_root_mean_squared_error: 505.7366\n",
      "Epoch 700/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 702647.3750 - root_mean_squared_error: 838.2407 - val_loss: 251498.5000 - val_root_mean_squared_error: 501.4962\n",
      "Epoch 701/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 667201.5625 - root_mean_squared_error: 816.8241 - val_loss: 261589.1094 - val_root_mean_squared_error: 511.4578\n",
      "Epoch 702/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 677380.6875 - root_mean_squared_error: 823.0314 - val_loss: 263607.5000 - val_root_mean_squared_error: 513.4272\n",
      "Epoch 703/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 669087.6875 - root_mean_squared_error: 817.9778 - val_loss: 265786.0625 - val_root_mean_squared_error: 515.5444\n",
      "Epoch 704/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 662868.8125 - root_mean_squared_error: 814.1675 - val_loss: 266722.9688 - val_root_mean_squared_error: 516.4523\n",
      "Epoch 705/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 685710.3750 - root_mean_squared_error: 828.0763 - val_loss: 252702.7344 - val_root_mean_squared_error: 502.6955\n",
      "Epoch 706/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 720079.9375 - root_mean_squared_error: 848.5753 - val_loss: 255036.5156 - val_root_mean_squared_error: 505.0114\n",
      "Epoch 707/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 705259.7500 - root_mean_squared_error: 839.7974 - val_loss: 251705.0625 - val_root_mean_squared_error: 501.7022\n",
      "Epoch 708/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 665946.1250 - root_mean_squared_error: 816.0552 - val_loss: 248392.6250 - val_root_mean_squared_error: 498.3900\n",
      "Epoch 709/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 678480.1875 - root_mean_squared_error: 823.6991 - val_loss: 249782.3906 - val_root_mean_squared_error: 499.7823\n",
      "Epoch 710/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 669321.1250 - root_mean_squared_error: 818.1205 - val_loss: 250008.1562 - val_root_mean_squared_error: 500.0081\n",
      "Epoch 711/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 663361.8125 - root_mean_squared_error: 814.4703 - val_loss: 263360.5000 - val_root_mean_squared_error: 513.1866\n",
      "Epoch 712/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 622189.9375 - root_mean_squared_error: 788.7902 - val_loss: 257748.4531 - val_root_mean_squared_error: 507.6893\n",
      "Epoch 713/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 666179.8750 - root_mean_squared_error: 816.1984 - val_loss: 254045.1250 - val_root_mean_squared_error: 504.0289\n",
      "Epoch 714/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 621301.2500 - root_mean_squared_error: 788.2266 - val_loss: 265617.5312 - val_root_mean_squared_error: 515.3810\n",
      "Epoch 715/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 666215.1875 - root_mean_squared_error: 816.2200 - val_loss: 260333.5469 - val_root_mean_squared_error: 510.2289\n",
      "Epoch 716/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 655129.6875 - root_mean_squared_error: 809.4008 - val_loss: 271601.1250 - val_root_mean_squared_error: 521.1536\n",
      "Epoch 717/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 630381.0625 - root_mean_squared_error: 793.9654 - val_loss: 263645.7188 - val_root_mean_squared_error: 513.4644\n",
      "Epoch 718/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 618737.1250 - root_mean_squared_error: 786.5984 - val_loss: 267032.9375 - val_root_mean_squared_error: 516.7523\n",
      "Epoch 719/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 637403.8125 - root_mean_squared_error: 798.3757 - val_loss: 276285.3125 - val_root_mean_squared_error: 525.6285\n",
      "Epoch 720/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 641585.0000 - root_mean_squared_error: 800.9900 - val_loss: 265630.8438 - val_root_mean_squared_error: 515.3939\n",
      "Epoch 721/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 605203.8125 - root_mean_squared_error: 777.9485 - val_loss: 254586.5312 - val_root_mean_squared_error: 504.5657\n",
      "Epoch 722/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 605807.5625 - root_mean_squared_error: 778.3364 - val_loss: 247897.9219 - val_root_mean_squared_error: 497.8935\n",
      "Epoch 723/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 632429.7500 - root_mean_squared_error: 795.2545 - val_loss: 255552.9062 - val_root_mean_squared_error: 505.5224\n",
      "Epoch 724/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 594872.1250 - root_mean_squared_error: 771.2795 - val_loss: 254304.0781 - val_root_mean_squared_error: 504.2857\n",
      "Epoch 725/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 598652.6875 - root_mean_squared_error: 773.7265 - val_loss: 247083.7188 - val_root_mean_squared_error: 497.0752\n",
      "Epoch 726/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 612863.1875 - root_mean_squared_error: 782.8558 - val_loss: 241552.7500 - val_root_mean_squared_error: 491.4802\n",
      "Epoch 727/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 602909.8750 - root_mean_squared_error: 776.4727 - val_loss: 246323.6719 - val_root_mean_squared_error: 496.3101\n",
      "Epoch 728/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 615601.1875 - root_mean_squared_error: 784.6025 - val_loss: 242245.1719 - val_root_mean_squared_error: 492.1841\n",
      "Epoch 729/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 633963.1250 - root_mean_squared_error: 796.2180 - val_loss: 238390.3281 - val_root_mean_squared_error: 488.2523\n",
      "Epoch 730/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 598431.0625 - root_mean_squared_error: 773.5833 - val_loss: 249126.9375 - val_root_mean_squared_error: 499.1262\n",
      "Epoch 731/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 603037.4375 - root_mean_squared_error: 776.5549 - val_loss: 242023.8281 - val_root_mean_squared_error: 491.9592\n",
      "Epoch 732/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 633911.6875 - root_mean_squared_error: 796.1857 - val_loss: 245845.8281 - val_root_mean_squared_error: 495.8284\n",
      "Epoch 733/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 595348.5625 - root_mean_squared_error: 771.5883 - val_loss: 245788.3281 - val_root_mean_squared_error: 495.7704\n",
      "Epoch 734/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 587170.5000 - root_mean_squared_error: 766.2705 - val_loss: 250813.9844 - val_root_mean_squared_error: 500.8133\n",
      "Epoch 735/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 616688.0625 - root_mean_squared_error: 785.2949 - val_loss: 255471.6094 - val_root_mean_squared_error: 505.4420\n",
      "Epoch 736/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 587602.5625 - root_mean_squared_error: 766.5524 - val_loss: 255044.0625 - val_root_mean_squared_error: 505.0189\n",
      "Epoch 737/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 566990.1875 - root_mean_squared_error: 752.9875 - val_loss: 255109.9062 - val_root_mean_squared_error: 505.0840\n",
      "Epoch 738/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 607794.8750 - root_mean_squared_error: 779.6120 - val_loss: 232250.6406 - val_root_mean_squared_error: 481.9239\n",
      "Epoch 739/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 578490.2500 - root_mean_squared_error: 760.5854 - val_loss: 245785.7031 - val_root_mean_squared_error: 495.7678\n",
      "Epoch 740/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 540654.1875 - root_mean_squared_error: 735.2919 - val_loss: 242362.0000 - val_root_mean_squared_error: 492.3028\n",
      "Epoch 741/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 567458.6250 - root_mean_squared_error: 753.2985 - val_loss: 243215.0312 - val_root_mean_squared_error: 493.1684\n",
      "Epoch 742/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 584291.6875 - root_mean_squared_error: 764.3898 - val_loss: 242718.8438 - val_root_mean_squared_error: 492.6650\n",
      "Epoch 743/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 551823.0625 - root_mean_squared_error: 742.8480 - val_loss: 234828.5781 - val_root_mean_squared_error: 484.5912\n",
      "Epoch 744/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 591771.3750 - root_mean_squared_error: 769.2668 - val_loss: 255477.4375 - val_root_mean_squared_error: 505.4478\n",
      "Epoch 745/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 557260.7500 - root_mean_squared_error: 746.4990 - val_loss: 244513.5469 - val_root_mean_squared_error: 494.4831\n",
      "Epoch 746/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 536578.3125 - root_mean_squared_error: 732.5151 - val_loss: 253152.2969 - val_root_mean_squared_error: 503.1424\n",
      "Epoch 747/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 548702.5000 - root_mean_squared_error: 740.7446 - val_loss: 250943.7812 - val_root_mean_squared_error: 500.9429\n",
      "Epoch 748/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 549759.7500 - root_mean_squared_error: 741.4578 - val_loss: 246516.9219 - val_root_mean_squared_error: 496.5047\n",
      "Epoch 749/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 559247.9375 - root_mean_squared_error: 747.8288 - val_loss: 246330.7500 - val_root_mean_squared_error: 496.3172\n",
      "Epoch 750/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 567960.3125 - root_mean_squared_error: 753.6314 - val_loss: 241940.0938 - val_root_mean_squared_error: 491.8741\n",
      "Epoch 751/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 570839.6875 - root_mean_squared_error: 755.5393 - val_loss: 247447.9844 - val_root_mean_squared_error: 497.4414\n",
      "Epoch 752/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 546423.0625 - root_mean_squared_error: 739.2043 - val_loss: 270461.5000 - val_root_mean_squared_error: 520.0591\n",
      "Epoch 753/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 563197.7500 - root_mean_squared_error: 750.4650 - val_loss: 257863.5781 - val_root_mean_squared_error: 507.8027\n",
      "Epoch 754/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 543150.1875 - root_mean_squared_error: 736.9872 - val_loss: 247833.3906 - val_root_mean_squared_error: 497.8287\n",
      "Epoch 755/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 553502.1250 - root_mean_squared_error: 743.9772 - val_loss: 246398.5000 - val_root_mean_squared_error: 496.3854\n",
      "Epoch 756/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 557137.6875 - root_mean_squared_error: 746.4166 - val_loss: 252320.0625 - val_root_mean_squared_error: 502.3147\n",
      "Epoch 757/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 542137.5625 - root_mean_squared_error: 736.2999 - val_loss: 252851.4219 - val_root_mean_squared_error: 502.8433\n",
      "Epoch 758/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 521861.1562 - root_mean_squared_error: 722.3996 - val_loss: 247199.8906 - val_root_mean_squared_error: 497.1920\n",
      "Epoch 759/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 528432.6875 - root_mean_squared_error: 726.9338 - val_loss: 253118.2812 - val_root_mean_squared_error: 503.1086\n",
      "Epoch 760/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 493696.7500 - root_mean_squared_error: 702.6356 - val_loss: 241883.6875 - val_root_mean_squared_error: 491.8167\n",
      "Epoch 761/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 503341.5000 - root_mean_squared_error: 709.4656 - val_loss: 249503.1406 - val_root_mean_squared_error: 499.5029\n",
      "Epoch 762/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 539138.5625 - root_mean_squared_error: 734.2606 - val_loss: 240526.2031 - val_root_mean_squared_error: 490.4347\n",
      "Epoch 763/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 530644.4375 - root_mean_squared_error: 728.4534 - val_loss: 237979.5156 - val_root_mean_squared_error: 487.8315\n",
      "Epoch 764/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 513204.9062 - root_mean_squared_error: 716.3832 - val_loss: 233277.0469 - val_root_mean_squared_error: 482.9876\n",
      "Epoch 765/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 506541.8750 - root_mean_squared_error: 711.7175 - val_loss: 249434.8594 - val_root_mean_squared_error: 499.4345\n",
      "Epoch 766/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 497308.4688 - root_mean_squared_error: 705.2010 - val_loss: 247391.6406 - val_root_mean_squared_error: 497.3848\n",
      "Epoch 767/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 513635.2812 - root_mean_squared_error: 716.6835 - val_loss: 249204.8906 - val_root_mean_squared_error: 499.2043\n",
      "Epoch 768/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 495613.6250 - root_mean_squared_error: 703.9983 - val_loss: 245884.2969 - val_root_mean_squared_error: 495.8672\n",
      "Epoch 769/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 492307.0312 - root_mean_squared_error: 701.6459 - val_loss: 246862.9531 - val_root_mean_squared_error: 496.8531\n",
      "Epoch 770/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 525608.1250 - root_mean_squared_error: 724.9883 - val_loss: 247169.6562 - val_root_mean_squared_error: 497.1616\n",
      "Epoch 771/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 507592.0312 - root_mean_squared_error: 712.4550 - val_loss: 250202.6875 - val_root_mean_squared_error: 500.2026\n",
      "Epoch 772/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 503298.5938 - root_mean_squared_error: 709.4354 - val_loss: 240595.5312 - val_root_mean_squared_error: 490.5054\n",
      "Epoch 773/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 488186.5312 - root_mean_squared_error: 698.7035 - val_loss: 248425.5938 - val_root_mean_squared_error: 498.4231\n",
      "Epoch 774/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 516328.3125 - root_mean_squared_error: 718.5599 - val_loss: 233873.6094 - val_root_mean_squared_error: 483.6048\n",
      "Epoch 775/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 497553.2812 - root_mean_squared_error: 705.3746 - val_loss: 257168.3281 - val_root_mean_squared_error: 507.1177\n",
      "Epoch 776/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 490034.4688 - root_mean_squared_error: 700.0246 - val_loss: 247098.8125 - val_root_mean_squared_error: 497.0903\n",
      "Epoch 777/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 490648.2188 - root_mean_squared_error: 700.4629 - val_loss: 246811.9062 - val_root_mean_squared_error: 496.8017\n",
      "Epoch 778/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 480187.5000 - root_mean_squared_error: 692.9556 - val_loss: 237011.7812 - val_root_mean_squared_error: 486.8386\n",
      "Epoch 779/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 510904.8125 - root_mean_squared_error: 714.7761 - val_loss: 245467.7969 - val_root_mean_squared_error: 495.4471\n",
      "Epoch 780/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 508228.7188 - root_mean_squared_error: 712.9016 - val_loss: 257754.6875 - val_root_mean_squared_error: 507.6955\n",
      "Epoch 781/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 490306.6250 - root_mean_squared_error: 700.2190 - val_loss: 244103.2188 - val_root_mean_squared_error: 494.0680\n",
      "Epoch 782/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 506015.1562 - root_mean_squared_error: 711.3474 - val_loss: 245425.2656 - val_root_mean_squared_error: 495.4041\n",
      "Epoch 783/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 492293.9688 - root_mean_squared_error: 701.6367 - val_loss: 249650.3750 - val_root_mean_squared_error: 499.6502\n",
      "Epoch 784/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 484891.1250 - root_mean_squared_error: 696.3412 - val_loss: 258616.6719 - val_root_mean_squared_error: 508.5437\n",
      "Epoch 785/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 485911.9688 - root_mean_squared_error: 697.0739 - val_loss: 261091.6250 - val_root_mean_squared_error: 510.9713\n",
      "Epoch 786/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 464154.5625 - root_mean_squared_error: 681.2889 - val_loss: 270701.5000 - val_root_mean_squared_error: 520.2898\n",
      "Epoch 787/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 482416.3125 - root_mean_squared_error: 694.5620 - val_loss: 252956.6250 - val_root_mean_squared_error: 502.9479\n",
      "Epoch 788/1600\n",
      "13/13 [==============================] - ETA: 0s - loss: 466380.5000 - root_mean_squared_error: 682.92 - 0s 5ms/step - loss: 486417.6562 - root_mean_squared_error: 697.4365 - val_loss: 239497.5938 - val_root_mean_squared_error: 489.3849\n",
      "Epoch 789/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 478569.7500 - root_mean_squared_error: 691.7874 - val_loss: 239444.4531 - val_root_mean_squared_error: 489.3306\n",
      "Epoch 790/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 468724.7500 - root_mean_squared_error: 684.6348 - val_loss: 234330.2031 - val_root_mean_squared_error: 484.0767\n",
      "Epoch 791/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 464519.5938 - root_mean_squared_error: 681.5568 - val_loss: 244292.4219 - val_root_mean_squared_error: 494.2595\n",
      "Epoch 792/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 476060.3438 - root_mean_squared_error: 689.9713 - val_loss: 227069.2500 - val_root_mean_squared_error: 476.5179\n",
      "Epoch 793/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 449152.2500 - root_mean_squared_error: 670.1882 - val_loss: 248577.9844 - val_root_mean_squared_error: 498.5760\n",
      "Epoch 794/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 468220.1250 - root_mean_squared_error: 684.2661 - val_loss: 237906.5156 - val_root_mean_squared_error: 487.7566\n",
      "Epoch 795/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 446954.3125 - root_mean_squared_error: 668.5464 - val_loss: 243775.4688 - val_root_mean_squared_error: 493.7362\n",
      "Epoch 796/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 439740.6875 - root_mean_squared_error: 663.1295 - val_loss: 231842.6875 - val_root_mean_squared_error: 481.5005\n",
      "Epoch 797/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 446180.8125 - root_mean_squared_error: 667.9677 - val_loss: 238083.4219 - val_root_mean_squared_error: 487.9379\n",
      "Epoch 798/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 438394.3438 - root_mean_squared_error: 662.1135 - val_loss: 243098.0312 - val_root_mean_squared_error: 493.0497\n",
      "Epoch 799/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 443083.7188 - root_mean_squared_error: 665.6453 - val_loss: 246703.4219 - val_root_mean_squared_error: 496.6925\n",
      "Epoch 800/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 429069.6875 - root_mean_squared_error: 655.0341 - val_loss: 234652.6250 - val_root_mean_squared_error: 484.4096\n",
      "Epoch 801/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 459129.3125 - root_mean_squared_error: 677.5908 - val_loss: 231905.5781 - val_root_mean_squared_error: 481.5658\n",
      "Epoch 802/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 428538.2188 - root_mean_squared_error: 654.6283 - val_loss: 240342.9375 - val_root_mean_squared_error: 490.2478\n",
      "Epoch 803/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 415064.5312 - root_mean_squared_error: 644.2550 - val_loss: 235481.1406 - val_root_mean_squared_error: 485.2640\n",
      "Epoch 804/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 440631.0625 - root_mean_squared_error: 663.8005 - val_loss: 233507.8125 - val_root_mean_squared_error: 483.2265\n",
      "Epoch 805/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 424581.8750 - root_mean_squared_error: 651.5995 - val_loss: 229486.3750 - val_root_mean_squared_error: 479.0474\n",
      "Epoch 806/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 439063.0625 - root_mean_squared_error: 662.6183 - val_loss: 247661.4531 - val_root_mean_squared_error: 497.6559\n",
      "Epoch 807/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 456637.2500 - root_mean_squared_error: 675.7494 - val_loss: 242408.1250 - val_root_mean_squared_error: 492.3496\n",
      "Epoch 808/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 439148.5625 - root_mean_squared_error: 662.6829 - val_loss: 243536.3594 - val_root_mean_squared_error: 493.4940\n",
      "Epoch 809/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 460001.0000 - root_mean_squared_error: 678.2338 - val_loss: 233989.1875 - val_root_mean_squared_error: 483.7243\n",
      "Epoch 810/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 425114.2188 - root_mean_squared_error: 652.0078 - val_loss: 232511.8594 - val_root_mean_squared_error: 482.1948\n",
      "Epoch 811/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 416413.9688 - root_mean_squared_error: 645.3015 - val_loss: 237809.5469 - val_root_mean_squared_error: 487.6572\n",
      "Epoch 812/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 408586.5000 - root_mean_squared_error: 639.2077 - val_loss: 221125.4844 - val_root_mean_squared_error: 470.2398\n",
      "Epoch 813/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 375081.0625 - root_mean_squared_error: 612.4386 - val_loss: 240682.3906 - val_root_mean_squared_error: 490.5939\n",
      "Epoch 814/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 433693.0000 - root_mean_squared_error: 658.5537 - val_loss: 243918.0938 - val_root_mean_squared_error: 493.8806\n",
      "Epoch 815/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 411387.5312 - root_mean_squared_error: 641.3950 - val_loss: 235980.5469 - val_root_mean_squared_error: 485.7783\n",
      "Epoch 816/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 399015.7188 - root_mean_squared_error: 631.6769 - val_loss: 225654.9375 - val_root_mean_squared_error: 475.0315\n",
      "Epoch 817/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 415620.1562 - root_mean_squared_error: 644.6861 - val_loss: 232706.6562 - val_root_mean_squared_error: 482.3968\n",
      "Epoch 818/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 399509.0000 - root_mean_squared_error: 632.0673 - val_loss: 226007.9219 - val_root_mean_squared_error: 475.4029\n",
      "Epoch 819/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 403858.2500 - root_mean_squared_error: 635.4984 - val_loss: 219360.3281 - val_root_mean_squared_error: 468.3592\n",
      "Epoch 820/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 407346.6875 - root_mean_squared_error: 638.2372 - val_loss: 217954.4531 - val_root_mean_squared_error: 466.8559\n",
      "Epoch 821/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 402303.1875 - root_mean_squared_error: 634.2737 - val_loss: 215497.3750 - val_root_mean_squared_error: 464.2169\n",
      "Epoch 822/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 373066.5000 - root_mean_squared_error: 610.7917 - val_loss: 226442.5469 - val_root_mean_squared_error: 475.8598\n",
      "Epoch 823/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 391033.7812 - root_mean_squared_error: 625.3270 - val_loss: 237434.5312 - val_root_mean_squared_error: 487.2726\n",
      "Epoch 824/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 400338.3125 - root_mean_squared_error: 632.7230 - val_loss: 222111.3906 - val_root_mean_squared_error: 471.2870\n",
      "Epoch 825/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 428157.8438 - root_mean_squared_error: 654.3377 - val_loss: 239061.1719 - val_root_mean_squared_error: 488.9388\n",
      "Epoch 826/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 445629.7188 - root_mean_squared_error: 667.5551 - val_loss: 214155.7188 - val_root_mean_squared_error: 462.7696\n",
      "Epoch 827/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 382087.7500 - root_mean_squared_error: 618.1324 - val_loss: 229635.1250 - val_root_mean_squared_error: 479.2026\n",
      "Epoch 828/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 390100.0312 - root_mean_squared_error: 624.5799 - val_loss: 199276.1875 - val_root_mean_squared_error: 446.4036\n",
      "Epoch 829/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 418880.2500 - root_mean_squared_error: 647.2096 - val_loss: 230012.8438 - val_root_mean_squared_error: 479.5966\n",
      "Epoch 830/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 412350.0000 - root_mean_squared_error: 642.1448 - val_loss: 231776.1562 - val_root_mean_squared_error: 481.4314\n",
      "Epoch 831/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 421668.1562 - root_mean_squared_error: 649.3598 - val_loss: 222392.6719 - val_root_mean_squared_error: 471.5853\n",
      "Epoch 832/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 404469.5000 - root_mean_squared_error: 635.9792 - val_loss: 221096.1406 - val_root_mean_squared_error: 470.2086\n",
      "Epoch 833/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 389844.5625 - root_mean_squared_error: 624.3754 - val_loss: 211680.0781 - val_root_mean_squared_error: 460.0870\n",
      "Epoch 834/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 390728.4688 - root_mean_squared_error: 625.0828 - val_loss: 214404.7500 - val_root_mean_squared_error: 463.0386\n",
      "Epoch 835/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 380669.8750 - root_mean_squared_error: 616.9845 - val_loss: 220470.1562 - val_root_mean_squared_error: 469.5425\n",
      "Epoch 836/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 372389.0938 - root_mean_squared_error: 610.2369 - val_loss: 224870.4062 - val_root_mean_squared_error: 474.2050\n",
      "Epoch 837/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 352160.9062 - root_mean_squared_error: 593.4315 - val_loss: 224631.9062 - val_root_mean_squared_error: 473.9535\n",
      "Epoch 838/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 377916.9375 - root_mean_squared_error: 614.7495 - val_loss: 216289.0000 - val_root_mean_squared_error: 465.0688\n",
      "Epoch 839/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 418573.9375 - root_mean_squared_error: 646.9729 - val_loss: 213564.0781 - val_root_mean_squared_error: 462.1299\n",
      "Epoch 840/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 373301.0000 - root_mean_squared_error: 610.9836 - val_loss: 218691.0938 - val_root_mean_squared_error: 467.6442\n",
      "Epoch 841/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 386617.9688 - root_mean_squared_error: 621.7861 - val_loss: 220007.6094 - val_root_mean_squared_error: 469.0497\n",
      "Epoch 842/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 389215.6875 - root_mean_squared_error: 623.8715 - val_loss: 224060.2344 - val_root_mean_squared_error: 473.3500\n",
      "Epoch 843/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 369541.2500 - root_mean_squared_error: 607.8990 - val_loss: 222948.4219 - val_root_mean_squared_error: 472.1741\n",
      "Epoch 844/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 378591.6250 - root_mean_squared_error: 615.2980 - val_loss: 238777.4531 - val_root_mean_squared_error: 488.6486\n",
      "Epoch 845/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 381184.9062 - root_mean_squared_error: 617.4017 - val_loss: 222284.1875 - val_root_mean_squared_error: 471.4702\n",
      "Epoch 846/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 361491.7812 - root_mean_squared_error: 601.2419 - val_loss: 212945.3594 - val_root_mean_squared_error: 461.4600\n",
      "Epoch 847/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 388138.5938 - root_mean_squared_error: 623.0077 - val_loss: 216064.9531 - val_root_mean_squared_error: 464.8279\n",
      "Epoch 848/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 421193.7188 - root_mean_squared_error: 648.9944 - val_loss: 211369.3750 - val_root_mean_squared_error: 459.7492\n",
      "Epoch 849/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 325123.2812 - root_mean_squared_error: 570.1958 - val_loss: 220723.0469 - val_root_mean_squared_error: 469.8117\n",
      "Epoch 850/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 353064.2812 - root_mean_squared_error: 594.1921 - val_loss: 232271.3125 - val_root_mean_squared_error: 481.9453\n",
      "Epoch 851/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 374472.3750 - root_mean_squared_error: 611.9415 - val_loss: 234050.1875 - val_root_mean_squared_error: 483.7873\n",
      "Epoch 852/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 347201.7812 - root_mean_squared_error: 589.2383 - val_loss: 234221.5469 - val_root_mean_squared_error: 483.9644\n",
      "Epoch 853/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 355030.6562 - root_mean_squared_error: 595.8445 - val_loss: 232402.7031 - val_root_mean_squared_error: 482.0816\n",
      "Epoch 854/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 355697.3125 - root_mean_squared_error: 596.4036 - val_loss: 222883.5156 - val_root_mean_squared_error: 472.1054\n",
      "Epoch 855/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 377723.7812 - root_mean_squared_error: 614.5923 - val_loss: 226900.4219 - val_root_mean_squared_error: 476.3407\n",
      "Epoch 856/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 395029.0000 - root_mean_squared_error: 628.5133 - val_loss: 215169.7969 - val_root_mean_squared_error: 463.8640\n",
      "Epoch 857/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 338951.6562 - root_mean_squared_error: 582.1956 - val_loss: 200183.8281 - val_root_mean_squared_error: 447.4191\n",
      "Epoch 858/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 346476.7188 - root_mean_squared_error: 588.6227 - val_loss: 221979.5781 - val_root_mean_squared_error: 471.1471\n",
      "Epoch 859/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 350278.9062 - root_mean_squared_error: 591.8436 - val_loss: 233192.8438 - val_root_mean_squared_error: 482.9005\n",
      "Epoch 860/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 393226.5938 - root_mean_squared_error: 627.0778 - val_loss: 220021.7500 - val_root_mean_squared_error: 469.0648\n",
      "Epoch 861/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 360309.2500 - root_mean_squared_error: 600.2576 - val_loss: 227373.8281 - val_root_mean_squared_error: 476.8373\n",
      "Epoch 862/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 344553.2188 - root_mean_squared_error: 586.9866 - val_loss: 218064.3906 - val_root_mean_squared_error: 466.9737\n",
      "Epoch 863/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 375781.1875 - root_mean_squared_error: 613.0099 - val_loss: 218904.4688 - val_root_mean_squared_error: 467.8723\n",
      "Epoch 864/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 359981.7812 - root_mean_squared_error: 599.9848 - val_loss: 226262.6094 - val_root_mean_squared_error: 475.6707\n",
      "Epoch 865/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 339650.5938 - root_mean_squared_error: 582.7955 - val_loss: 230535.2344 - val_root_mean_squared_error: 480.1408\n",
      "Epoch 866/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 359911.8750 - root_mean_squared_error: 599.9266 - val_loss: 212967.9844 - val_root_mean_squared_error: 461.4846\n",
      "Epoch 867/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 357908.1562 - root_mean_squared_error: 598.2543 - val_loss: 204710.2500 - val_root_mean_squared_error: 452.4492\n",
      "Epoch 868/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 372020.1250 - root_mean_squared_error: 609.9345 - val_loss: 206423.0938 - val_root_mean_squared_error: 454.3381\n",
      "Epoch 869/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 372232.4375 - root_mean_squared_error: 610.1085 - val_loss: 211444.0781 - val_root_mean_squared_error: 459.8305\n",
      "Epoch 870/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 354848.0938 - root_mean_squared_error: 595.6913 - val_loss: 210066.7969 - val_root_mean_squared_error: 458.3304\n",
      "Epoch 871/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 327708.5625 - root_mean_squared_error: 572.4584 - val_loss: 226995.2031 - val_root_mean_squared_error: 476.4401\n",
      "Epoch 872/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 332147.4062 - root_mean_squared_error: 576.3223 - val_loss: 223478.6250 - val_root_mean_squared_error: 472.7353\n",
      "Epoch 873/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 401542.9062 - root_mean_squared_error: 633.6741 - val_loss: 219769.6406 - val_root_mean_squared_error: 468.7960\n",
      "Epoch 874/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 353639.6562 - root_mean_squared_error: 594.6761 - val_loss: 223743.6875 - val_root_mean_squared_error: 473.0155\n",
      "Epoch 875/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 346797.0938 - root_mean_squared_error: 588.8948 - val_loss: 207523.0938 - val_root_mean_squared_error: 455.5470\n",
      "Epoch 876/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 362105.5938 - root_mean_squared_error: 601.7521 - val_loss: 209115.8750 - val_root_mean_squared_error: 457.2919\n",
      "Epoch 877/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 358299.0938 - root_mean_squared_error: 598.5809 - val_loss: 224895.4688 - val_root_mean_squared_error: 474.2314\n",
      "Epoch 878/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 329024.1875 - root_mean_squared_error: 573.6063 - val_loss: 219642.4531 - val_root_mean_squared_error: 468.6603\n",
      "Epoch 879/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 314500.8750 - root_mean_squared_error: 560.8038 - val_loss: 220362.8281 - val_root_mean_squared_error: 469.4282\n",
      "Epoch 880/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 362602.2812 - root_mean_squared_error: 602.1647 - val_loss: 216654.7344 - val_root_mean_squared_error: 465.4619\n",
      "Epoch 881/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 347031.4062 - root_mean_squared_error: 589.0937 - val_loss: 211642.4219 - val_root_mean_squared_error: 460.0461\n",
      "Epoch 882/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 330349.7812 - root_mean_squared_error: 574.7606 - val_loss: 212793.7500 - val_root_mean_squared_error: 461.2957\n",
      "Epoch 883/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 315856.8750 - root_mean_squared_error: 562.0115 - val_loss: 219618.0469 - val_root_mean_squared_error: 468.6342\n",
      "Epoch 884/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 340232.4375 - root_mean_squared_error: 583.2945 - val_loss: 227898.4375 - val_root_mean_squared_error: 477.3871\n",
      "Epoch 885/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 321859.4375 - root_mean_squared_error: 567.3266 - val_loss: 218782.6250 - val_root_mean_squared_error: 467.7420\n",
      "Epoch 886/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 308340.2188 - root_mean_squared_error: 555.2839 - val_loss: 223920.4219 - val_root_mean_squared_error: 473.2023\n",
      "Epoch 887/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 333928.7188 - root_mean_squared_error: 577.8657 - val_loss: 223008.3281 - val_root_mean_squared_error: 472.2376\n",
      "Epoch 888/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 342926.3438 - root_mean_squared_error: 585.5991 - val_loss: 223731.0312 - val_root_mean_squared_error: 473.0021\n",
      "Epoch 889/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 347313.4062 - root_mean_squared_error: 589.3330 - val_loss: 222943.2188 - val_root_mean_squared_error: 472.1686\n",
      "Epoch 890/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 313112.5000 - root_mean_squared_error: 559.5646 - val_loss: 224203.2500 - val_root_mean_squared_error: 473.5011\n",
      "Epoch 891/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 333423.4688 - root_mean_squared_error: 577.4283 - val_loss: 232232.7344 - val_root_mean_squared_error: 481.9053\n",
      "Epoch 892/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 325054.3438 - root_mean_squared_error: 570.1354 - val_loss: 235301.9062 - val_root_mean_squared_error: 485.0793\n",
      "Epoch 893/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 314062.7188 - root_mean_squared_error: 560.4130 - val_loss: 239730.9062 - val_root_mean_squared_error: 489.6232\n",
      "Epoch 894/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 298699.3125 - root_mean_squared_error: 546.5339 - val_loss: 227809.5312 - val_root_mean_squared_error: 477.2940\n",
      "Epoch 895/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 287871.9062 - root_mean_squared_error: 536.5370 - val_loss: 256449.4062 - val_root_mean_squared_error: 506.4083\n",
      "Epoch 896/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 315395.1250 - root_mean_squared_error: 561.6005 - val_loss: 253770.1719 - val_root_mean_squared_error: 503.7561\n",
      "Epoch 897/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 324947.3438 - root_mean_squared_error: 570.0415 - val_loss: 238170.9219 - val_root_mean_squared_error: 488.0276\n",
      "Epoch 898/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 305668.8125 - root_mean_squared_error: 552.8732 - val_loss: 250037.7812 - val_root_mean_squared_error: 500.0378\n",
      "Epoch 899/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 336465.9688 - root_mean_squared_error: 580.0569 - val_loss: 235459.9219 - val_root_mean_squared_error: 485.2421\n",
      "Epoch 900/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 308813.8438 - root_mean_squared_error: 555.7102 - val_loss: 230230.2812 - val_root_mean_squared_error: 479.8232\n",
      "Epoch 901/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 330744.8125 - root_mean_squared_error: 575.1042 - val_loss: 222688.4844 - val_root_mean_squared_error: 471.8988\n",
      "Epoch 902/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 315587.4375 - root_mean_squared_error: 561.7717 - val_loss: 244586.2500 - val_root_mean_squared_error: 494.5566\n",
      "Epoch 903/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 293242.1875 - root_mean_squared_error: 541.5184 - val_loss: 235202.5938 - val_root_mean_squared_error: 484.9769\n",
      "Epoch 904/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 331583.6250 - root_mean_squared_error: 575.8330 - val_loss: 241386.6406 - val_root_mean_squared_error: 491.3112\n",
      "Epoch 905/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 315247.5938 - root_mean_squared_error: 561.4691 - val_loss: 241412.3125 - val_root_mean_squared_error: 491.3373\n",
      "Epoch 906/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 296226.2188 - root_mean_squared_error: 544.2667 - val_loss: 262857.7188 - val_root_mean_squared_error: 512.6965\n",
      "Epoch 907/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 299852.5312 - root_mean_squared_error: 547.5879 - val_loss: 248439.0469 - val_root_mean_squared_error: 498.4366\n",
      "Epoch 908/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 300727.0625 - root_mean_squared_error: 548.3859 - val_loss: 243668.9688 - val_root_mean_squared_error: 493.6284\n",
      "Epoch 909/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 326247.6250 - root_mean_squared_error: 571.1809 - val_loss: 239412.7500 - val_root_mean_squared_error: 489.2982\n",
      "Epoch 910/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 298077.8125 - root_mean_squared_error: 545.9650 - val_loss: 240621.2969 - val_root_mean_squared_error: 490.5316\n",
      "Epoch 911/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 307420.6250 - root_mean_squared_error: 554.4553 - val_loss: 252105.7031 - val_root_mean_squared_error: 502.1013\n",
      "Epoch 912/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 279345.2812 - root_mean_squared_error: 528.5312 - val_loss: 242180.9688 - val_root_mean_squared_error: 492.1189\n",
      "Epoch 913/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 312891.8750 - root_mean_squared_error: 559.3674 - val_loss: 238930.7969 - val_root_mean_squared_error: 488.8055\n",
      "Epoch 914/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 310320.2812 - root_mean_squared_error: 557.0640 - val_loss: 240273.9688 - val_root_mean_squared_error: 490.1775\n",
      "Epoch 915/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 323654.6875 - root_mean_squared_error: 568.9066 - val_loss: 253348.0625 - val_root_mean_squared_error: 503.3369\n",
      "Epoch 916/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 319118.3750 - root_mean_squared_error: 564.9056 - val_loss: 245042.4062 - val_root_mean_squared_error: 495.0176\n",
      "Epoch 917/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 297103.7188 - root_mean_squared_error: 545.0722 - val_loss: 254355.7031 - val_root_mean_squared_error: 504.3369\n",
      "Epoch 918/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 317376.1875 - root_mean_squared_error: 563.3615 - val_loss: 237891.2344 - val_root_mean_squared_error: 487.7409\n",
      "Epoch 919/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 298553.7188 - root_mean_squared_error: 546.4007 - val_loss: 223220.2344 - val_root_mean_squared_error: 472.4619\n",
      "Epoch 920/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 322930.4062 - root_mean_squared_error: 568.2697 - val_loss: 227923.4062 - val_root_mean_squared_error: 477.4132\n",
      "Epoch 921/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 299439.9062 - root_mean_squared_error: 547.2110 - val_loss: 229251.7031 - val_root_mean_squared_error: 478.8024\n",
      "Epoch 922/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 336559.7188 - root_mean_squared_error: 580.1377 - val_loss: 231869.7344 - val_root_mean_squared_error: 481.5285\n",
      "Epoch 923/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 287775.0938 - root_mean_squared_error: 536.4467 - val_loss: 230247.0781 - val_root_mean_squared_error: 479.8407\n",
      "Epoch 924/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 313000.7188 - root_mean_squared_error: 559.4647 - val_loss: 230086.6250 - val_root_mean_squared_error: 479.6735\n",
      "Epoch 925/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 288659.3125 - root_mean_squared_error: 537.2703 - val_loss: 216575.8125 - val_root_mean_squared_error: 465.3771\n",
      "Epoch 926/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 311853.1562 - root_mean_squared_error: 558.4381 - val_loss: 236316.7969 - val_root_mean_squared_error: 486.1243\n",
      "Epoch 927/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 315232.5000 - root_mean_squared_error: 561.4557 - val_loss: 228426.7500 - val_root_mean_squared_error: 477.9401\n",
      "Epoch 928/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 291683.3750 - root_mean_squared_error: 540.0772 - val_loss: 228417.6875 - val_root_mean_squared_error: 477.9306\n",
      "Epoch 929/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 311422.2188 - root_mean_squared_error: 558.0522 - val_loss: 234362.9375 - val_root_mean_squared_error: 484.1105\n",
      "Epoch 930/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 282507.6875 - root_mean_squared_error: 531.5145 - val_loss: 220710.5000 - val_root_mean_squared_error: 469.7984\n",
      "Epoch 931/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 315837.2500 - root_mean_squared_error: 561.9940 - val_loss: 239052.2969 - val_root_mean_squared_error: 488.9297\n",
      "Epoch 932/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 327024.7500 - root_mean_squared_error: 571.8608 - val_loss: 226352.5000 - val_root_mean_squared_error: 475.7652\n",
      "Epoch 933/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 294831.8125 - root_mean_squared_error: 542.9842 - val_loss: 225519.8125 - val_root_mean_squared_error: 474.8893\n",
      "Epoch 934/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 258606.8281 - root_mean_squared_error: 508.5340 - val_loss: 244955.5000 - val_root_mean_squared_error: 494.9298\n",
      "Epoch 935/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 280414.2188 - root_mean_squared_error: 529.5415 - val_loss: 223379.8906 - val_root_mean_squared_error: 472.6308\n",
      "Epoch 936/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 322745.8125 - root_mean_squared_error: 568.1072 - val_loss: 248915.6406 - val_root_mean_squared_error: 498.9145\n",
      "Epoch 937/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 282886.6562 - root_mean_squared_error: 531.8709 - val_loss: 235640.4219 - val_root_mean_squared_error: 485.4281\n",
      "Epoch 938/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 261761.7031 - root_mean_squared_error: 511.6265 - val_loss: 237040.7969 - val_root_mean_squared_error: 486.8683\n",
      "Epoch 939/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 301602.9688 - root_mean_squared_error: 549.1839 - val_loss: 241221.0000 - val_root_mean_squared_error: 491.1425\n",
      "Epoch 940/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 307789.5312 - root_mean_squared_error: 554.7878 - val_loss: 231205.2188 - val_root_mean_squared_error: 480.8380\n",
      "Epoch 941/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 340120.6562 - root_mean_squared_error: 583.1987 - val_loss: 223358.6562 - val_root_mean_squared_error: 472.6084\n",
      "Epoch 942/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 305728.7188 - root_mean_squared_error: 552.9274 - val_loss: 231876.9688 - val_root_mean_squared_error: 481.5360\n",
      "Epoch 943/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 270568.3125 - root_mean_squared_error: 520.1618 - val_loss: 225870.9688 - val_root_mean_squared_error: 475.2589\n",
      "Epoch 944/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 285029.3750 - root_mean_squared_error: 533.8814 - val_loss: 244573.8125 - val_root_mean_squared_error: 494.5440\n",
      "Epoch 945/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 275472.1562 - root_mean_squared_error: 524.8544 - val_loss: 229301.1875 - val_root_mean_squared_error: 478.8540\n",
      "Epoch 946/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 297902.8125 - root_mean_squared_error: 545.8047 - val_loss: 231232.6406 - val_root_mean_squared_error: 480.8665\n",
      "Epoch 947/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 316083.7500 - root_mean_squared_error: 562.2133 - val_loss: 237310.1562 - val_root_mean_squared_error: 487.1449\n",
      "Epoch 948/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 302975.5000 - root_mean_squared_error: 550.4321 - val_loss: 220954.5625 - val_root_mean_squared_error: 470.0580\n",
      "Epoch 949/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 313249.5625 - root_mean_squared_error: 559.6870 - val_loss: 206040.1406 - val_root_mean_squared_error: 453.9164\n",
      "Epoch 950/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 298010.3125 - root_mean_squared_error: 545.9032 - val_loss: 240260.0000 - val_root_mean_squared_error: 490.1632\n",
      "Epoch 951/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 275699.7812 - root_mean_squared_error: 525.0712 - val_loss: 224928.7812 - val_root_mean_squared_error: 474.2666\n",
      "Epoch 952/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 308630.4062 - root_mean_squared_error: 555.5452 - val_loss: 216997.8906 - val_root_mean_squared_error: 465.8303\n",
      "Epoch 953/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 277152.5938 - root_mean_squared_error: 526.4528 - val_loss: 259000.6250 - val_root_mean_squared_error: 508.9211\n",
      "Epoch 954/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 279516.9375 - root_mean_squared_error: 528.6936 - val_loss: 230459.6719 - val_root_mean_squared_error: 480.0622\n",
      "Epoch 955/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 259634.3281 - root_mean_squared_error: 509.5432 - val_loss: 243592.2344 - val_root_mean_squared_error: 493.5506\n",
      "Epoch 956/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 308032.5938 - root_mean_squared_error: 555.0068 - val_loss: 243797.8594 - val_root_mean_squared_error: 493.7589\n",
      "Epoch 957/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 253607.0469 - root_mean_squared_error: 503.5941 - val_loss: 246188.1094 - val_root_mean_squared_error: 496.1735\n",
      "Epoch 958/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 287505.3438 - root_mean_squared_error: 536.1953 - val_loss: 211336.4219 - val_root_mean_squared_error: 459.7134\n",
      "Epoch 959/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 288140.6250 - root_mean_squared_error: 536.7873 - val_loss: 236047.1094 - val_root_mean_squared_error: 485.8468\n",
      "Epoch 960/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 286884.5000 - root_mean_squared_error: 535.6160 - val_loss: 235118.1719 - val_root_mean_squared_error: 484.8899\n",
      "Epoch 961/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 294775.5625 - root_mean_squared_error: 542.9324 - val_loss: 226623.4062 - val_root_mean_squared_error: 476.0498\n",
      "Epoch 962/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 280345.9062 - root_mean_squared_error: 529.4770 - val_loss: 210079.3594 - val_root_mean_squared_error: 458.3441\n",
      "Epoch 963/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 273030.9688 - root_mean_squared_error: 522.5237 - val_loss: 209125.2188 - val_root_mean_squared_error: 457.3021\n",
      "Epoch 964/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 299049.5625 - root_mean_squared_error: 546.8542 - val_loss: 217153.3125 - val_root_mean_squared_error: 465.9971\n",
      "Epoch 965/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 268177.1250 - root_mean_squared_error: 517.8582 - val_loss: 201267.5625 - val_root_mean_squared_error: 448.6285\n",
      "Epoch 966/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 260124.0000 - root_mean_squared_error: 510.0235 - val_loss: 203571.3594 - val_root_mean_squared_error: 451.1888\n",
      "Epoch 967/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 263913.0938 - root_mean_squared_error: 513.7247 - val_loss: 214425.8438 - val_root_mean_squared_error: 463.0614\n",
      "Epoch 968/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 303717.7812 - root_mean_squared_error: 551.1060 - val_loss: 204846.0156 - val_root_mean_squared_error: 452.5992\n",
      "Epoch 969/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 298384.0625 - root_mean_squared_error: 546.2454 - val_loss: 200280.1719 - val_root_mean_squared_error: 447.5267\n",
      "Epoch 970/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 281464.4688 - root_mean_squared_error: 530.5322 - val_loss: 185108.9531 - val_root_mean_squared_error: 430.2429\n",
      "Epoch 971/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 281174.1562 - root_mean_squared_error: 530.2585 - val_loss: 211520.2031 - val_root_mean_squared_error: 459.9133\n",
      "Epoch 972/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 261152.4844 - root_mean_squared_error: 511.0308 - val_loss: 204922.7188 - val_root_mean_squared_error: 452.6839\n",
      "Epoch 973/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 285131.9688 - root_mean_squared_error: 533.9775 - val_loss: 204622.0781 - val_root_mean_squared_error: 452.3517\n",
      "Epoch 974/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 258127.4062 - root_mean_squared_error: 508.0624 - val_loss: 221607.6406 - val_root_mean_squared_error: 470.7522\n",
      "Epoch 975/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 306741.0625 - root_mean_squared_error: 553.8421 - val_loss: 208569.3750 - val_root_mean_squared_error: 456.6940\n",
      "Epoch 976/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 298055.5625 - root_mean_squared_error: 545.9446 - val_loss: 229552.5469 - val_root_mean_squared_error: 479.1164\n",
      "Epoch 977/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 278661.6875 - root_mean_squared_error: 527.8842 - val_loss: 211556.8594 - val_root_mean_squared_error: 459.9531\n",
      "Epoch 978/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 304076.4375 - root_mean_squared_error: 551.4313 - val_loss: 230829.3906 - val_root_mean_squared_error: 480.4471\n",
      "Epoch 979/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 258097.6094 - root_mean_squared_error: 508.0331 - val_loss: 224140.7188 - val_root_mean_squared_error: 473.4350\n",
      "Epoch 980/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 265832.2812 - root_mean_squared_error: 515.5892 - val_loss: 218187.6406 - val_root_mean_squared_error: 467.1056\n",
      "Epoch 981/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 259868.3594 - root_mean_squared_error: 509.7729 - val_loss: 222782.0781 - val_root_mean_squared_error: 471.9980\n",
      "Epoch 982/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 293144.5000 - root_mean_squared_error: 541.4282 - val_loss: 228680.6250 - val_root_mean_squared_error: 478.2056\n",
      "Epoch 983/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 279650.6875 - root_mean_squared_error: 528.8201 - val_loss: 195622.1406 - val_root_mean_squared_error: 442.2919\n",
      "Epoch 984/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 271866.7500 - root_mean_squared_error: 521.4084 - val_loss: 219734.7500 - val_root_mean_squared_error: 468.7587\n",
      "Epoch 985/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 290971.7188 - root_mean_squared_error: 539.4180 - val_loss: 210521.0312 - val_root_mean_squared_error: 458.8257\n",
      "Epoch 986/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 243805.8594 - root_mean_squared_error: 493.7670 - val_loss: 206779.9062 - val_root_mean_squared_error: 454.7306\n",
      "Epoch 987/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245970.6250 - root_mean_squared_error: 495.9543 - val_loss: 204510.4531 - val_root_mean_squared_error: 452.2283\n",
      "Epoch 988/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 255795.2031 - root_mean_squared_error: 505.7620 - val_loss: 195477.7031 - val_root_mean_squared_error: 442.1286\n",
      "Epoch 989/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254407.1719 - root_mean_squared_error: 504.3879 - val_loss: 195124.1875 - val_root_mean_squared_error: 441.7286\n",
      "Epoch 990/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 264767.9688 - root_mean_squared_error: 514.5561 - val_loss: 181733.6094 - val_root_mean_squared_error: 426.3022\n",
      "Epoch 991/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 285589.3438 - root_mean_squared_error: 534.4056 - val_loss: 191419.2188 - val_root_mean_squared_error: 437.5148\n",
      "Epoch 992/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 277304.4688 - root_mean_squared_error: 526.5970 - val_loss: 212273.3750 - val_root_mean_squared_error: 460.7314\n",
      "Epoch 993/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 278197.0312 - root_mean_squared_error: 527.4438 - val_loss: 223049.1875 - val_root_mean_squared_error: 472.2808\n",
      "Epoch 994/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 234385.6719 - root_mean_squared_error: 484.1339 - val_loss: 223221.0469 - val_root_mean_squared_error: 472.4627\n",
      "Epoch 995/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 275453.9688 - root_mean_squared_error: 524.8371 - val_loss: 231497.9375 - val_root_mean_squared_error: 481.1423\n",
      "Epoch 996/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 272950.8750 - root_mean_squared_error: 522.4470 - val_loss: 212889.4375 - val_root_mean_squared_error: 461.3994\n",
      "Epoch 997/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 274711.9062 - root_mean_squared_error: 524.1296 - val_loss: 238022.8281 - val_root_mean_squared_error: 487.8758\n",
      "Epoch 998/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255162.8281 - root_mean_squared_error: 505.1364 - val_loss: 238154.9375 - val_root_mean_squared_error: 488.0112\n",
      "Epoch 999/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 242476.7969 - root_mean_squared_error: 492.4193 - val_loss: 217554.9531 - val_root_mean_squared_error: 466.4279\n",
      "Epoch 1000/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 256464.0000 - root_mean_squared_error: 506.4228 - val_loss: 222723.8125 - val_root_mean_squared_error: 471.9362\n",
      "Epoch 1001/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 257400.6719 - root_mean_squared_error: 507.3467 - val_loss: 217909.7969 - val_root_mean_squared_error: 466.8081\n",
      "Epoch 1002/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 259483.6719 - root_mean_squared_error: 509.3954 - val_loss: 205178.7812 - val_root_mean_squared_error: 452.9666\n",
      "Epoch 1003/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 273360.9375 - root_mean_squared_error: 522.8393 - val_loss: 209371.4688 - val_root_mean_squared_error: 457.5713\n",
      "Epoch 1004/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 287457.4688 - root_mean_squared_error: 536.1506 - val_loss: 227315.4219 - val_root_mean_squared_error: 476.7761\n",
      "Epoch 1005/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 271353.7812 - root_mean_squared_error: 520.9163 - val_loss: 214403.4688 - val_root_mean_squared_error: 463.0372\n",
      "Epoch 1006/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241825.5000 - root_mean_squared_error: 491.7576 - val_loss: 231871.3125 - val_root_mean_squared_error: 481.5302\n",
      "Epoch 1007/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 281818.0000 - root_mean_squared_error: 530.8654 - val_loss: 230018.5156 - val_root_mean_squared_error: 479.6024\n",
      "Epoch 1008/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252298.0000 - root_mean_squared_error: 502.2928 - val_loss: 228626.9062 - val_root_mean_squared_error: 478.1495\n",
      "Epoch 1009/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247405.5469 - root_mean_squared_error: 497.3988 - val_loss: 234645.0625 - val_root_mean_squared_error: 484.4018\n",
      "Epoch 1010/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 252931.1094 - root_mean_squared_error: 502.9226 - val_loss: 230045.5000 - val_root_mean_squared_error: 479.6306\n",
      "Epoch 1011/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 280075.0938 - root_mean_squared_error: 529.2212 - val_loss: 241805.8438 - val_root_mean_squared_error: 491.7376\n",
      "Epoch 1012/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249398.6719 - root_mean_squared_error: 499.3983 - val_loss: 233616.8906 - val_root_mean_squared_error: 483.3393\n",
      "Epoch 1013/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245266.1875 - root_mean_squared_error: 495.2436 - val_loss: 229668.2344 - val_root_mean_squared_error: 479.2372\n",
      "Epoch 1014/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262745.3438 - root_mean_squared_error: 512.5869 - val_loss: 228537.3281 - val_root_mean_squared_error: 478.0558\n",
      "Epoch 1015/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 277513.5625 - root_mean_squared_error: 526.7955 - val_loss: 210831.6094 - val_root_mean_squared_error: 459.1640\n",
      "Epoch 1016/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244724.9062 - root_mean_squared_error: 494.6968 - val_loss: 218480.3281 - val_root_mean_squared_error: 467.4188\n",
      "Epoch 1017/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245180.7969 - root_mean_squared_error: 495.1573 - val_loss: 243277.9844 - val_root_mean_squared_error: 493.2322\n",
      "Epoch 1018/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 311658.5625 - root_mean_squared_error: 558.2639 - val_loss: 235726.2656 - val_root_mean_squared_error: 485.5165\n",
      "Epoch 1019/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 267010.6875 - root_mean_squared_error: 516.7308 - val_loss: 230386.9688 - val_root_mean_squared_error: 479.9864\n",
      "Epoch 1020/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 256183.9531 - root_mean_squared_error: 506.1462 - val_loss: 224439.9531 - val_root_mean_squared_error: 473.7509\n",
      "Epoch 1021/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 259948.5000 - root_mean_squared_error: 509.8514 - val_loss: 228736.6406 - val_root_mean_squared_error: 478.2642\n",
      "Epoch 1022/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 315710.2812 - root_mean_squared_error: 561.8810 - val_loss: 238072.1562 - val_root_mean_squared_error: 487.9264\n",
      "Epoch 1023/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254189.8125 - root_mean_squared_error: 504.1724 - val_loss: 230867.7656 - val_root_mean_squared_error: 480.4870\n",
      "Epoch 1024/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233969.8125 - root_mean_squared_error: 483.7043 - val_loss: 212547.5781 - val_root_mean_squared_error: 461.0288\n",
      "Epoch 1025/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 288847.2500 - root_mean_squared_error: 537.4451 - val_loss: 243605.6406 - val_root_mean_squared_error: 493.5642\n",
      "Epoch 1026/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 272995.0000 - root_mean_squared_error: 522.4893 - val_loss: 215021.0156 - val_root_mean_squared_error: 463.7036\n",
      "Epoch 1027/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 240604.1250 - root_mean_squared_error: 490.5141 - val_loss: 229367.2812 - val_root_mean_squared_error: 478.9230\n",
      "Epoch 1028/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 263532.5938 - root_mean_squared_error: 513.3542 - val_loss: 247983.2812 - val_root_mean_squared_error: 497.9792\n",
      "Epoch 1029/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 265262.2812 - root_mean_squared_error: 515.0362 - val_loss: 228530.4531 - val_root_mean_squared_error: 478.0486\n",
      "Epoch 1030/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 278355.3125 - root_mean_squared_error: 527.5939 - val_loss: 235601.9375 - val_root_mean_squared_error: 485.3884\n",
      "Epoch 1031/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 264500.7188 - root_mean_squared_error: 514.2963 - val_loss: 229055.5781 - val_root_mean_squared_error: 478.5975\n",
      "Epoch 1032/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 290706.8750 - root_mean_squared_error: 539.1724 - val_loss: 238175.9375 - val_root_mean_squared_error: 488.0327\n",
      "Epoch 1033/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 222315.7031 - root_mean_squared_error: 471.5037 - val_loss: 264844.7188 - val_root_mean_squared_error: 514.6307\n",
      "Epoch 1034/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 264339.5312 - root_mean_squared_error: 514.1396 - val_loss: 224779.8125 - val_root_mean_squared_error: 474.1095\n",
      "Epoch 1035/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 249280.1250 - root_mean_squared_error: 499.2796 - val_loss: 225437.5312 - val_root_mean_squared_error: 474.8026\n",
      "Epoch 1036/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 290886.4062 - root_mean_squared_error: 539.3389 - val_loss: 221665.1406 - val_root_mean_squared_error: 470.8133\n",
      "Epoch 1037/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243158.8125 - root_mean_squared_error: 493.1114 - val_loss: 228290.6094 - val_root_mean_squared_error: 477.7977\n",
      "Epoch 1038/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249152.6250 - root_mean_squared_error: 499.1519 - val_loss: 220662.2500 - val_root_mean_squared_error: 469.7470\n",
      "Epoch 1039/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 264195.6250 - root_mean_squared_error: 513.9996 - val_loss: 237181.5938 - val_root_mean_squared_error: 487.0129\n",
      "Epoch 1040/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 279436.6875 - root_mean_squared_error: 528.6177 - val_loss: 195461.5625 - val_root_mean_squared_error: 442.1104\n",
      "Epoch 1041/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 272389.7812 - root_mean_squared_error: 521.9097 - val_loss: 215275.9531 - val_root_mean_squared_error: 463.9784\n",
      "Epoch 1042/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220786.6875 - root_mean_squared_error: 469.8794 - val_loss: 220514.2031 - val_root_mean_squared_error: 469.5894\n",
      "Epoch 1043/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 290509.4375 - root_mean_squared_error: 538.9893 - val_loss: 218567.0938 - val_root_mean_squared_error: 467.5116\n",
      "Epoch 1044/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 217403.7031 - root_mean_squared_error: 466.2657 - val_loss: 223140.4062 - val_root_mean_squared_error: 472.3774\n",
      "Epoch 1045/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 230122.5000 - root_mean_squared_error: 479.7108 - val_loss: 230250.7031 - val_root_mean_squared_error: 479.8445\n",
      "Epoch 1046/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 256598.9219 - root_mean_squared_error: 506.5559 - val_loss: 198311.2812 - val_root_mean_squared_error: 445.3215\n",
      "Epoch 1047/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 295816.0938 - root_mean_squared_error: 543.8898 - val_loss: 205415.6875 - val_root_mean_squared_error: 453.2281\n",
      "Epoch 1048/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 289673.3750 - root_mean_squared_error: 538.2131 - val_loss: 218774.5469 - val_root_mean_squared_error: 467.7334\n",
      "Epoch 1049/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 235415.0625 - root_mean_squared_error: 485.1959 - val_loss: 206455.1094 - val_root_mean_squared_error: 454.3733\n",
      "Epoch 1050/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 253559.0938 - root_mean_squared_error: 503.5465 - val_loss: 218264.4219 - val_root_mean_squared_error: 467.1878\n",
      "Epoch 1051/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 231915.5469 - root_mean_squared_error: 481.5761 - val_loss: 230765.6406 - val_root_mean_squared_error: 480.3807\n",
      "Epoch 1052/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 256924.7031 - root_mean_squared_error: 506.8774 - val_loss: 219482.6250 - val_root_mean_squared_error: 468.4897\n",
      "Epoch 1053/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251724.0938 - root_mean_squared_error: 501.7211 - val_loss: 219613.6875 - val_root_mean_squared_error: 468.6296\n",
      "Epoch 1054/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 251548.8125 - root_mean_squared_error: 501.5464 - val_loss: 227763.4219 - val_root_mean_squared_error: 477.2457\n",
      "Epoch 1055/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 226562.7500 - root_mean_squared_error: 475.9861 - val_loss: 226199.4375 - val_root_mean_squared_error: 475.6043\n",
      "Epoch 1056/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 244901.4062 - root_mean_squared_error: 494.8752 - val_loss: 210037.5469 - val_root_mean_squared_error: 458.2985\n",
      "Epoch 1057/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 258947.7969 - root_mean_squared_error: 508.8691 - val_loss: 227441.2188 - val_root_mean_squared_error: 476.9080\n",
      "Epoch 1058/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 270335.4375 - root_mean_squared_error: 519.9379 - val_loss: 196587.5781 - val_root_mean_squared_error: 443.3820\n",
      "Epoch 1059/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 240800.8125 - root_mean_squared_error: 490.7146 - val_loss: 220475.1719 - val_root_mean_squared_error: 469.5479\n",
      "Epoch 1060/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 239237.9062 - root_mean_squared_error: 489.1195 - val_loss: 225816.1250 - val_root_mean_squared_error: 475.2011\n",
      "Epoch 1061/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 267032.2500 - root_mean_squared_error: 516.7516 - val_loss: 209159.5625 - val_root_mean_squared_error: 457.3397\n",
      "Epoch 1062/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 271999.9688 - root_mean_squared_error: 521.5361 - val_loss: 229926.9688 - val_root_mean_squared_error: 479.5070\n",
      "Epoch 1063/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 281184.0938 - root_mean_squared_error: 530.2679 - val_loss: 221634.1719 - val_root_mean_squared_error: 470.7804\n",
      "Epoch 1064/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 238572.8438 - root_mean_squared_error: 488.4392 - val_loss: 209960.8750 - val_root_mean_squared_error: 458.2149\n",
      "Epoch 1065/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 284172.9062 - root_mean_squared_error: 533.0787 - val_loss: 228428.7500 - val_root_mean_squared_error: 477.9422\n",
      "Epoch 1066/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 232608.8750 - root_mean_squared_error: 482.2954 - val_loss: 216897.3750 - val_root_mean_squared_error: 465.7224\n",
      "Epoch 1067/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 283080.9688 - root_mean_squared_error: 532.0535 - val_loss: 218957.7500 - val_root_mean_squared_error: 467.9292\n",
      "Epoch 1068/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 235951.9531 - root_mean_squared_error: 485.7489 - val_loss: 208498.2656 - val_root_mean_squared_error: 456.6161\n",
      "Epoch 1069/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 230061.4375 - root_mean_squared_error: 479.6472 - val_loss: 225147.7500 - val_root_mean_squared_error: 474.4974\n",
      "Epoch 1070/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 237213.5312 - root_mean_squared_error: 487.0457 - val_loss: 223200.3125 - val_root_mean_squared_error: 472.4408\n",
      "Epoch 1071/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 235474.5469 - root_mean_squared_error: 485.2572 - val_loss: 198815.5312 - val_root_mean_squared_error: 445.8874\n",
      "Epoch 1072/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 267294.6562 - root_mean_squared_error: 517.0055 - val_loss: 205081.3125 - val_root_mean_squared_error: 452.8590\n",
      "Epoch 1073/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 270131.1562 - root_mean_squared_error: 519.7415 - val_loss: 208909.6719 - val_root_mean_squared_error: 457.0664\n",
      "Epoch 1074/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 225256.5469 - root_mean_squared_error: 474.6120 - val_loss: 205672.1719 - val_root_mean_squared_error: 453.5109\n",
      "Epoch 1075/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245892.1250 - root_mean_squared_error: 495.8751 - val_loss: 200080.8750 - val_root_mean_squared_error: 447.3040\n",
      "Epoch 1076/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 250524.9688 - root_mean_squared_error: 500.5247 - val_loss: 221107.0469 - val_root_mean_squared_error: 470.2202\n",
      "Epoch 1077/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 217653.5625 - root_mean_squared_error: 466.5336 - val_loss: 226715.5625 - val_root_mean_squared_error: 476.1466\n",
      "Epoch 1078/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 217258.6406 - root_mean_squared_error: 466.1101 - val_loss: 219196.5156 - val_root_mean_squared_error: 468.1843\n",
      "Epoch 1079/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 242037.7656 - root_mean_squared_error: 491.9733 - val_loss: 218640.0625 - val_root_mean_squared_error: 467.5896\n",
      "Epoch 1080/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 286629.7188 - root_mean_squared_error: 535.3781 - val_loss: 218794.0000 - val_root_mean_squared_error: 467.7542\n",
      "Epoch 1081/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247630.5938 - root_mean_squared_error: 497.6249 - val_loss: 190256.7188 - val_root_mean_squared_error: 436.1843\n",
      "Epoch 1082/1600\n",
      "13/13 [==============================] - 0s 10ms/step - loss: 261968.4688 - root_mean_squared_error: 511.8286 - val_loss: 209265.2344 - val_root_mean_squared_error: 457.4552\n",
      "Epoch 1083/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246596.2812 - root_mean_squared_error: 496.5846 - val_loss: 211220.3594 - val_root_mean_squared_error: 459.5872\n",
      "Epoch 1084/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 235562.5625 - root_mean_squared_error: 485.3479 - val_loss: 204779.5156 - val_root_mean_squared_error: 452.5257\n",
      "Epoch 1085/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248889.9688 - root_mean_squared_error: 498.8887 - val_loss: 211654.1406 - val_root_mean_squared_error: 460.0588\n",
      "Epoch 1086/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 275316.0312 - root_mean_squared_error: 524.7057 - val_loss: 237996.2344 - val_root_mean_squared_error: 487.8486\n",
      "Epoch 1087/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251917.1094 - root_mean_squared_error: 501.9135 - val_loss: 199992.7031 - val_root_mean_squared_error: 447.2054\n",
      "Epoch 1088/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 274782.7812 - root_mean_squared_error: 524.1973 - val_loss: 205497.5000 - val_root_mean_squared_error: 453.3183\n",
      "Epoch 1089/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 239827.9062 - root_mean_squared_error: 489.7223 - val_loss: 217507.9688 - val_root_mean_squared_error: 466.3775\n",
      "Epoch 1090/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244134.9062 - root_mean_squared_error: 494.1001 - val_loss: 209289.2969 - val_root_mean_squared_error: 457.4815\n",
      "Epoch 1091/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 271710.2188 - root_mean_squared_error: 521.2583 - val_loss: 219772.8281 - val_root_mean_squared_error: 468.7993\n",
      "Epoch 1092/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252278.8125 - root_mean_squared_error: 502.2737 - val_loss: 229739.3438 - val_root_mean_squared_error: 479.3113\n",
      "Epoch 1093/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 256563.5938 - root_mean_squared_error: 506.5211 - val_loss: 210432.8906 - val_root_mean_squared_error: 458.7296\n",
      "Epoch 1094/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248186.9062 - root_mean_squared_error: 498.1836 - val_loss: 230264.9531 - val_root_mean_squared_error: 479.8593\n",
      "Epoch 1095/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 226757.2188 - root_mean_squared_error: 476.1903 - val_loss: 232718.9844 - val_root_mean_squared_error: 482.4095\n",
      "Epoch 1096/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237801.9688 - root_mean_squared_error: 487.6494 - val_loss: 217714.5625 - val_root_mean_squared_error: 466.5989\n",
      "Epoch 1097/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 269459.5000 - root_mean_squared_error: 519.0949 - val_loss: 213093.5469 - val_root_mean_squared_error: 461.6206\n",
      "Epoch 1098/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 257604.1406 - root_mean_squared_error: 507.5472 - val_loss: 216888.0938 - val_root_mean_squared_error: 465.7125\n",
      "Epoch 1099/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248217.3594 - root_mean_squared_error: 498.2142 - val_loss: 209333.0469 - val_root_mean_squared_error: 457.5293\n",
      "Epoch 1100/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245254.6094 - root_mean_squared_error: 495.2319 - val_loss: 228848.8594 - val_root_mean_squared_error: 478.3815\n",
      "Epoch 1101/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 218339.4219 - root_mean_squared_error: 467.2680 - val_loss: 214254.6406 - val_root_mean_squared_error: 462.8765\n",
      "Epoch 1102/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 226463.3906 - root_mean_squared_error: 475.8817 - val_loss: 227408.5938 - val_root_mean_squared_error: 476.8738\n",
      "Epoch 1103/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 252833.1406 - root_mean_squared_error: 502.8252 - val_loss: 223049.0938 - val_root_mean_squared_error: 472.2807\n",
      "Epoch 1104/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245036.0938 - root_mean_squared_error: 495.0112 - val_loss: 226720.5938 - val_root_mean_squared_error: 476.1519\n",
      "Epoch 1105/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254940.3281 - root_mean_squared_error: 504.9162 - val_loss: 221163.7656 - val_root_mean_squared_error: 470.2805\n",
      "Epoch 1106/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246889.4531 - root_mean_squared_error: 496.8797 - val_loss: 217532.2344 - val_root_mean_squared_error: 466.4035\n",
      "Epoch 1107/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 283345.2500 - root_mean_squared_error: 532.3018 - val_loss: 223671.5312 - val_root_mean_squared_error: 472.9392\n",
      "Epoch 1108/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 270186.6250 - root_mean_squared_error: 519.7948 - val_loss: 220448.4531 - val_root_mean_squared_error: 469.5194\n",
      "Epoch 1109/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241443.4688 - root_mean_squared_error: 491.3690 - val_loss: 223230.4531 - val_root_mean_squared_error: 472.4727\n",
      "Epoch 1110/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 229846.2188 - root_mean_squared_error: 479.4228 - val_loss: 214554.6875 - val_root_mean_squared_error: 463.2005\n",
      "Epoch 1111/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 225065.2344 - root_mean_squared_error: 474.4104 - val_loss: 222126.5156 - val_root_mean_squared_error: 471.3030\n",
      "Epoch 1112/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262554.1875 - root_mean_squared_error: 512.4004 - val_loss: 224411.4062 - val_root_mean_squared_error: 473.7208\n",
      "Epoch 1113/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 244010.7031 - root_mean_squared_error: 493.9744 - val_loss: 219423.8281 - val_root_mean_squared_error: 468.4270\n",
      "Epoch 1114/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 258056.2656 - root_mean_squared_error: 507.9924 - val_loss: 218018.5781 - val_root_mean_squared_error: 466.9246\n",
      "Epoch 1115/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 257308.1562 - root_mean_squared_error: 507.2555 - val_loss: 248913.1875 - val_root_mean_squared_error: 498.9120\n",
      "Epoch 1116/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 248271.1406 - root_mean_squared_error: 498.2681 - val_loss: 240731.2188 - val_root_mean_squared_error: 490.6437\n",
      "Epoch 1117/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 261578.5312 - root_mean_squared_error: 511.4475 - val_loss: 233460.3906 - val_root_mean_squared_error: 483.1774\n",
      "Epoch 1118/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 222413.9062 - root_mean_squared_error: 471.6078 - val_loss: 251337.3906 - val_root_mean_squared_error: 501.3356\n",
      "Epoch 1119/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 224740.4062 - root_mean_squared_error: 474.0679 - val_loss: 240606.6406 - val_root_mean_squared_error: 490.5167\n",
      "Epoch 1120/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 229103.2969 - root_mean_squared_error: 478.6474 - val_loss: 221294.3594 - val_root_mean_squared_error: 470.4193\n",
      "Epoch 1121/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 218891.7188 - root_mean_squared_error: 467.8586 - val_loss: 234193.0156 - val_root_mean_squared_error: 483.9349\n",
      "Epoch 1122/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 280536.9688 - root_mean_squared_error: 529.6574 - val_loss: 239315.9219 - val_root_mean_squared_error: 489.1993\n",
      "Epoch 1123/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 229748.3750 - root_mean_squared_error: 479.3207 - val_loss: 235640.5312 - val_root_mean_squared_error: 485.4282\n",
      "Epoch 1124/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 235582.1250 - root_mean_squared_error: 485.3680 - val_loss: 239041.2656 - val_root_mean_squared_error: 488.9185\n",
      "Epoch 1125/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 251659.0625 - root_mean_squared_error: 501.6563 - val_loss: 227741.4219 - val_root_mean_squared_error: 477.2226\n",
      "Epoch 1126/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245733.0000 - root_mean_squared_error: 495.7146 - val_loss: 239147.5000 - val_root_mean_squared_error: 489.0271\n",
      "Epoch 1127/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 293119.9062 - root_mean_squared_error: 541.4055 - val_loss: 231122.4219 - val_root_mean_squared_error: 480.7519\n",
      "Epoch 1128/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247401.8438 - root_mean_squared_error: 497.3951 - val_loss: 219735.0625 - val_root_mean_squared_error: 468.7591\n",
      "Epoch 1129/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233424.2031 - root_mean_squared_error: 483.1400 - val_loss: 216618.7188 - val_root_mean_squared_error: 465.4232\n",
      "Epoch 1130/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 276769.5938 - root_mean_squared_error: 526.0890 - val_loss: 228641.8438 - val_root_mean_squared_error: 478.1651\n",
      "Epoch 1131/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 262282.9375 - root_mean_squared_error: 512.1357 - val_loss: 235314.0938 - val_root_mean_squared_error: 485.0918\n",
      "Epoch 1132/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 268951.8438 - root_mean_squared_error: 518.6057 - val_loss: 231058.6875 - val_root_mean_squared_error: 480.6856\n",
      "Epoch 1133/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 241739.6094 - root_mean_squared_error: 491.6702 - val_loss: 239907.7031 - val_root_mean_squared_error: 489.8037\n",
      "Epoch 1134/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239620.1562 - root_mean_squared_error: 489.5101 - val_loss: 232789.1875 - val_root_mean_squared_error: 482.4823\n",
      "Epoch 1135/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 270563.8750 - root_mean_squared_error: 520.1575 - val_loss: 239236.3125 - val_root_mean_squared_error: 489.1179\n",
      "Epoch 1136/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 238213.6250 - root_mean_squared_error: 488.0713 - val_loss: 233224.0938 - val_root_mean_squared_error: 482.9328\n",
      "Epoch 1137/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 266522.6250 - root_mean_squared_error: 516.2583 - val_loss: 231661.4062 - val_root_mean_squared_error: 481.3122\n",
      "Epoch 1138/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 260215.5625 - root_mean_squared_error: 510.1133 - val_loss: 225467.8594 - val_root_mean_squared_error: 474.8346\n",
      "Epoch 1139/1600\n",
      "13/13 [==============================] - 0s 10ms/step - loss: 210182.7344 - root_mean_squared_error: 458.4569 - val_loss: 251762.1250 - val_root_mean_squared_error: 501.7590\n",
      "Epoch 1140/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 240827.6406 - root_mean_squared_error: 490.7419 - val_loss: 233501.7656 - val_root_mean_squared_error: 483.2202\n",
      "Epoch 1141/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231599.2188 - root_mean_squared_error: 481.2476 - val_loss: 223045.2969 - val_root_mean_squared_error: 472.2767\n",
      "Epoch 1142/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 263895.9375 - root_mean_squared_error: 513.7080 - val_loss: 224011.0000 - val_root_mean_squared_error: 473.2980\n",
      "Epoch 1143/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252593.5156 - root_mean_squared_error: 502.5868 - val_loss: 213359.9844 - val_root_mean_squared_error: 461.9091\n",
      "Epoch 1144/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 236076.0625 - root_mean_squared_error: 485.8766 - val_loss: 221429.1094 - val_root_mean_squared_error: 470.5625\n",
      "Epoch 1145/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 217757.7031 - root_mean_squared_error: 466.6452 - val_loss: 218031.6250 - val_root_mean_squared_error: 466.9386\n",
      "Epoch 1146/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 217493.0781 - root_mean_squared_error: 466.3615 - val_loss: 218464.2500 - val_root_mean_squared_error: 467.4016\n",
      "Epoch 1147/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 222761.7969 - root_mean_squared_error: 471.9765 - val_loss: 227814.9062 - val_root_mean_squared_error: 477.2996\n",
      "Epoch 1148/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 205732.8750 - root_mean_squared_error: 453.5779 - val_loss: 239737.5000 - val_root_mean_squared_error: 489.6300\n",
      "Epoch 1149/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 255396.9375 - root_mean_squared_error: 505.3681 - val_loss: 227148.7969 - val_root_mean_squared_error: 476.6013\n",
      "Epoch 1150/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 267520.5938 - root_mean_squared_error: 517.2239 - val_loss: 226326.7188 - val_root_mean_squared_error: 475.7381\n",
      "Epoch 1151/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231613.4062 - root_mean_squared_error: 481.2623 - val_loss: 224045.1875 - val_root_mean_squared_error: 473.3341\n",
      "Epoch 1152/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 251835.1719 - root_mean_squared_error: 501.8318 - val_loss: 222377.3750 - val_root_mean_squared_error: 471.5691\n",
      "Epoch 1153/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 261840.7812 - root_mean_squared_error: 511.7038 - val_loss: 235001.2500 - val_root_mean_squared_error: 484.7693\n",
      "Epoch 1154/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 269242.9062 - root_mean_squared_error: 518.8862 - val_loss: 217945.7031 - val_root_mean_squared_error: 466.8466\n",
      "Epoch 1155/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231361.0000 - root_mean_squared_error: 481.0000 - val_loss: 232483.2656 - val_root_mean_squared_error: 482.1652\n",
      "Epoch 1156/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255671.7656 - root_mean_squared_error: 505.6400 - val_loss: 218561.0781 - val_root_mean_squared_error: 467.5052\n",
      "Epoch 1157/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243072.6719 - root_mean_squared_error: 493.0240 - val_loss: 216285.4688 - val_root_mean_squared_error: 465.0650\n",
      "Epoch 1158/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 268972.7500 - root_mean_squared_error: 518.6259 - val_loss: 232834.6562 - val_root_mean_squared_error: 482.5294\n",
      "Epoch 1159/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 213753.2031 - root_mean_squared_error: 462.3345 - val_loss: 229409.0469 - val_root_mean_squared_error: 478.9666\n",
      "Epoch 1160/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 239035.0000 - root_mean_squared_error: 488.9120 - val_loss: 228773.0781 - val_root_mean_squared_error: 478.3023\n",
      "Epoch 1161/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237218.9219 - root_mean_squared_error: 487.0512 - val_loss: 210641.8594 - val_root_mean_squared_error: 458.9574\n",
      "Epoch 1162/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 236803.8125 - root_mean_squared_error: 486.6249 - val_loss: 217600.4219 - val_root_mean_squared_error: 466.4766\n",
      "Epoch 1163/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 263569.5000 - root_mean_squared_error: 513.3902 - val_loss: 237622.4375 - val_root_mean_squared_error: 487.4653\n",
      "Epoch 1164/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 229390.3281 - root_mean_squared_error: 478.9471 - val_loss: 243310.6406 - val_root_mean_squared_error: 493.2653\n",
      "Epoch 1165/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 239802.2344 - root_mean_squared_error: 489.6961 - val_loss: 245469.7969 - val_root_mean_squared_error: 495.4491\n",
      "Epoch 1166/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 218264.2188 - root_mean_squared_error: 467.1876 - val_loss: 224176.8438 - val_root_mean_squared_error: 473.4732\n",
      "Epoch 1167/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 252292.0469 - root_mean_squared_error: 502.2868 - val_loss: 219018.7656 - val_root_mean_squared_error: 467.9944\n",
      "Epoch 1168/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243496.2188 - root_mean_squared_error: 493.4534 - val_loss: 234834.7656 - val_root_mean_squared_error: 484.5975\n",
      "Epoch 1169/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 238073.7500 - root_mean_squared_error: 487.9280 - val_loss: 224685.5156 - val_root_mean_squared_error: 474.0100\n",
      "Epoch 1170/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 231422.1875 - root_mean_squared_error: 481.0636 - val_loss: 216631.2969 - val_root_mean_squared_error: 465.4367\n",
      "Epoch 1171/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 245634.1719 - root_mean_squared_error: 495.6149 - val_loss: 227585.4844 - val_root_mean_squared_error: 477.0592\n",
      "Epoch 1172/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 252580.5625 - root_mean_squared_error: 502.5739 - val_loss: 241997.5156 - val_root_mean_squared_error: 491.9324\n",
      "Epoch 1173/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 256830.4531 - root_mean_squared_error: 506.7844 - val_loss: 226301.7656 - val_root_mean_squared_error: 475.7119\n",
      "Epoch 1174/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245483.5156 - root_mean_squared_error: 495.4629 - val_loss: 220185.6719 - val_root_mean_squared_error: 469.2395\n",
      "Epoch 1175/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 272355.4062 - root_mean_squared_error: 521.8768 - val_loss: 217800.2188 - val_root_mean_squared_error: 466.6907\n",
      "Epoch 1176/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246808.6250 - root_mean_squared_error: 496.7984 - val_loss: 228749.3125 - val_root_mean_squared_error: 478.2774\n",
      "Epoch 1177/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252262.9219 - root_mean_squared_error: 502.2578 - val_loss: 220561.6406 - val_root_mean_squared_error: 469.6399\n",
      "Epoch 1178/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 253274.2188 - root_mean_squared_error: 503.2636 - val_loss: 217268.2969 - val_root_mean_squared_error: 466.1205\n",
      "Epoch 1179/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 227192.7031 - root_mean_squared_error: 476.6474 - val_loss: 215384.0625 - val_root_mean_squared_error: 464.0949\n",
      "Epoch 1180/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 230950.0469 - root_mean_squared_error: 480.5726 - val_loss: 245194.3281 - val_root_mean_squared_error: 495.1710\n",
      "Epoch 1181/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 226636.3750 - root_mean_squared_error: 476.0634 - val_loss: 234271.0000 - val_root_mean_squared_error: 484.0155\n",
      "Epoch 1182/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 243973.5469 - root_mean_squared_error: 493.9368 - val_loss: 226788.2344 - val_root_mean_squared_error: 476.2229\n",
      "Epoch 1183/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 238155.5312 - root_mean_squared_error: 488.0118 - val_loss: 226686.3906 - val_root_mean_squared_error: 476.1159\n",
      "Epoch 1184/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 287949.1250 - root_mean_squared_error: 536.6089 - val_loss: 228477.8750 - val_root_mean_squared_error: 477.9936\n",
      "Epoch 1185/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 252631.2500 - root_mean_squared_error: 502.6244 - val_loss: 221355.0469 - val_root_mean_squared_error: 470.4839\n",
      "Epoch 1186/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247652.0625 - root_mean_squared_error: 497.6465 - val_loss: 224848.9375 - val_root_mean_squared_error: 474.1824\n",
      "Epoch 1187/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 220521.2969 - root_mean_squared_error: 469.5970 - val_loss: 235484.2656 - val_root_mean_squared_error: 485.2672\n",
      "Epoch 1188/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 255028.7656 - root_mean_squared_error: 505.0037 - val_loss: 244064.6094 - val_root_mean_squared_error: 494.0290\n",
      "Epoch 1189/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 222475.0938 - root_mean_squared_error: 471.6727 - val_loss: 241252.1719 - val_root_mean_squared_error: 491.1743\n",
      "Epoch 1190/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 258598.1875 - root_mean_squared_error: 508.5255 - val_loss: 222927.1562 - val_root_mean_squared_error: 472.1516\n",
      "Epoch 1191/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 231716.9531 - root_mean_squared_error: 481.3699 - val_loss: 223622.1250 - val_root_mean_squared_error: 472.8870\n",
      "Epoch 1192/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 237859.0156 - root_mean_squared_error: 487.7079 - val_loss: 220653.9062 - val_root_mean_squared_error: 469.7381\n",
      "Epoch 1193/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 243864.7969 - root_mean_squared_error: 493.8267 - val_loss: 253661.2656 - val_root_mean_squared_error: 503.6479\n",
      "Epoch 1194/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249016.0469 - root_mean_squared_error: 499.0151 - val_loss: 250705.9375 - val_root_mean_squared_error: 500.7054\n",
      "Epoch 1195/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 240339.7344 - root_mean_squared_error: 490.2446 - val_loss: 240078.0625 - val_root_mean_squared_error: 489.9776\n",
      "Epoch 1196/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 236207.8125 - root_mean_squared_error: 486.0121 - val_loss: 243565.2656 - val_root_mean_squared_error: 493.5233\n",
      "Epoch 1197/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231908.9688 - root_mean_squared_error: 481.5693 - val_loss: 247393.1562 - val_root_mean_squared_error: 497.3863\n",
      "Epoch 1198/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 264056.8438 - root_mean_squared_error: 513.8646 - val_loss: 250279.8281 - val_root_mean_squared_error: 500.2798\n",
      "Epoch 1199/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 199234.8594 - root_mean_squared_error: 446.3573 - val_loss: 239041.2656 - val_root_mean_squared_error: 488.9185\n",
      "Epoch 1200/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 259914.7031 - root_mean_squared_error: 509.8183 - val_loss: 250690.8125 - val_root_mean_squared_error: 500.6903\n",
      "Epoch 1201/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 283712.1250 - root_mean_squared_error: 532.6464 - val_loss: 250367.3906 - val_root_mean_squared_error: 500.3672\n",
      "Epoch 1202/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 229054.1562 - root_mean_squared_error: 478.5960 - val_loss: 229245.4062 - val_root_mean_squared_error: 478.7958\n",
      "Epoch 1203/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254256.3906 - root_mean_squared_error: 504.2384 - val_loss: 244551.4062 - val_root_mean_squared_error: 494.5214\n",
      "Epoch 1204/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 244064.2500 - root_mean_squared_error: 494.0286 - val_loss: 231699.6094 - val_root_mean_squared_error: 481.3519\n",
      "Epoch 1205/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 242040.1094 - root_mean_squared_error: 491.9757 - val_loss: 240902.5000 - val_root_mean_squared_error: 490.8182\n",
      "Epoch 1206/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 240007.4219 - root_mean_squared_error: 489.9055 - val_loss: 226737.7344 - val_root_mean_squared_error: 476.1699\n",
      "Epoch 1207/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 228651.8281 - root_mean_squared_error: 478.1755 - val_loss: 226100.0469 - val_root_mean_squared_error: 475.4998\n",
      "Epoch 1208/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 232111.1406 - root_mean_squared_error: 481.7791 - val_loss: 218131.9688 - val_root_mean_squared_error: 467.0460\n",
      "Epoch 1209/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 250182.2344 - root_mean_squared_error: 500.1822 - val_loss: 219311.3750 - val_root_mean_squared_error: 468.3069\n",
      "Epoch 1210/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 215347.6094 - root_mean_squared_error: 464.0556 - val_loss: 223189.8125 - val_root_mean_squared_error: 472.4297\n",
      "Epoch 1211/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 263589.3438 - root_mean_squared_error: 513.4095 - val_loss: 234463.0938 - val_root_mean_squared_error: 484.2139\n",
      "Epoch 1212/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 258165.1406 - root_mean_squared_error: 508.0995 - val_loss: 234371.7656 - val_root_mean_squared_error: 484.1196\n",
      "Epoch 1213/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 261124.5000 - root_mean_squared_error: 511.0034 - val_loss: 225102.3281 - val_root_mean_squared_error: 474.4495\n",
      "Epoch 1214/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 256144.8125 - root_mean_squared_error: 506.1075 - val_loss: 220235.2188 - val_root_mean_squared_error: 469.2923\n",
      "Epoch 1215/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 260673.4844 - root_mean_squared_error: 510.5619 - val_loss: 228746.5312 - val_root_mean_squared_error: 478.2745\n",
      "Epoch 1216/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249554.7344 - root_mean_squared_error: 499.5545 - val_loss: 222531.0625 - val_root_mean_squared_error: 471.7320\n",
      "Epoch 1217/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 255381.2188 - root_mean_squared_error: 505.3526 - val_loss: 240708.9219 - val_root_mean_squared_error: 490.6209\n",
      "Epoch 1218/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 224110.5469 - root_mean_squared_error: 473.4032 - val_loss: 252629.0625 - val_root_mean_squared_error: 502.6222\n",
      "Epoch 1219/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 246272.9219 - root_mean_squared_error: 496.2589 - val_loss: 256558.5000 - val_root_mean_squared_error: 506.5161\n",
      "Epoch 1220/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253827.5156 - root_mean_squared_error: 503.8130 - val_loss: 236080.9375 - val_root_mean_squared_error: 485.8816\n",
      "Epoch 1221/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 248131.5312 - root_mean_squared_error: 498.1280 - val_loss: 235676.8438 - val_root_mean_squared_error: 485.4656\n",
      "Epoch 1222/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 261568.5938 - root_mean_squared_error: 511.4378 - val_loss: 245125.8594 - val_root_mean_squared_error: 495.1019\n",
      "Epoch 1223/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 251830.9375 - root_mean_squared_error: 501.8276 - val_loss: 227374.2500 - val_root_mean_squared_error: 476.8378\n",
      "Epoch 1224/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 267573.6562 - root_mean_squared_error: 517.2752 - val_loss: 227261.9844 - val_root_mean_squared_error: 476.7200\n",
      "Epoch 1225/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 250846.8906 - root_mean_squared_error: 500.8462 - val_loss: 226164.1094 - val_root_mean_squared_error: 475.5671\n",
      "Epoch 1226/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239684.3438 - root_mean_squared_error: 489.5757 - val_loss: 212812.7812 - val_root_mean_squared_error: 461.3163\n",
      "Epoch 1227/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 261224.6719 - root_mean_squared_error: 511.1014 - val_loss: 238133.6094 - val_root_mean_squared_error: 487.9893\n",
      "Epoch 1228/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 274525.5000 - root_mean_squared_error: 523.9518 - val_loss: 210690.2969 - val_root_mean_squared_error: 459.0101\n",
      "Epoch 1229/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 228506.6094 - root_mean_squared_error: 478.0237 - val_loss: 230912.0000 - val_root_mean_squared_error: 480.5331\n",
      "Epoch 1230/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 220901.0156 - root_mean_squared_error: 470.0011 - val_loss: 244293.3281 - val_root_mean_squared_error: 494.2604\n",
      "Epoch 1231/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 233118.9219 - root_mean_squared_error: 482.8239 - val_loss: 234365.5625 - val_root_mean_squared_error: 484.1132\n",
      "Epoch 1232/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 240803.5469 - root_mean_squared_error: 490.7174 - val_loss: 235391.9219 - val_root_mean_squared_error: 485.1721\n",
      "Epoch 1233/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 212980.3906 - root_mean_squared_error: 461.4980 - val_loss: 226127.0000 - val_root_mean_squared_error: 475.5281\n",
      "Epoch 1234/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 219660.8594 - root_mean_squared_error: 468.6799 - val_loss: 242800.2969 - val_root_mean_squared_error: 492.7477\n",
      "Epoch 1235/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 266529.7812 - root_mean_squared_error: 516.2652 - val_loss: 231393.1719 - val_root_mean_squared_error: 481.0334\n",
      "Epoch 1236/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241319.7031 - root_mean_squared_error: 491.2430 - val_loss: 232118.4688 - val_root_mean_squared_error: 481.7867\n",
      "Epoch 1237/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253340.9375 - root_mean_squared_error: 503.3299 - val_loss: 208788.9531 - val_root_mean_squared_error: 456.9343\n",
      "Epoch 1238/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 230662.8125 - root_mean_squared_error: 480.2737 - val_loss: 216151.1094 - val_root_mean_squared_error: 464.9205\n",
      "Epoch 1239/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239764.7188 - root_mean_squared_error: 489.6577 - val_loss: 240107.1094 - val_root_mean_squared_error: 490.0073\n",
      "Epoch 1240/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233690.4219 - root_mean_squared_error: 483.4154 - val_loss: 238221.7344 - val_root_mean_squared_error: 488.0797\n",
      "Epoch 1241/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252906.9219 - root_mean_squared_error: 502.8985 - val_loss: 237488.7500 - val_root_mean_squared_error: 487.3282\n",
      "Epoch 1242/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 235154.8438 - root_mean_squared_error: 484.9277 - val_loss: 239676.0469 - val_root_mean_squared_error: 489.5672\n",
      "Epoch 1243/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 254284.4844 - root_mean_squared_error: 504.2663 - val_loss: 229547.6094 - val_root_mean_squared_error: 479.1113\n",
      "Epoch 1244/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241482.6562 - root_mean_squared_error: 491.4088 - val_loss: 232454.5469 - val_root_mean_squared_error: 482.1354\n",
      "Epoch 1245/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 238182.9219 - root_mean_squared_error: 488.0399 - val_loss: 228840.9844 - val_root_mean_squared_error: 478.3733\n",
      "Epoch 1246/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 256562.2031 - root_mean_squared_error: 506.5197 - val_loss: 225879.2656 - val_root_mean_squared_error: 475.2676\n",
      "Epoch 1247/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 219956.8750 - root_mean_squared_error: 468.9956 - val_loss: 242947.0469 - val_root_mean_squared_error: 492.8966\n",
      "Epoch 1248/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246564.3750 - root_mean_squared_error: 496.5525 - val_loss: 234895.1719 - val_root_mean_squared_error: 484.6599\n",
      "Epoch 1249/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 263085.3125 - root_mean_squared_error: 512.9185 - val_loss: 240998.1094 - val_root_mean_squared_error: 490.9156\n",
      "Epoch 1250/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 259716.3750 - root_mean_squared_error: 509.6237 - val_loss: 234951.5156 - val_root_mean_squared_error: 484.7180\n",
      "Epoch 1251/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 292463.5625 - root_mean_squared_error: 540.7990 - val_loss: 263059.3750 - val_root_mean_squared_error: 512.8931\n",
      "Epoch 1252/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 241042.5781 - root_mean_squared_error: 490.9609 - val_loss: 246896.7969 - val_root_mean_squared_error: 496.8871\n",
      "Epoch 1253/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 260364.1406 - root_mean_squared_error: 510.2589 - val_loss: 237225.3438 - val_root_mean_squared_error: 487.0578\n",
      "Epoch 1254/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 219959.1250 - root_mean_squared_error: 468.9980 - val_loss: 229764.3594 - val_root_mean_squared_error: 479.3374\n",
      "Epoch 1255/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251002.5156 - root_mean_squared_error: 501.0015 - val_loss: 236563.9531 - val_root_mean_squared_error: 486.3784\n",
      "Epoch 1256/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 251344.5312 - root_mean_squared_error: 501.3427 - val_loss: 225974.0625 - val_root_mean_squared_error: 475.3673\n",
      "Epoch 1257/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245938.9531 - root_mean_squared_error: 495.9223 - val_loss: 232900.5000 - val_root_mean_squared_error: 482.5977\n",
      "Epoch 1258/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 210003.8906 - root_mean_squared_error: 458.2618 - val_loss: 221916.7656 - val_root_mean_squared_error: 471.0804\n",
      "Epoch 1259/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 227729.0000 - root_mean_squared_error: 477.2096 - val_loss: 235455.0938 - val_root_mean_squared_error: 485.2372\n",
      "Epoch 1260/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 210747.9219 - root_mean_squared_error: 459.0729 - val_loss: 237430.2656 - val_root_mean_squared_error: 487.2682\n",
      "Epoch 1261/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254593.1562 - root_mean_squared_error: 504.5723 - val_loss: 244704.7969 - val_root_mean_squared_error: 494.6765\n",
      "Epoch 1262/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233974.3125 - root_mean_squared_error: 483.7089 - val_loss: 231288.6562 - val_root_mean_squared_error: 480.9248\n",
      "Epoch 1263/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 276073.1250 - root_mean_squared_error: 525.4266 - val_loss: 239624.8750 - val_root_mean_squared_error: 489.5150\n",
      "Epoch 1264/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244812.6562 - root_mean_squared_error: 494.7855 - val_loss: 225852.9219 - val_root_mean_squared_error: 475.2399\n",
      "Epoch 1265/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241110.3438 - root_mean_squared_error: 491.0299 - val_loss: 232268.8750 - val_root_mean_squared_error: 481.9428\n",
      "Epoch 1266/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 250114.8281 - root_mean_squared_error: 500.1148 - val_loss: 230736.0781 - val_root_mean_squared_error: 480.3499\n",
      "Epoch 1267/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 241959.7812 - root_mean_squared_error: 491.8941 - val_loss: 231695.3750 - val_root_mean_squared_error: 481.3474\n",
      "Epoch 1268/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 240009.3281 - root_mean_squared_error: 489.9075 - val_loss: 230732.6406 - val_root_mean_squared_error: 480.3464\n",
      "Epoch 1269/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 247327.0938 - root_mean_squared_error: 497.3199 - val_loss: 251193.0938 - val_root_mean_squared_error: 501.1917\n",
      "Epoch 1270/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241213.2656 - root_mean_squared_error: 491.1347 - val_loss: 243903.0000 - val_root_mean_squared_error: 493.8654\n",
      "Epoch 1271/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 236366.4062 - root_mean_squared_error: 486.1753 - val_loss: 247978.9219 - val_root_mean_squared_error: 497.9748\n",
      "Epoch 1272/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 252444.8906 - root_mean_squared_error: 502.4389 - val_loss: 249179.7656 - val_root_mean_squared_error: 499.1791\n",
      "Epoch 1273/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 257861.1719 - root_mean_squared_error: 507.8003 - val_loss: 247376.1875 - val_root_mean_squared_error: 497.3693\n",
      "Epoch 1274/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 218005.5625 - root_mean_squared_error: 466.9106 - val_loss: 230559.8438 - val_root_mean_squared_error: 480.1665\n",
      "Epoch 1275/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 229485.5156 - root_mean_squared_error: 479.0465 - val_loss: 246953.2031 - val_root_mean_squared_error: 496.9439\n",
      "Epoch 1276/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 230372.7812 - root_mean_squared_error: 479.9716 - val_loss: 232710.5156 - val_root_mean_squared_error: 482.4008\n",
      "Epoch 1277/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239469.0000 - root_mean_squared_error: 489.3557 - val_loss: 228388.9531 - val_root_mean_squared_error: 477.9006\n",
      "Epoch 1278/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 241802.6250 - root_mean_squared_error: 491.7343 - val_loss: 238799.7812 - val_root_mean_squared_error: 488.6714\n",
      "Epoch 1279/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 234573.2656 - root_mean_squared_error: 484.3276 - val_loss: 246314.8281 - val_root_mean_squared_error: 496.3011\n",
      "Epoch 1280/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 199168.4688 - root_mean_squared_error: 446.2830 - val_loss: 224673.7031 - val_root_mean_squared_error: 473.9976\n",
      "Epoch 1281/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 279665.8750 - root_mean_squared_error: 528.8345 - val_loss: 225373.5469 - val_root_mean_squared_error: 474.7352\n",
      "Epoch 1282/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 224280.1094 - root_mean_squared_error: 473.5822 - val_loss: 223388.0938 - val_root_mean_squared_error: 472.6395\n",
      "Epoch 1283/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251378.2031 - root_mean_squared_error: 501.3763 - val_loss: 215986.3750 - val_root_mean_squared_error: 464.7433\n",
      "Epoch 1284/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 272792.3750 - root_mean_squared_error: 522.2953 - val_loss: 234175.0938 - val_root_mean_squared_error: 483.9164\n",
      "Epoch 1285/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 234218.2031 - root_mean_squared_error: 483.9609 - val_loss: 222413.9531 - val_root_mean_squared_error: 471.6078\n",
      "Epoch 1286/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 252250.6406 - root_mean_squared_error: 502.2456 - val_loss: 219582.0000 - val_root_mean_squared_error: 468.5958\n",
      "Epoch 1287/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 257845.8750 - root_mean_squared_error: 507.7853 - val_loss: 218822.6875 - val_root_mean_squared_error: 467.7849\n",
      "Epoch 1288/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 235241.0156 - root_mean_squared_error: 485.0165 - val_loss: 241092.5312 - val_root_mean_squared_error: 491.0117\n",
      "Epoch 1289/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 224646.7656 - root_mean_squared_error: 473.9691 - val_loss: 231886.9531 - val_root_mean_squared_error: 481.5464\n",
      "Epoch 1290/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 214119.4688 - root_mean_squared_error: 462.7304 - val_loss: 242690.1562 - val_root_mean_squared_error: 492.6359\n",
      "Epoch 1291/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 228931.2500 - root_mean_squared_error: 478.4676 - val_loss: 236243.2344 - val_root_mean_squared_error: 486.0486\n",
      "Epoch 1292/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 235655.1406 - root_mean_squared_error: 485.4432 - val_loss: 228339.4688 - val_root_mean_squared_error: 477.8488\n",
      "Epoch 1293/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255011.3438 - root_mean_squared_error: 504.9865 - val_loss: 232746.2812 - val_root_mean_squared_error: 482.4379\n",
      "Epoch 1294/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 253920.0938 - root_mean_squared_error: 503.9048 - val_loss: 213409.9062 - val_root_mean_squared_error: 461.9631\n",
      "Epoch 1295/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248543.0469 - root_mean_squared_error: 498.5409 - val_loss: 212073.9375 - val_root_mean_squared_error: 460.5149\n",
      "Epoch 1296/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 232785.6875 - root_mean_squared_error: 482.4787 - val_loss: 217754.3125 - val_root_mean_squared_error: 466.6415\n",
      "Epoch 1297/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 278800.0938 - root_mean_squared_error: 528.0153 - val_loss: 215516.6562 - val_root_mean_squared_error: 464.2377\n",
      "Epoch 1298/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 226760.5938 - root_mean_squared_error: 476.1938 - val_loss: 208025.0000 - val_root_mean_squared_error: 456.0976\n",
      "Epoch 1299/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 259468.9375 - root_mean_squared_error: 509.3809 - val_loss: 201807.8594 - val_root_mean_squared_error: 449.2303\n",
      "Epoch 1300/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247833.9375 - root_mean_squared_error: 497.8292 - val_loss: 191737.8906 - val_root_mean_squared_error: 437.8788\n",
      "Epoch 1301/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 277614.1250 - root_mean_squared_error: 526.8910 - val_loss: 204863.0938 - val_root_mean_squared_error: 452.6180\n",
      "Epoch 1302/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 258677.8750 - root_mean_squared_error: 508.6039 - val_loss: 214275.4219 - val_root_mean_squared_error: 462.8989\n",
      "Epoch 1303/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245170.7188 - root_mean_squared_error: 495.1472 - val_loss: 224388.3281 - val_root_mean_squared_error: 473.6964\n",
      "Epoch 1304/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 234617.9062 - root_mean_squared_error: 484.3737 - val_loss: 209746.3281 - val_root_mean_squared_error: 457.9807\n",
      "Epoch 1305/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 223950.7031 - root_mean_squared_error: 473.2343 - val_loss: 211055.0000 - val_root_mean_squared_error: 459.4072\n",
      "Epoch 1306/1600\n",
      "13/13 [==============================] - 0s 10ms/step - loss: 269369.4062 - root_mean_squared_error: 519.0081 - val_loss: 209294.5156 - val_root_mean_squared_error: 457.4872\n",
      "Epoch 1307/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 245028.7812 - root_mean_squared_error: 495.0038 - val_loss: 214176.6719 - val_root_mean_squared_error: 462.7923\n",
      "Epoch 1308/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255654.7188 - root_mean_squared_error: 505.6231 - val_loss: 219022.0469 - val_root_mean_squared_error: 467.9979\n",
      "Epoch 1309/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247452.1562 - root_mean_squared_error: 497.4456 - val_loss: 208075.6094 - val_root_mean_squared_error: 456.1530\n",
      "Epoch 1310/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 237419.2344 - root_mean_squared_error: 487.2568 - val_loss: 216747.1562 - val_root_mean_squared_error: 465.5611\n",
      "Epoch 1311/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 236424.2344 - root_mean_squared_error: 486.2347 - val_loss: 213355.9219 - val_root_mean_squared_error: 461.9047\n",
      "Epoch 1312/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 278989.9062 - root_mean_squared_error: 528.1949 - val_loss: 218769.4688 - val_root_mean_squared_error: 467.7280\n",
      "Epoch 1313/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 229430.1562 - root_mean_squared_error: 478.9887 - val_loss: 218186.9688 - val_root_mean_squared_error: 467.1049\n",
      "Epoch 1314/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 277706.8125 - root_mean_squared_error: 526.9789 - val_loss: 226297.9688 - val_root_mean_squared_error: 475.7079\n",
      "Epoch 1315/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 253707.2500 - root_mean_squared_error: 503.6936 - val_loss: 213069.1406 - val_root_mean_squared_error: 461.5941\n",
      "Epoch 1316/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 217016.0312 - root_mean_squared_error: 465.8498 - val_loss: 226099.4219 - val_root_mean_squared_error: 475.4991\n",
      "Epoch 1317/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 246738.0156 - root_mean_squared_error: 496.7273 - val_loss: 215484.2969 - val_root_mean_squared_error: 464.2029\n",
      "Epoch 1318/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 228010.9531 - root_mean_squared_error: 477.5049 - val_loss: 216776.9531 - val_root_mean_squared_error: 465.5931\n",
      "Epoch 1319/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244015.1094 - root_mean_squared_error: 493.9789 - val_loss: 210885.0625 - val_root_mean_squared_error: 459.2222\n",
      "Epoch 1320/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253192.3906 - root_mean_squared_error: 503.1823 - val_loss: 218676.0469 - val_root_mean_squared_error: 467.6281\n",
      "Epoch 1321/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 280574.5312 - root_mean_squared_error: 529.6929 - val_loss: 223299.4531 - val_root_mean_squared_error: 472.5457\n",
      "Epoch 1322/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 235957.5000 - root_mean_squared_error: 485.7546 - val_loss: 222017.5156 - val_root_mean_squared_error: 471.1873\n",
      "Epoch 1323/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 271573.2500 - root_mean_squared_error: 521.1269 - val_loss: 220086.6406 - val_root_mean_squared_error: 469.1339\n",
      "Epoch 1324/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251773.0156 - root_mean_squared_error: 501.7699 - val_loss: 225987.2656 - val_root_mean_squared_error: 475.3812\n",
      "Epoch 1325/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 261962.1406 - root_mean_squared_error: 511.8224 - val_loss: 228724.9375 - val_root_mean_squared_error: 478.2520\n",
      "Epoch 1326/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 284125.6562 - root_mean_squared_error: 533.0344 - val_loss: 222131.0469 - val_root_mean_squared_error: 471.3078\n",
      "Epoch 1327/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 265933.1562 - root_mean_squared_error: 515.6871 - val_loss: 215435.3125 - val_root_mean_squared_error: 464.1501\n",
      "Epoch 1328/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 221283.9531 - root_mean_squared_error: 470.4083 - val_loss: 216539.7969 - val_root_mean_squared_error: 465.3384\n",
      "Epoch 1329/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237913.1094 - root_mean_squared_error: 487.7634 - val_loss: 221226.9688 - val_root_mean_squared_error: 470.3477\n",
      "Epoch 1330/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 269376.2500 - root_mean_squared_error: 519.0147 - val_loss: 222155.2500 - val_root_mean_squared_error: 471.3335\n",
      "Epoch 1331/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 234083.2500 - root_mean_squared_error: 483.8215 - val_loss: 245560.7344 - val_root_mean_squared_error: 495.5409\n",
      "Epoch 1332/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 272368.9375 - root_mean_squared_error: 521.8898 - val_loss: 217973.0312 - val_root_mean_squared_error: 466.8758\n",
      "Epoch 1333/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 241713.5312 - root_mean_squared_error: 491.6437 - val_loss: 219110.1562 - val_root_mean_squared_error: 468.0920\n",
      "Epoch 1334/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 235276.1719 - root_mean_squared_error: 485.0528 - val_loss: 210514.6094 - val_root_mean_squared_error: 458.8187\n",
      "Epoch 1335/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 260030.7344 - root_mean_squared_error: 509.9321 - val_loss: 212580.8281 - val_root_mean_squared_error: 461.0649\n",
      "Epoch 1336/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239031.6875 - root_mean_squared_error: 488.9087 - val_loss: 234335.5625 - val_root_mean_squared_error: 484.0822\n",
      "Epoch 1337/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 249720.3438 - root_mean_squared_error: 499.7203 - val_loss: 225692.4844 - val_root_mean_squared_error: 475.0710\n",
      "Epoch 1338/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 240459.4531 - root_mean_squared_error: 490.3666 - val_loss: 238826.8125 - val_root_mean_squared_error: 488.6991\n",
      "Epoch 1339/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233549.6250 - root_mean_squared_error: 483.2697 - val_loss: 222102.5000 - val_root_mean_squared_error: 471.2775\n",
      "Epoch 1340/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262218.2500 - root_mean_squared_error: 512.0725 - val_loss: 225463.3750 - val_root_mean_squared_error: 474.8298\n",
      "Epoch 1341/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 224694.8281 - root_mean_squared_error: 474.0198 - val_loss: 227237.8594 - val_root_mean_squared_error: 476.6947\n",
      "Epoch 1342/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 238770.8125 - root_mean_squared_error: 488.6418 - val_loss: 220648.9531 - val_root_mean_squared_error: 469.7328\n",
      "Epoch 1343/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 254410.8281 - root_mean_squared_error: 504.3915 - val_loss: 216963.9219 - val_root_mean_squared_error: 465.7939\n",
      "Epoch 1344/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 208884.9844 - root_mean_squared_error: 457.0394 - val_loss: 232026.4219 - val_root_mean_squared_error: 481.6912\n",
      "Epoch 1345/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 236368.8906 - root_mean_squared_error: 486.1778 - val_loss: 227435.0469 - val_root_mean_squared_error: 476.9015\n",
      "Epoch 1346/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249235.5625 - root_mean_squared_error: 499.2350 - val_loss: 221584.0000 - val_root_mean_squared_error: 470.7271\n",
      "Epoch 1347/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 253384.0000 - root_mean_squared_error: 503.3726 - val_loss: 219905.1250 - val_root_mean_squared_error: 468.9404\n",
      "Epoch 1348/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 247927.5312 - root_mean_squared_error: 497.9232 - val_loss: 214288.8906 - val_root_mean_squared_error: 462.9135\n",
      "Epoch 1349/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 221532.2188 - root_mean_squared_error: 470.6721 - val_loss: 212372.0625 - val_root_mean_squared_error: 460.8384\n",
      "Epoch 1350/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 228145.3281 - root_mean_squared_error: 477.6456 - val_loss: 218011.7344 - val_root_mean_squared_error: 466.9173\n",
      "Epoch 1351/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 263960.9375 - root_mean_squared_error: 513.7713 - val_loss: 206160.6875 - val_root_mean_squared_error: 454.0492\n",
      "Epoch 1352/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 216346.8438 - root_mean_squared_error: 465.1310 - val_loss: 207558.1875 - val_root_mean_squared_error: 455.5855\n",
      "Epoch 1353/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 260748.5469 - root_mean_squared_error: 510.6354 - val_loss: 190206.5625 - val_root_mean_squared_error: 436.1268\n",
      "Epoch 1354/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 257341.3438 - root_mean_squared_error: 507.2882 - val_loss: 203462.2656 - val_root_mean_squared_error: 451.0679\n",
      "Epoch 1355/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 246637.1875 - root_mean_squared_error: 496.6258 - val_loss: 194932.4531 - val_root_mean_squared_error: 441.5116\n",
      "Epoch 1356/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 236128.1094 - root_mean_squared_error: 485.9301 - val_loss: 198200.1250 - val_root_mean_squared_error: 445.1967\n",
      "Epoch 1357/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245184.4062 - root_mean_squared_error: 495.1610 - val_loss: 196747.8125 - val_root_mean_squared_error: 443.5626\n",
      "Epoch 1358/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 245823.0625 - root_mean_squared_error: 495.8055 - val_loss: 196982.0781 - val_root_mean_squared_error: 443.8266\n",
      "Epoch 1359/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 287536.9375 - root_mean_squared_error: 536.2247 - val_loss: 194959.3438 - val_root_mean_squared_error: 441.5420\n",
      "Epoch 1360/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 228776.0938 - root_mean_squared_error: 478.3055 - val_loss: 184637.8750 - val_root_mean_squared_error: 429.6951\n",
      "Epoch 1361/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246435.8594 - root_mean_squared_error: 496.4231 - val_loss: 189059.8750 - val_root_mean_squared_error: 434.8102\n",
      "Epoch 1362/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247091.8594 - root_mean_squared_error: 497.0833 - val_loss: 185890.6719 - val_root_mean_squared_error: 431.1504\n",
      "Epoch 1363/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249820.5000 - root_mean_squared_error: 499.8205 - val_loss: 200614.4531 - val_root_mean_squared_error: 447.9001\n",
      "Epoch 1364/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 262286.9062 - root_mean_squared_error: 512.1395 - val_loss: 201566.1406 - val_root_mean_squared_error: 448.9612\n",
      "Epoch 1365/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 240900.1094 - root_mean_squared_error: 490.8158 - val_loss: 212660.2969 - val_root_mean_squared_error: 461.1511\n",
      "Epoch 1366/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220401.2031 - root_mean_squared_error: 469.4691 - val_loss: 214086.7500 - val_root_mean_squared_error: 462.6951\n",
      "Epoch 1367/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233469.3750 - root_mean_squared_error: 483.1867 - val_loss: 195753.1094 - val_root_mean_squared_error: 442.4399\n",
      "Epoch 1368/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 239490.9375 - root_mean_squared_error: 489.3781 - val_loss: 185732.5625 - val_root_mean_squared_error: 430.9670\n",
      "Epoch 1369/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 263416.7812 - root_mean_squared_error: 513.2415 - val_loss: 201373.2656 - val_root_mean_squared_error: 448.7463\n",
      "Epoch 1370/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 210671.3906 - root_mean_squared_error: 458.9895 - val_loss: 204524.0000 - val_root_mean_squared_error: 452.2433\n",
      "Epoch 1371/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 229185.3125 - root_mean_squared_error: 478.7330 - val_loss: 203896.5312 - val_root_mean_squared_error: 451.5490\n",
      "Epoch 1372/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233360.5781 - root_mean_squared_error: 483.0741 - val_loss: 200512.2969 - val_root_mean_squared_error: 447.7860\n",
      "Epoch 1373/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 240919.2500 - root_mean_squared_error: 490.8353 - val_loss: 202164.3281 - val_root_mean_squared_error: 449.6269\n",
      "Epoch 1374/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 284337.1875 - root_mean_squared_error: 533.2328 - val_loss: 204566.7188 - val_root_mean_squared_error: 452.2905\n",
      "Epoch 1375/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 221208.2500 - root_mean_squared_error: 470.3278 - val_loss: 212941.9062 - val_root_mean_squared_error: 461.4563\n",
      "Epoch 1376/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 214003.9062 - root_mean_squared_error: 462.6056 - val_loss: 214700.2656 - val_root_mean_squared_error: 463.3576\n",
      "Epoch 1377/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 262963.1875 - root_mean_squared_error: 512.7994 - val_loss: 228256.5938 - val_root_mean_squared_error: 477.7621\n",
      "Epoch 1378/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 266984.7812 - root_mean_squared_error: 516.7057 - val_loss: 223774.7812 - val_root_mean_squared_error: 473.0484\n",
      "Epoch 1379/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 260700.1250 - root_mean_squared_error: 510.5880 - val_loss: 217679.3906 - val_root_mean_squared_error: 466.5612\n",
      "Epoch 1380/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 237601.9375 - root_mean_squared_error: 487.4443 - val_loss: 224893.4375 - val_root_mean_squared_error: 474.2293\n",
      "Epoch 1381/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 253085.0625 - root_mean_squared_error: 503.0756 - val_loss: 221554.6719 - val_root_mean_squared_error: 470.6960\n",
      "Epoch 1382/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 256426.4062 - root_mean_squared_error: 506.3856 - val_loss: 220461.9531 - val_root_mean_squared_error: 469.5338\n",
      "Epoch 1383/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 268419.3125 - root_mean_squared_error: 518.0920 - val_loss: 226220.6562 - val_root_mean_squared_error: 475.6266\n",
      "Epoch 1384/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 253327.7969 - root_mean_squared_error: 503.3168 - val_loss: 234994.3750 - val_root_mean_squared_error: 484.7622\n",
      "Epoch 1385/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 231588.6250 - root_mean_squared_error: 481.2366 - val_loss: 236045.7969 - val_root_mean_squared_error: 485.8455\n",
      "Epoch 1386/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 203049.0000 - root_mean_squared_error: 450.6096 - val_loss: 222182.6250 - val_root_mean_squared_error: 471.3625\n",
      "Epoch 1387/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 227047.3594 - root_mean_squared_error: 476.4949 - val_loss: 222703.0156 - val_root_mean_squared_error: 471.9142\n",
      "Epoch 1388/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243296.9219 - root_mean_squared_error: 493.2514 - val_loss: 222714.3438 - val_root_mean_squared_error: 471.9262\n",
      "Epoch 1389/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 207016.4688 - root_mean_squared_error: 454.9906 - val_loss: 231235.5469 - val_root_mean_squared_error: 480.8696\n",
      "Epoch 1390/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231138.9375 - root_mean_squared_error: 480.7691 - val_loss: 222345.1875 - val_root_mean_squared_error: 471.5349\n",
      "Epoch 1391/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 226090.5625 - root_mean_squared_error: 475.4898 - val_loss: 226917.2500 - val_root_mean_squared_error: 476.3583\n",
      "Epoch 1392/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 268686.3125 - root_mean_squared_error: 518.3496 - val_loss: 213669.5938 - val_root_mean_squared_error: 462.2441\n",
      "Epoch 1393/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233087.5625 - root_mean_squared_error: 482.7914 - val_loss: 207508.4688 - val_root_mean_squared_error: 455.5310\n",
      "Epoch 1394/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 211390.5781 - root_mean_squared_error: 459.7723 - val_loss: 212450.2656 - val_root_mean_squared_error: 460.9233\n",
      "Epoch 1395/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 236886.7344 - root_mean_squared_error: 486.7101 - val_loss: 215505.2031 - val_root_mean_squared_error: 464.2254\n",
      "Epoch 1396/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 207379.5938 - root_mean_squared_error: 455.3895 - val_loss: 215724.2969 - val_root_mean_squared_error: 464.4613\n",
      "Epoch 1397/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 228428.0625 - root_mean_squared_error: 477.9415 - val_loss: 217028.5469 - val_root_mean_squared_error: 465.8632\n",
      "Epoch 1398/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 234632.2344 - root_mean_squared_error: 484.3885 - val_loss: 208642.9219 - val_root_mean_squared_error: 456.7745\n",
      "Epoch 1399/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 254950.0938 - root_mean_squared_error: 504.9258 - val_loss: 213189.4219 - val_root_mean_squared_error: 461.7244\n",
      "Epoch 1400/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 230285.3281 - root_mean_squared_error: 479.8805 - val_loss: 214005.9219 - val_root_mean_squared_error: 462.6077\n",
      "Epoch 1401/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 218715.7656 - root_mean_squared_error: 467.6706 - val_loss: 219161.3750 - val_root_mean_squared_error: 468.1468\n",
      "Epoch 1402/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 258465.3594 - root_mean_squared_error: 508.3949 - val_loss: 235089.8438 - val_root_mean_squared_error: 484.8607\n",
      "Epoch 1403/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 251628.5781 - root_mean_squared_error: 501.6259 - val_loss: 230636.7500 - val_root_mean_squared_error: 480.2466\n",
      "Epoch 1404/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 206979.5938 - root_mean_squared_error: 454.9501 - val_loss: 228425.2500 - val_root_mean_squared_error: 477.9385\n",
      "Epoch 1405/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249671.2656 - root_mean_squared_error: 499.6711 - val_loss: 233291.1875 - val_root_mean_squared_error: 483.0023\n",
      "Epoch 1406/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 262982.6875 - root_mean_squared_error: 512.8184 - val_loss: 224184.1719 - val_root_mean_squared_error: 473.4809\n",
      "Epoch 1407/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233963.5156 - root_mean_squared_error: 483.6978 - val_loss: 226328.0625 - val_root_mean_squared_error: 475.7395\n",
      "Epoch 1408/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 218936.9062 - root_mean_squared_error: 467.9070 - val_loss: 215076.6094 - val_root_mean_squared_error: 463.7635\n",
      "Epoch 1409/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 269254.0625 - root_mean_squared_error: 518.8970 - val_loss: 225222.5000 - val_root_mean_squared_error: 474.5761\n",
      "Epoch 1410/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247457.4531 - root_mean_squared_error: 497.4510 - val_loss: 222906.2031 - val_root_mean_squared_error: 472.1294\n",
      "Epoch 1411/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 242192.2188 - root_mean_squared_error: 492.1303 - val_loss: 217317.5469 - val_root_mean_squared_error: 466.1733\n",
      "Epoch 1412/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262838.6875 - root_mean_squared_error: 512.6780 - val_loss: 218805.0938 - val_root_mean_squared_error: 467.7661\n",
      "Epoch 1413/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253272.8906 - root_mean_squared_error: 503.2622 - val_loss: 235489.3281 - val_root_mean_squared_error: 485.2724\n",
      "Epoch 1414/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241837.6250 - root_mean_squared_error: 491.7699 - val_loss: 238221.8750 - val_root_mean_squared_error: 488.0798\n",
      "Epoch 1415/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 211241.8438 - root_mean_squared_error: 459.6105 - val_loss: 232254.1250 - val_root_mean_squared_error: 481.9275\n",
      "Epoch 1416/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 219550.0625 - root_mean_squared_error: 468.5617 - val_loss: 227704.6094 - val_root_mean_squared_error: 477.1841\n",
      "Epoch 1417/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 229168.1562 - root_mean_squared_error: 478.7151 - val_loss: 221545.1406 - val_root_mean_squared_error: 470.6858\n",
      "Epoch 1418/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 217242.9531 - root_mean_squared_error: 466.0933 - val_loss: 222812.4844 - val_root_mean_squared_error: 472.0302\n",
      "Epoch 1419/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254767.5000 - root_mean_squared_error: 504.7450 - val_loss: 223441.9688 - val_root_mean_squared_error: 472.6965\n",
      "Epoch 1420/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 237538.7188 - root_mean_squared_error: 487.3794 - val_loss: 220777.9375 - val_root_mean_squared_error: 469.8701\n",
      "Epoch 1421/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 225006.1719 - root_mean_squared_error: 474.3481 - val_loss: 233540.8906 - val_root_mean_squared_error: 483.2607\n",
      "Epoch 1422/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 275188.4375 - root_mean_squared_error: 524.5840 - val_loss: 239897.6406 - val_root_mean_squared_error: 489.7935\n",
      "Epoch 1423/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 228065.0312 - root_mean_squared_error: 477.5616 - val_loss: 233286.6250 - val_root_mean_squared_error: 482.9975\n",
      "Epoch 1424/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 209452.7656 - root_mean_squared_error: 457.6601 - val_loss: 248307.3125 - val_root_mean_squared_error: 498.3044\n",
      "Epoch 1425/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 276128.7188 - root_mean_squared_error: 525.4795 - val_loss: 234786.4844 - val_root_mean_squared_error: 484.5477\n",
      "Epoch 1426/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233510.5625 - root_mean_squared_error: 483.2293 - val_loss: 227004.5156 - val_root_mean_squared_error: 476.4499\n",
      "Epoch 1427/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 240272.2031 - root_mean_squared_error: 490.1757 - val_loss: 233376.3906 - val_root_mean_squared_error: 483.0905\n",
      "Epoch 1428/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 286228.6875 - root_mean_squared_error: 535.0034 - val_loss: 232137.9688 - val_root_mean_squared_error: 481.8070\n",
      "Epoch 1429/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 199760.8438 - root_mean_squared_error: 446.9461 - val_loss: 235305.6406 - val_root_mean_squared_error: 485.0831\n",
      "Epoch 1430/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 267535.6250 - root_mean_squared_error: 517.2385 - val_loss: 233239.5781 - val_root_mean_squared_error: 482.9488\n",
      "Epoch 1431/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 255959.1562 - root_mean_squared_error: 505.9241 - val_loss: 225614.1250 - val_root_mean_squared_error: 474.9886\n",
      "Epoch 1432/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 250446.3281 - root_mean_squared_error: 500.4461 - val_loss: 214105.7188 - val_root_mean_squared_error: 462.7156\n",
      "Epoch 1433/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 261615.2969 - root_mean_squared_error: 511.4834 - val_loss: 216479.1406 - val_root_mean_squared_error: 465.2732\n",
      "Epoch 1434/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 250360.9531 - root_mean_squared_error: 500.3608 - val_loss: 215963.9219 - val_root_mean_squared_error: 464.7192\n",
      "Epoch 1435/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 244495.9844 - root_mean_squared_error: 494.4654 - val_loss: 219100.2344 - val_root_mean_squared_error: 468.0815\n",
      "Epoch 1436/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 250751.3281 - root_mean_squared_error: 500.7508 - val_loss: 226832.2812 - val_root_mean_squared_error: 476.2691\n",
      "Epoch 1437/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233301.5312 - root_mean_squared_error: 483.0130 - val_loss: 228075.6094 - val_root_mean_squared_error: 477.5726\n",
      "Epoch 1438/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244413.5312 - root_mean_squared_error: 494.3820 - val_loss: 236249.8125 - val_root_mean_squared_error: 486.0554\n",
      "Epoch 1439/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245328.2344 - root_mean_squared_error: 495.3062 - val_loss: 238661.6875 - val_root_mean_squared_error: 488.5301\n",
      "Epoch 1440/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 259986.3281 - root_mean_squared_error: 509.8885 - val_loss: 244917.1719 - val_root_mean_squared_error: 494.8911\n",
      "Epoch 1441/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 217213.9844 - root_mean_squared_error: 466.0622 - val_loss: 229507.8750 - val_root_mean_squared_error: 479.0698\n",
      "Epoch 1442/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 231658.2031 - root_mean_squared_error: 481.3088 - val_loss: 236872.7656 - val_root_mean_squared_error: 486.6958\n",
      "Epoch 1443/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 235558.7656 - root_mean_squared_error: 485.3440 - val_loss: 218186.0312 - val_root_mean_squared_error: 467.1039\n",
      "Epoch 1444/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249526.8594 - root_mean_squared_error: 499.5266 - val_loss: 223651.6406 - val_root_mean_squared_error: 472.9182\n",
      "Epoch 1445/1600\n",
      "13/13 [==============================] - 0s 11ms/step - loss: 240392.6250 - root_mean_squared_error: 490.2985 - val_loss: 230540.0938 - val_root_mean_squared_error: 480.1459\n",
      "Epoch 1446/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 264454.5312 - root_mean_squared_error: 514.2514 - val_loss: 224534.8438 - val_root_mean_squared_error: 473.8511\n",
      "Epoch 1447/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253312.0000 - root_mean_squared_error: 503.3011 - val_loss: 208261.9844 - val_root_mean_squared_error: 456.3573\n",
      "Epoch 1448/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 247468.9219 - root_mean_squared_error: 497.4625 - val_loss: 202863.0156 - val_root_mean_squared_error: 450.4032\n",
      "Epoch 1449/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 225159.7656 - root_mean_squared_error: 474.5100 - val_loss: 202794.0312 - val_root_mean_squared_error: 450.3266\n",
      "Epoch 1450/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 230918.7969 - root_mean_squared_error: 480.5401 - val_loss: 211569.2188 - val_root_mean_squared_error: 459.9666\n",
      "Epoch 1451/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 231892.4219 - root_mean_squared_error: 481.5521 - val_loss: 209399.1250 - val_root_mean_squared_error: 457.6015\n",
      "Epoch 1452/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 256189.0156 - root_mean_squared_error: 506.1512 - val_loss: 205632.4688 - val_root_mean_squared_error: 453.4672\n",
      "Epoch 1453/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 254083.5000 - root_mean_squared_error: 504.0670 - val_loss: 222319.9688 - val_root_mean_squared_error: 471.5082\n",
      "Epoch 1454/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 214033.7656 - root_mean_squared_error: 462.6378 - val_loss: 229904.3750 - val_root_mean_squared_error: 479.4835\n",
      "Epoch 1455/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 222376.5469 - root_mean_squared_error: 471.5682 - val_loss: 225068.1250 - val_root_mean_squared_error: 474.4135\n",
      "Epoch 1456/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 205930.6406 - root_mean_squared_error: 453.7958 - val_loss: 223894.5000 - val_root_mean_squared_error: 473.1749\n",
      "Epoch 1457/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237274.7344 - root_mean_squared_error: 487.1086 - val_loss: 231475.5469 - val_root_mean_squared_error: 481.1190\n",
      "Epoch 1458/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 257237.2031 - root_mean_squared_error: 507.1856 - val_loss: 225631.2969 - val_root_mean_squared_error: 475.0066\n",
      "Epoch 1459/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 289234.9062 - root_mean_squared_error: 537.8057 - val_loss: 206077.3125 - val_root_mean_squared_error: 453.9574\n",
      "Epoch 1460/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 278475.0312 - root_mean_squared_error: 527.7073 - val_loss: 215114.9219 - val_root_mean_squared_error: 463.8048\n",
      "Epoch 1461/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 241503.0000 - root_mean_squared_error: 491.4295 - val_loss: 202445.0625 - val_root_mean_squared_error: 449.9390\n",
      "Epoch 1462/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249976.6719 - root_mean_squared_error: 499.9767 - val_loss: 205119.1250 - val_root_mean_squared_error: 452.9008\n",
      "Epoch 1463/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 218546.3750 - root_mean_squared_error: 467.4894 - val_loss: 207792.0469 - val_root_mean_squared_error: 455.8421\n",
      "Epoch 1464/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255922.5625 - root_mean_squared_error: 505.8879 - val_loss: 220718.3906 - val_root_mean_squared_error: 469.8068\n",
      "Epoch 1465/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 223193.7812 - root_mean_squared_error: 472.4339 - val_loss: 234111.3594 - val_root_mean_squared_error: 483.8506\n",
      "Epoch 1466/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 212815.2812 - root_mean_squared_error: 461.3191 - val_loss: 218879.8125 - val_root_mean_squared_error: 467.8459\n",
      "Epoch 1467/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 255065.7188 - root_mean_squared_error: 505.0403 - val_loss: 215323.1719 - val_root_mean_squared_error: 464.0293\n",
      "Epoch 1468/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 265017.5000 - root_mean_squared_error: 514.7985 - val_loss: 200732.0781 - val_root_mean_squared_error: 448.0313\n",
      "Epoch 1469/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 277555.6562 - root_mean_squared_error: 526.8355 - val_loss: 211041.7344 - val_root_mean_squared_error: 459.3928\n",
      "Epoch 1470/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248664.3438 - root_mean_squared_error: 498.6626 - val_loss: 211004.4375 - val_root_mean_squared_error: 459.3522\n",
      "Epoch 1471/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 253053.0469 - root_mean_squared_error: 503.0438 - val_loss: 205599.0312 - val_root_mean_squared_error: 453.4303\n",
      "Epoch 1472/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 279387.4688 - root_mean_squared_error: 528.5712 - val_loss: 229661.6250 - val_root_mean_squared_error: 479.2303\n",
      "Epoch 1473/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 254570.0312 - root_mean_squared_error: 504.5493 - val_loss: 231430.7656 - val_root_mean_squared_error: 481.0725\n",
      "Epoch 1474/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 236765.5156 - root_mean_squared_error: 486.5856 - val_loss: 218933.5312 - val_root_mean_squared_error: 467.9033\n",
      "Epoch 1475/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 234037.4219 - root_mean_squared_error: 483.7741 - val_loss: 216390.5000 - val_root_mean_squared_error: 465.1779\n",
      "Epoch 1476/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 233356.3438 - root_mean_squared_error: 483.0697 - val_loss: 227698.2656 - val_root_mean_squared_error: 477.1774\n",
      "Epoch 1477/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 224817.5938 - root_mean_squared_error: 474.1493 - val_loss: 224172.1562 - val_root_mean_squared_error: 473.4682\n",
      "Epoch 1478/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244142.9375 - root_mean_squared_error: 494.1082 - val_loss: 224215.1562 - val_root_mean_squared_error: 473.5136\n",
      "Epoch 1479/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 219015.3438 - root_mean_squared_error: 467.9908 - val_loss: 220578.4219 - val_root_mean_squared_error: 469.6578\n",
      "Epoch 1480/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 246419.9375 - root_mean_squared_error: 496.4070 - val_loss: 222505.2344 - val_root_mean_squared_error: 471.7046\n",
      "Epoch 1481/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 255069.4688 - root_mean_squared_error: 505.0440 - val_loss: 226243.9062 - val_root_mean_squared_error: 475.6510\n",
      "Epoch 1482/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 229442.7031 - root_mean_squared_error: 479.0018 - val_loss: 208752.9531 - val_root_mean_squared_error: 456.8949\n",
      "Epoch 1483/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 238837.1719 - root_mean_squared_error: 488.7097 - val_loss: 202257.1406 - val_root_mean_squared_error: 449.7301\n",
      "Epoch 1484/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 217450.2500 - root_mean_squared_error: 466.3156 - val_loss: 211529.2500 - val_root_mean_squared_error: 459.9231\n",
      "Epoch 1485/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 236037.4062 - root_mean_squared_error: 485.8368 - val_loss: 220271.3750 - val_root_mean_squared_error: 469.3308\n",
      "Epoch 1486/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220738.1562 - root_mean_squared_error: 469.8278 - val_loss: 215457.7500 - val_root_mean_squared_error: 464.1743\n",
      "Epoch 1487/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 223093.9375 - root_mean_squared_error: 472.3282 - val_loss: 222252.5781 - val_root_mean_squared_error: 471.4367\n",
      "Epoch 1488/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 215398.9688 - root_mean_squared_error: 464.1109 - val_loss: 235459.0938 - val_root_mean_squared_error: 485.2413\n",
      "Epoch 1489/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 236831.7344 - root_mean_squared_error: 486.6536 - val_loss: 232513.1094 - val_root_mean_squared_error: 482.1961\n",
      "Epoch 1490/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 224575.2500 - root_mean_squared_error: 473.8937 - val_loss: 234259.1562 - val_root_mean_squared_error: 484.0033\n",
      "Epoch 1491/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 265031.2812 - root_mean_squared_error: 514.8119 - val_loss: 241298.0469 - val_root_mean_squared_error: 491.2210\n",
      "Epoch 1492/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 238604.1719 - root_mean_squared_error: 488.4713 - val_loss: 233008.4688 - val_root_mean_squared_error: 482.7095\n",
      "Epoch 1493/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 211525.0781 - root_mean_squared_error: 459.9185 - val_loss: 213912.8906 - val_root_mean_squared_error: 462.5072\n",
      "Epoch 1494/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 242395.4688 - root_mean_squared_error: 492.3367 - val_loss: 210340.0781 - val_root_mean_squared_error: 458.6285\n",
      "Epoch 1495/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 208971.6406 - root_mean_squared_error: 457.1342 - val_loss: 214795.5781 - val_root_mean_squared_error: 463.4604\n",
      "Epoch 1496/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233194.0156 - root_mean_squared_error: 482.9017 - val_loss: 209472.4219 - val_root_mean_squared_error: 457.6816\n",
      "Epoch 1497/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 246983.7344 - root_mean_squared_error: 496.9746 - val_loss: 200887.0781 - val_root_mean_squared_error: 448.2043\n",
      "Epoch 1498/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 260921.8906 - root_mean_squared_error: 510.8051 - val_loss: 203773.2812 - val_root_mean_squared_error: 451.4125\n",
      "Epoch 1499/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 267174.0625 - root_mean_squared_error: 516.8889 - val_loss: 199233.8438 - val_root_mean_squared_error: 446.3562\n",
      "Epoch 1500/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 210961.3750 - root_mean_squared_error: 459.3053 - val_loss: 220048.6562 - val_root_mean_squared_error: 469.0934\n",
      "Epoch 1501/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237339.7656 - root_mean_squared_error: 487.1753 - val_loss: 213038.1406 - val_root_mean_squared_error: 461.5605\n",
      "Epoch 1502/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 226434.9375 - root_mean_squared_error: 475.8518 - val_loss: 218448.8281 - val_root_mean_squared_error: 467.3851\n",
      "Epoch 1503/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 230096.7812 - root_mean_squared_error: 479.6841 - val_loss: 232759.3750 - val_root_mean_squared_error: 482.4514\n",
      "Epoch 1504/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 232162.7188 - root_mean_squared_error: 481.8327 - val_loss: 207720.0469 - val_root_mean_squared_error: 455.7632\n",
      "Epoch 1505/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 251954.2656 - root_mean_squared_error: 501.9505 - val_loss: 198228.3594 - val_root_mean_squared_error: 445.2284\n",
      "Epoch 1506/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 245237.2656 - root_mean_squared_error: 495.2144 - val_loss: 192246.0312 - val_root_mean_squared_error: 438.4587\n",
      "Epoch 1507/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 243971.9844 - root_mean_squared_error: 493.9352 - val_loss: 204219.1875 - val_root_mean_squared_error: 451.9062\n",
      "Epoch 1508/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 214178.4375 - root_mean_squared_error: 462.7942 - val_loss: 210564.7188 - val_root_mean_squared_error: 458.8733\n",
      "Epoch 1509/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 200252.7188 - root_mean_squared_error: 447.4961 - val_loss: 206553.1719 - val_root_mean_squared_error: 454.4812\n",
      "Epoch 1510/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 219308.0469 - root_mean_squared_error: 468.3034 - val_loss: 196279.1094 - val_root_mean_squared_error: 443.0340\n",
      "Epoch 1511/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 283905.3438 - root_mean_squared_error: 532.8277 - val_loss: 199487.8906 - val_root_mean_squared_error: 446.6407\n",
      "Epoch 1512/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 230644.0469 - root_mean_squared_error: 480.2542 - val_loss: 208357.2656 - val_root_mean_squared_error: 456.4617\n",
      "Epoch 1513/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 216772.1562 - root_mean_squared_error: 465.5880 - val_loss: 202045.2188 - val_root_mean_squared_error: 449.4944\n",
      "Epoch 1514/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 225522.8438 - root_mean_squared_error: 474.8925 - val_loss: 211243.9375 - val_root_mean_squared_error: 459.6128\n",
      "Epoch 1515/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 255861.9688 - root_mean_squared_error: 505.8280 - val_loss: 205753.7500 - val_root_mean_squared_error: 453.6009\n",
      "Epoch 1516/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220966.2500 - root_mean_squared_error: 470.0705 - val_loss: 203118.2656 - val_root_mean_squared_error: 450.6864\n",
      "Epoch 1517/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 219200.6250 - root_mean_squared_error: 468.1887 - val_loss: 214556.9844 - val_root_mean_squared_error: 463.2030\n",
      "Epoch 1518/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243941.3125 - root_mean_squared_error: 493.9041 - val_loss: 214596.6094 - val_root_mean_squared_error: 463.2457\n",
      "Epoch 1519/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 239386.0938 - root_mean_squared_error: 489.2710 - val_loss: 214144.9844 - val_root_mean_squared_error: 462.7580\n",
      "Epoch 1520/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220022.6562 - root_mean_squared_error: 469.0657 - val_loss: 211399.7969 - val_root_mean_squared_error: 459.7823\n",
      "Epoch 1521/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 248832.9219 - root_mean_squared_error: 498.8315 - val_loss: 209398.9688 - val_root_mean_squared_error: 457.6013\n",
      "Epoch 1522/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 257665.6250 - root_mean_squared_error: 507.6078 - val_loss: 198019.9844 - val_root_mean_squared_error: 444.9944\n",
      "Epoch 1523/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 246359.9219 - root_mean_squared_error: 496.3466 - val_loss: 196142.0625 - val_root_mean_squared_error: 442.8793\n",
      "Epoch 1524/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 259048.5625 - root_mean_squared_error: 508.9681 - val_loss: 196516.4062 - val_root_mean_squared_error: 443.3017\n",
      "Epoch 1525/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 238142.8281 - root_mean_squared_error: 487.9988 - val_loss: 207186.4531 - val_root_mean_squared_error: 455.1774\n",
      "Epoch 1526/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262390.6875 - root_mean_squared_error: 512.2408 - val_loss: 205972.0938 - val_root_mean_squared_error: 453.8415\n",
      "Epoch 1527/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 223454.0000 - root_mean_squared_error: 472.7092 - val_loss: 208519.1562 - val_root_mean_squared_error: 456.6390\n",
      "Epoch 1528/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 235439.6250 - root_mean_squared_error: 485.2212 - val_loss: 212335.8125 - val_root_mean_squared_error: 460.7991\n",
      "Epoch 1529/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 261303.0156 - root_mean_squared_error: 511.1781 - val_loss: 203281.5469 - val_root_mean_squared_error: 450.8676\n",
      "Epoch 1530/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246890.6875 - root_mean_squared_error: 496.8810 - val_loss: 198229.6406 - val_root_mean_squared_error: 445.2299\n",
      "Epoch 1531/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239241.5156 - root_mean_squared_error: 489.1232 - val_loss: 212118.6094 - val_root_mean_squared_error: 460.5634\n",
      "Epoch 1532/1600\n",
      "13/13 [==============================] - 0s 10ms/step - loss: 228175.8125 - root_mean_squared_error: 477.6775 - val_loss: 231359.9062 - val_root_mean_squared_error: 480.9989\n",
      "Epoch 1533/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 199104.7812 - root_mean_squared_error: 446.2116 - val_loss: 232475.7188 - val_root_mean_squared_error: 482.1573\n",
      "Epoch 1534/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 215187.5625 - root_mean_squared_error: 463.8831 - val_loss: 208461.7500 - val_root_mean_squared_error: 456.5761\n",
      "Epoch 1535/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 213020.5312 - root_mean_squared_error: 461.5415 - val_loss: 206819.4219 - val_root_mean_squared_error: 454.7740\n",
      "Epoch 1536/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244974.8438 - root_mean_squared_error: 494.9493 - val_loss: 215231.4219 - val_root_mean_squared_error: 463.9304\n",
      "Epoch 1537/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 237391.7188 - root_mean_squared_error: 487.2286 - val_loss: 215806.1562 - val_root_mean_squared_error: 464.5494\n",
      "Epoch 1538/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 246475.2969 - root_mean_squared_error: 496.4628 - val_loss: 219087.2344 - val_root_mean_squared_error: 468.0676\n",
      "Epoch 1539/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 243744.2969 - root_mean_squared_error: 493.7047 - val_loss: 214769.1406 - val_root_mean_squared_error: 463.4319\n",
      "Epoch 1540/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 242132.4531 - root_mean_squared_error: 492.0695 - val_loss: 217470.4531 - val_root_mean_squared_error: 466.3373\n",
      "Epoch 1541/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 225841.5938 - root_mean_squared_error: 475.2279 - val_loss: 211690.1562 - val_root_mean_squared_error: 460.0980\n",
      "Epoch 1542/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 201218.5938 - root_mean_squared_error: 448.5739 - val_loss: 213500.6875 - val_root_mean_squared_error: 462.0613\n",
      "Epoch 1543/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 269255.5938 - root_mean_squared_error: 518.8984 - val_loss: 223771.7656 - val_root_mean_squared_error: 473.0452\n",
      "Epoch 1544/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 212059.0781 - root_mean_squared_error: 460.4987 - val_loss: 222279.8281 - val_root_mean_squared_error: 471.4656\n",
      "Epoch 1545/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 229658.5312 - root_mean_squared_error: 479.2270 - val_loss: 218958.6250 - val_root_mean_squared_error: 467.9301\n",
      "Epoch 1546/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247320.0000 - root_mean_squared_error: 497.3128 - val_loss: 192417.0156 - val_root_mean_squared_error: 438.6536\n",
      "Epoch 1547/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 233977.6250 - root_mean_squared_error: 483.7123 - val_loss: 204467.2656 - val_root_mean_squared_error: 452.1806\n",
      "Epoch 1548/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 261018.0312 - root_mean_squared_error: 510.8992 - val_loss: 206919.8438 - val_root_mean_squared_error: 454.8844\n",
      "Epoch 1549/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 222615.7969 - root_mean_squared_error: 471.8218 - val_loss: 209641.6406 - val_root_mean_squared_error: 457.8664\n",
      "Epoch 1550/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 247826.3750 - root_mean_squared_error: 497.8216 - val_loss: 215524.0625 - val_root_mean_squared_error: 464.2457\n",
      "Epoch 1551/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 214526.2969 - root_mean_squared_error: 463.1698 - val_loss: 210225.4688 - val_root_mean_squared_error: 458.5035\n",
      "Epoch 1552/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 235160.9531 - root_mean_squared_error: 484.9340 - val_loss: 209814.2344 - val_root_mean_squared_error: 458.0548\n",
      "Epoch 1553/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 235082.8125 - root_mean_squared_error: 484.8534 - val_loss: 208971.7188 - val_root_mean_squared_error: 457.1342\n",
      "Epoch 1554/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 232987.0312 - root_mean_squared_error: 482.6873 - val_loss: 208288.5625 - val_root_mean_squared_error: 456.3864\n",
      "Epoch 1555/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 262188.1562 - root_mean_squared_error: 512.0431 - val_loss: 216905.0312 - val_root_mean_squared_error: 465.7307\n",
      "Epoch 1556/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 265604.0938 - root_mean_squared_error: 515.3679 - val_loss: 224029.4844 - val_root_mean_squared_error: 473.3175\n",
      "Epoch 1557/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249219.3281 - root_mean_squared_error: 499.2187 - val_loss: 227284.8281 - val_root_mean_squared_error: 476.7440\n",
      "Epoch 1558/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 264483.8125 - root_mean_squared_error: 514.2799 - val_loss: 233675.3906 - val_root_mean_squared_error: 483.3998\n",
      "Epoch 1559/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 236434.1875 - root_mean_squared_error: 486.2450 - val_loss: 218183.0781 - val_root_mean_squared_error: 467.1007\n",
      "Epoch 1560/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 289607.5000 - root_mean_squared_error: 538.1519 - val_loss: 218812.5938 - val_root_mean_squared_error: 467.7741\n",
      "Epoch 1561/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 257749.9062 - root_mean_squared_error: 507.6908 - val_loss: 208819.1250 - val_root_mean_squared_error: 456.9673\n",
      "Epoch 1562/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 277073.8125 - root_mean_squared_error: 526.3780 - val_loss: 206493.9531 - val_root_mean_squared_error: 454.4160\n",
      "Epoch 1563/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 231257.2656 - root_mean_squared_error: 480.8922 - val_loss: 193753.7656 - val_root_mean_squared_error: 440.1747\n",
      "Epoch 1564/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 254760.8750 - root_mean_squared_error: 504.7384 - val_loss: 216543.7344 - val_root_mean_squared_error: 465.3426\n",
      "Epoch 1565/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 228999.0469 - root_mean_squared_error: 478.5385 - val_loss: 209681.0312 - val_root_mean_squared_error: 457.9094\n",
      "Epoch 1566/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 224684.6094 - root_mean_squared_error: 474.0091 - val_loss: 215443.4688 - val_root_mean_squared_error: 464.1589\n",
      "Epoch 1567/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 218561.7812 - root_mean_squared_error: 467.5059 - val_loss: 234032.4688 - val_root_mean_squared_error: 483.7690\n",
      "Epoch 1568/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 234331.0469 - root_mean_squared_error: 484.0775 - val_loss: 209317.7812 - val_root_mean_squared_error: 457.5126\n",
      "Epoch 1569/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 228738.9844 - root_mean_squared_error: 478.2666 - val_loss: 200748.1719 - val_root_mean_squared_error: 448.0493\n",
      "Epoch 1570/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 258470.4844 - root_mean_squared_error: 508.3999 - val_loss: 192178.4531 - val_root_mean_squared_error: 438.3816\n",
      "Epoch 1571/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 249952.2656 - root_mean_squared_error: 499.9523 - val_loss: 183829.7812 - val_root_mean_squared_error: 428.7538\n",
      "Epoch 1572/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 232365.3125 - root_mean_squared_error: 482.0428 - val_loss: 193393.4688 - val_root_mean_squared_error: 439.7652\n",
      "Epoch 1573/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 256973.0469 - root_mean_squared_error: 506.9251 - val_loss: 196241.8438 - val_root_mean_squared_error: 442.9919\n",
      "Epoch 1574/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 255053.3906 - root_mean_squared_error: 505.0281 - val_loss: 188683.5469 - val_root_mean_squared_error: 434.3772\n",
      "Epoch 1575/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 216619.8906 - root_mean_squared_error: 465.4244 - val_loss: 187570.6250 - val_root_mean_squared_error: 433.0942\n",
      "Epoch 1576/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 203103.7188 - root_mean_squared_error: 450.6703 - val_loss: 198176.1094 - val_root_mean_squared_error: 445.1697\n",
      "Epoch 1577/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 239377.1250 - root_mean_squared_error: 489.2618 - val_loss: 190615.3438 - val_root_mean_squared_error: 436.5952\n",
      "Epoch 1578/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 241884.6250 - root_mean_squared_error: 491.8177 - val_loss: 197871.7969 - val_root_mean_squared_error: 444.8278\n",
      "Epoch 1579/1600\n",
      "13/13 [==============================] - 0s 9ms/step - loss: 217945.1250 - root_mean_squared_error: 466.8459 - val_loss: 192577.2812 - val_root_mean_squared_error: 438.8363\n",
      "Epoch 1580/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 233692.0781 - root_mean_squared_error: 483.4171 - val_loss: 202745.1094 - val_root_mean_squared_error: 450.2722\n",
      "Epoch 1581/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 242889.8125 - root_mean_squared_error: 492.8385 - val_loss: 195351.7500 - val_root_mean_squared_error: 441.9861\n",
      "Epoch 1582/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 215190.7812 - root_mean_squared_error: 463.8866 - val_loss: 210999.2031 - val_root_mean_squared_error: 459.3465\n",
      "Epoch 1583/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 209422.9844 - root_mean_squared_error: 457.6276 - val_loss: 208823.5781 - val_root_mean_squared_error: 456.9722\n",
      "Epoch 1584/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 220251.5469 - root_mean_squared_error: 469.3097 - val_loss: 224053.5938 - val_root_mean_squared_error: 473.3430\n",
      "Epoch 1585/1600\n",
      "13/13 [==============================] - 0s 7ms/step - loss: 283374.9062 - root_mean_squared_error: 532.3297 - val_loss: 229639.7500 - val_root_mean_squared_error: 479.2074\n",
      "Epoch 1586/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 227414.8281 - root_mean_squared_error: 476.8803 - val_loss: 235353.2500 - val_root_mean_squared_error: 485.1322\n",
      "Epoch 1587/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 231927.9531 - root_mean_squared_error: 481.5890 - val_loss: 206753.6094 - val_root_mean_squared_error: 454.7017\n",
      "Epoch 1588/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 212753.0000 - root_mean_squared_error: 461.2516 - val_loss: 210354.2812 - val_root_mean_squared_error: 458.6440\n",
      "Epoch 1589/1600\n",
      "13/13 [==============================] - 0s 8ms/step - loss: 232625.7969 - root_mean_squared_error: 482.3130 - val_loss: 195783.9531 - val_root_mean_squared_error: 442.4748\n",
      "Epoch 1590/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 242890.0000 - root_mean_squared_error: 492.8387 - val_loss: 208744.5156 - val_root_mean_squared_error: 456.8857\n",
      "Epoch 1591/1600\n",
      "13/13 [==============================] - 0s 6ms/step - loss: 218438.0312 - root_mean_squared_error: 467.3735 - val_loss: 212843.5781 - val_root_mean_squared_error: 461.3497\n",
      "Epoch 1592/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 244534.7812 - root_mean_squared_error: 494.5046 - val_loss: 224535.4375 - val_root_mean_squared_error: 473.8517\n",
      "Epoch 1593/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 229418.6406 - root_mean_squared_error: 478.9767 - val_loss: 221227.2812 - val_root_mean_squared_error: 470.3481\n",
      "Epoch 1594/1600\n",
      "13/13 [==============================] - 0s 10ms/step - loss: 213805.3750 - root_mean_squared_error: 462.3909 - val_loss: 219931.1719 - val_root_mean_squared_error: 468.9682\n",
      "Epoch 1595/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 267953.6875 - root_mean_squared_error: 517.6425 - val_loss: 222661.1094 - val_root_mean_squared_error: 471.8698\n",
      "Epoch 1596/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 250438.7656 - root_mean_squared_error: 500.4386 - val_loss: 213817.4219 - val_root_mean_squared_error: 462.4040\n",
      "Epoch 1597/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 270213.8750 - root_mean_squared_error: 519.8210 - val_loss: 217023.5312 - val_root_mean_squared_error: 465.8578\n",
      "Epoch 1598/1600\n",
      "13/13 [==============================] - 0s 5ms/step - loss: 216128.1250 - root_mean_squared_error: 464.8958 - val_loss: 234417.0781 - val_root_mean_squared_error: 484.1664\n",
      "Epoch 1599/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 225112.2969 - root_mean_squared_error: 474.4600 - val_loss: 231739.0625 - val_root_mean_squared_error: 481.3928\n",
      "Epoch 1600/1600\n",
      "13/13 [==============================] - 0s 4ms/step - loss: 221370.2188 - root_mean_squared_error: 470.5000 - val_loss: 234217.7344 - val_root_mean_squared_error: 483.9605\n",
      "INFO:tensorflow:Assets written to: model_3layers_with_dropout\\assets\n"
     ]
    }
   ],
   "source": [
    "\n",
    "if TRAIN:\n",
    "    history_nn = nn_model.fit(\n",
    "        x=train_data_sc,\n",
    "        y=train_labels,\n",
    "        batch_size=BATCH_SIZE,\n",
    "        epochs=EPOCHS*2,\n",
    "        verbose=\"auto\",\n",
    "        validation_split=VAL_SPLIT\n",
    "    )\n",
    "\n",
    "    nn_model.save('model_3layers_with_dropout')\n",
    "else:\n",
    "    nn_model = tf.keras.models.load_model(\"model_3layers_with_dropout\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "plt.plot(list(range(EPOCHS*2)), history_nn.history[\"loss\"], label=\"train\")\n",
    "plt.plot(list(range(EPOCHS*2)), history_nn.history[\"val_loss\"], label=\"val\")\n",
    "plt.title(\"Neural Network - Loss (MSE)\")\n",
    "plt.xlabel(\"epochs\")\n",
    "plt.ylabel(\"MSE\")\n",
    "plt.grid()\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "plt.plot(list(range(EPOCHS*2)), history_nn.history[\"root_mean_squared_error\"], label=\"train\")\n",
    "plt.plot(list(range(EPOCHS*2)), history_nn.history[\"val_root_mean_squared_error\"], label=\"val\")\n",
    "plt.title(\"Neural Network - RMSE\")\n",
    "plt.xlabel(\"epochs\")\n",
    "plt.ylabel(\"RMSE\")\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.grid()\n",
    "plt.savefig('NN_learning_curve.png')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "nn_pred = nn_model.predict(test_data_sc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NN model 2 (with dropout)RMSE:\t 325.04384320334185\n",
      "Black-Scholes RMSE:\t\t 771.9785736192837\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(\"NN model 2 (with dropout)RMSE:\\t\",mean_squared_error(nn_pred, test_labels)**0.5)\n",
    "print(\"Black-Scholes RMSE:\\t\\t\",mean_squared_error(test_df[\"BS\"], test_labels)**0.5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "params_dict = {\n",
    "    'kernel' : ['poly', 'rbf', 'sigmoid'],\n",
    "    'degree' : np.arange(3,6),\n",
    "    'gamma': [1, 0.1, 0.5, 0.01, 0.005, 0.001, 0.0001, 0.00001],\n",
    "    'C': list(range(5,500,20)),\n",
    "    'shrinking':[True, False]\n",
    "}\n",
    "svr_model = SVR()\n",
    "\n",
    "search = RandomizedSearchCV(\n",
    "    svr_model,\n",
    "    params_dict,\n",
    "    cv=3,\n",
    "    n_iter=15, scoring='neg_mean_squared_error',\n",
    "    verbose=3\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "if TRAIN:\n",
    "    # search.fit(\n",
    "    #     train_data_sc,\n",
    "    #     train_labels\n",
    "    # )\n",
    "    svr_model = SVR(\n",
    "        C= 305,\n",
    "        cache_size= 200,\n",
    "        coef0= 0.0,\n",
    "        degree= 3,\n",
    "        epsilon= 0.1,\n",
    "        gamma= 0.1,\n",
    "        kernel= 'rbf',\n",
    "        max_iter= -1,\n",
    "        shrinking= False,\n",
    "        tol= 0.001,\n",
    "        verbose= False\n",
    "    )\n",
    "    svr_model.fit(\n",
    "        train_data_sc,\n",
    "        train_labels\n",
    "    )\n",
    "    best_svr_model = svr_model\n",
    "    # print(search.best_params_)\n",
    "    \n",
    "    filename = 'best_SVR.sav'\n",
    "    pickle.dump(best_svr_model, open(filename, 'wb'))\n",
    "else:\n",
    "    # load the model from disk\n",
    "    filename = 'best_SVR.sav'\n",
    "    best_svr_model = pickle.load(open(filename, 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 305,\n",
       " 'cache_size': 200,\n",
       " 'coef0': 0.0,\n",
       " 'degree': 3,\n",
       " 'epsilon': 0.1,\n",
       " 'gamma': 0.1,\n",
       " 'kernel': 'rbf',\n",
       " 'max_iter': -1,\n",
       " 'shrinking': False,\n",
       " 'tol': 0.001,\n",
       " 'verbose': False}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_svr_model.get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "svr_pred = best_svr_model.predict(test_data_sc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVR RMSE:\t\t 600.5260059146941\n",
      "Black-Scholes RMSE:\t 771.9785736192837\n"
     ]
    }
   ],
   "source": [
    "print('SVR RMSE:\\t\\t', mean_squared_error(svr_pred, test_labels)**0.5)\n",
    "print('Black-Scholes RMSE:\\t', mean_squared_error(test_df[\"BS\"], test_labels)**0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# nn_model = tf.keras.models.load_model(\"model_3layers_with_dropout\")\n",
    "# nn_pred = nn_model.predict(test_data_sc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Final Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neural Network RMSE:\t 325.044\n",
      "LightGBM RMSE:\t\t 332.393\n",
      "SVR RMSE:\t\t 600.526\n",
      "Black Sholes RMSE:\t 771.979\n"
     ]
    }
   ],
   "source": [
    "print(\"Neural Network RMSE:\\t\",round(mean_squared_error(nn_pred, test_labels, squared=False),3))\n",
    "print(\"LightGBM RMSE:\\t\\t\",round(mean_squared_error(lgbm_pred, test_labels, squared=False),3))\n",
    "print(\"SVR RMSE:\\t\\t\",round(mean_squared_error(svr_pred, test_labels, squared=False),3))\n",
    "print(\"Black Sholes RMSE:\\t\",round(mean_squared_error(test_df[\"BS\"], test_labels, squared=False),3))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is shown that three ML models have outperformed the Black-sholes model. The Neural Net has had the best performance among the 3."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(data=pd.DataFrame(\n",
    "    {\n",
    "        \"NN\": [mean_squared_error(nn_pred, test_labels, squared=False)],\n",
    "        \"LGBM\": [mean_squared_error(lgbm_pred, test_labels, squared=False)],\n",
    "        \"SVR\": [mean_squared_error(svr_pred, test_labels, squared=False)],\n",
    "        \"BS\": [mean_squared_error(test_df[\"BS\"], test_labels, squared=False)]\n",
    "    }\n",
    "))\n",
    "plt.ylabel(\"error (RMSE)\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neural Network full RMSE:\t 306.011\n"
     ]
    }
   ],
   "source": [
    "full_nn_pred = nn_model.predict(scaler.transform(df[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values))\n",
    "print(\"Neural Network full RMSE:\\t\",round(mean_squared_error(full_nn_pred,\n",
    "                                                    df[[\"close_price\"]].values, squared=False),3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LightGBM full RMSE:\t 295.605\n"
     ]
    }
   ],
   "source": [
    "full_lgb_pred = best_lgb_model.predict(df[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values)\n",
    "print(\"LightGBM full RMSE:\\t\",round(mean_squared_error(full_lgb_pred,\n",
    "                                                    df[[\"close_price\"]].values, squared=False),3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVR full RMSE:\t 613.005\n"
     ]
    }
   ],
   "source": [
    "full_svr_pred = best_svr_model.predict(scaler.transform(df[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values))\n",
    "print(\"SVR full RMSE:\\t\",round(mean_squared_error(full_svr_pred,\n",
    "                                                    df[[\"close_price\"]].values, squared=False),3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Black Sholes full RMSE:\t 846.745\n"
     ]
    }
   ],
   "source": [
    "print(\"Black Sholes full RMSE:\\t\",round(mean_squared_error(df[\"BS\"].values, df[[\"close_price\"]].values, squared=False),3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rmse(x,y):\n",
    "    return(round(mean_squared_error(x, y, squared=False),2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = {}\n",
    "for term in [\"short_term\", \"long_term\", \"ALL\"]:\n",
    "    # print(\">>\",term+\":\",)    \n",
    "    result[term] = {}    \n",
    "    for moneyness in [\"ITM\", \"OTM\", \"ATM\", \"ALL\"]:\n",
    "        if (moneyness == \"ALL\") and (term ==\"ALL\"):\n",
    "            data = test_df.copy()\n",
    "        elif moneyness == \"ALL\":\n",
    "            data = test_df[test_df[\"time_group\"] == term].copy()\n",
    "        elif term ==\"ALL\":\n",
    "            data = test_df[test_df[\"moneyness\"] == moneyness].copy()\n",
    "        else:\n",
    "            data = test_df[(test_df[\"moneyness\"]==moneyness) & (test_df[\"time_group\"] == term)].copy()\n",
    "        x = data[[\"strike_price\", \"equity_price\", \"annual_volatility\", \"riskfree_rate\", \"time_until_exp\"]].values\n",
    "        y = data[[\"close_price\"]].values\n",
    "\n",
    "        # print(\"\\t\"+moneyness+\":\",end=\"\")\n",
    "        result[term][moneyness] = []\n",
    "        if len(data)==0:\n",
    "            result[term][moneyness].append(\"No Data\")\n",
    "            result[term][moneyness].append(\"No Data\")\n",
    "            result[term][moneyness].append(\"No Data\")\n",
    "            result[term][moneyness].append(\"No Data\")\n",
    "            # print(\"\\tNN: \", \"No Data\", end=\"\\n\\t\\t\")\n",
    "            # print(\"SVR: \", \"No Data\", end=\"\\n\\t\\t\")\n",
    "            # print(\"LGBM: \", \"No Data\")\n",
    "            continue\n",
    "        result[term][moneyness].append(rmse(nn_model.predict(scaler.transform(x)), y))\n",
    "        result[term][moneyness].append(rmse(best_svr_model.predict(scaler.transform(x)), y))\n",
    "        result[term][moneyness].append(rmse(best_lgb_model.predict(x), y))\n",
    "        result[term][moneyness].append(rmse(data[\"BS\"].values, y))\n",
    "        # print(\"\\tNN: \",rmse(model_2.predict(scaler.transform(x)), y) , end=\"\\n\\t\\t\")\n",
    "        # print(\"SVR: \",rmse(best_svr_model.predict(scaler.transform(x)), y), end=\"\\n\\t\\t\")\n",
    "        # print(\"LGBM: \",rmse(best_lgb_model.predict(x), y))\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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></th>\n",
       "      <th>NN</th>\n",
       "      <th>SVR</th>\n",
       "      <th>LGBM</th>\n",
       "      <th>BS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">short_term</th>\n",
       "      <th>ITM</th>\n",
       "      <td>225.17</td>\n",
       "      <td>342.1</td>\n",
       "      <td>219.09</td>\n",
       "      <td>443.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OTM</th>\n",
       "      <td>340.56</td>\n",
       "      <td>732.9</td>\n",
       "      <td>396.41</td>\n",
       "      <td>914.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ATM</th>\n",
       "      <td>165.15</td>\n",
       "      <td>300.01</td>\n",
       "      <td>194.95</td>\n",
       "      <td>290.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ALL</th>\n",
       "      <td>298.53</td>\n",
       "      <td>603.77</td>\n",
       "      <td>335.1</td>\n",
       "      <td>757.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">long_term</th>\n",
       "      <th>ITM</th>\n",
       "      <td>808.96</td>\n",
       "      <td>386.37</td>\n",
       "      <td>254.59</td>\n",
       "      <td>614.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OTM</th>\n",
       "      <td>526.61</td>\n",
       "      <td>631.14</td>\n",
       "      <td>279.24</td>\n",
       "      <td>1319.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ATM</th>\n",
       "      <td>No Data</td>\n",
       "      <td>No Data</td>\n",
       "      <td>No Data</td>\n",
       "      <td>No Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ALL</th>\n",
       "      <td>679.46</td>\n",
       "      <td>525.9</td>\n",
       "      <td>267.47</td>\n",
       "      <td>1036.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">ALL</th>\n",
       "      <th>ITM</th>\n",
       "      <td>287.18</td>\n",
       "      <td>344.57</td>\n",
       "      <td>221.1</td>\n",
       "      <td>453.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OTM</th>\n",
       "      <td>349.61</td>\n",
       "      <td>729.23</td>\n",
       "      <td>392.53</td>\n",
       "      <td>933.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ATM</th>\n",
       "      <td>165.15</td>\n",
       "      <td>300.01</td>\n",
       "      <td>194.95</td>\n",
       "      <td>290.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ALL</th>\n",
       "      <td>325.04</td>\n",
       "      <td>600.53</td>\n",
       "      <td>332.39</td>\n",
       "      <td>771.98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     NN      SVR     LGBM       BS\n",
       "short_term ITM   225.17    342.1   219.09   443.07\n",
       "           OTM   340.56    732.9   396.41   914.97\n",
       "           ATM   165.15   300.01   194.95   290.11\n",
       "           ALL   298.53   603.77    335.1   757.44\n",
       "long_term  ITM   808.96   386.37   254.59   614.32\n",
       "           OTM   526.61   631.14   279.24  1319.76\n",
       "           ATM  No Data  No Data  No Data  No Data\n",
       "           ALL   679.46    525.9   267.47  1036.69\n",
       "ALL        ITM   287.18   344.57    221.1   453.69\n",
       "           OTM   349.61   729.23   392.53   933.92\n",
       "           ATM   165.15   300.01   194.95   290.11\n",
       "           ALL   325.04   600.53   332.39   771.98"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reform = {(outerKey, innerKey): values for outerKey, innerDict in result.items() for innerKey, values in innerDict.items()}\n",
    "reform_df = pd.DataFrame(reform)\n",
    "reform_df = reform_df.T\n",
    "reform_df.columns = [\"NN\", \"SVR\", \"LGBM\",\"BS\"]\n",
    "reform_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "''"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "errors = pd.DataFrame()\n",
    "\n",
    "errors ['NN'] = nn_pred.reshape(-1) - test_labels.reshape(-1) \n",
    "errors ['LGBM'] = lgbm_pred.reshape(-1) - test_labels.reshape(-1)\n",
    "errors ['SVR'] = svr_pred.reshape(-1)  - test_labels.reshape(-1)\n",
    "errors ['BS'] = test_df [\"BS\"].values.reshape(-1) - test_labels.reshape(-1)\n",
    "\n",
    "plt.figure(figsize=(12,8))\n",
    "sns.boxplot(data=errors, whis=5)\n",
    "plt.grid()\n",
    "plt.ylim((-3500,3500))\n",
    "plt.ylabel('Errors')\n",
    "plt.savefig(\"errors_boxplot.png\")\n",
    ";"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "''"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,6))\n",
    "sns.kdeplot(data=errors,fill=0.3)\n",
    "plt.grid()\n",
    "# plt.ylim((-3500,3500))\n",
    "plt.title(\"Errors Distribution\")\n",
    "plt.ylabel('Prob')\n",
    "plt.savefig(\"errors_dist.png\")\n",
    ";"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "err_desc = errors.describe()\n",
    "err_desc.columns = [col + ' errors' for col in err_desc.columns]\n",
    "err_desc.T.to_csv(\"errors_tosifi.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>NN errors</th>\n",
       "      <th>LGBM errors</th>\n",
       "      <th>SVR errors</th>\n",
       "      <th>BS errors</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1014.000000</td>\n",
       "      <td>1014.000000</td>\n",
       "      <td>1014.000000</td>\n",
       "      <td>1014.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-14.412748</td>\n",
       "      <td>-2.561110</td>\n",
       "      <td>-109.271517</td>\n",
       "      <td>111.210398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>324.884388</td>\n",
       "      <td>332.547545</td>\n",
       "      <td>590.792210</td>\n",
       "      <td>764.303120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-3512.150391</td>\n",
       "      <td>-3627.553093</td>\n",
       "      <td>-6617.949359</td>\n",
       "      <td>-4499.080793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-61.790066</td>\n",
       "      <td>-53.387706</td>\n",
       "      <td>-197.123307</td>\n",
       "      <td>-49.185938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>17.955208</td>\n",
       "      <td>2.932040</td>\n",
       "      <td>-4.642849</td>\n",
       "      <td>-2.457327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>70.064152</td>\n",
       "      <td>54.802499</td>\n",
       "      <td>88.874895</td>\n",
       "      <td>74.195112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2190.068115</td>\n",
       "      <td>2548.762534</td>\n",
       "      <td>2685.905120</td>\n",
       "      <td>9406.878405</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         NN errors  LGBM errors   SVR errors    BS errors\n",
       "count  1014.000000  1014.000000  1014.000000  1014.000000\n",
       "mean    -14.412748    -2.561110  -109.271517   111.210398\n",
       "std     324.884388   332.547545   590.792210   764.303120\n",
       "min   -3512.150391 -3627.553093 -6617.949359 -4499.080793\n",
       "25%     -61.790066   -53.387706  -197.123307   -49.185938\n",
       "50%      17.955208     2.932040    -4.642849    -2.457327\n",
       "75%      70.064152    54.802499    88.874895    74.195112\n",
       "max    2190.068115  2548.762534  2685.905120  9406.878405"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "err_desc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "vscode": {
   "interpreter": {
    "hash": "25034407fed5d681614dac11a1c0537e8cb49e3a8883c071303eea01322943d9"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}