train.ipynb

{
 "nbformat": 4,
 "nbformat_minor": 0,
 "metadata": {
  "colab": {
   "provenance": [],
   "collapsed_sections": []
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3"
  },
  "language_info": {
   "name": "python"
  },
  "accelerator": "GPU",
  "gpuClass": "standard"
 },
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "XUGsAbSgGa0y",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import pickle\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from keras.callbacks import EarlyStopping\n",
    "from tensorflow import keras\n",
    "\n",
    "\n",
    "year='2019'\n",
    "total_df = pd.read_csv('./src/data/dust_'+year+'.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "## EDA"
   ],
   "metadata": {
    "id": "GZEtgQD-TQut",
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "total_df.head(3)"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 143
    },
    "id": "BnpVid_9UDJR",
    "outputId": "d29f98f0-caab-4d4c-f21f-01577ad38a23",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 2,
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "   Unnamed: 0  MSRDT_DE MSRRGN_NM MSRSTE_NM  PM10  PM25     O3    NO2   CO  \\\n",
       "0           0  20190101       도심권        중구  38.0  26.0  0.019  0.032  0.6   \n",
       "1           1  20190101       도심권       종로구  35.0  22.0  0.016  0.033  0.7   \n",
       "2           2  20190101       도심권       용산구  31.0  25.0  0.013  0.033  0.6   \n",
       "\n",
       "     SO2  \n",
       "0  0.003  \n",
       "1  0.004  \n",
       "2  0.003  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-9bfbd0c1-1023-40f4-8b6f-97bae79ac1cb\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Unnamed: 0</th>\n",
       "      <th>MSRDT_DE</th>\n",
       "      <th>MSRRGN_NM</th>\n",
       "      <th>MSRSTE_NM</th>\n",
       "      <th>PM10</th>\n",
       "      <th>PM25</th>\n",
       "      <th>O3</th>\n",
       "      <th>NO2</th>\n",
       "      <th>CO</th>\n",
       "      <th>SO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>20190101</td>\n",
       "      <td>도심권</td>\n",
       "      <td>중구</td>\n",
       "      <td>38.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.019</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20190101</td>\n",
       "      <td>도심권</td>\n",
       "      <td>종로구</td>\n",
       "      <td>35.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0.016</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>20190101</td>\n",
       "      <td>도심권</td>\n",
       "      <td>용산구</td>\n",
       "      <td>31.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0.013</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.003</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9bfbd0c1-1023-40f4-8b6f-97bae79ac1cb')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-9bfbd0c1-1023-40f4-8b6f-97bae79ac1cb button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-9bfbd0c1-1023-40f4-8b6f-97bae79ac1cb');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ]
     },
     "metadata": {},
     "execution_count": 2
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "total_df.describe()"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "id": "ktqDl0ClO6yD",
    "outputId": "9f2449ba-66bd-4202-af48-4aadb86cdfd4",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 3,
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "        Unnamed: 0      MSRDT_DE         PM10         PM25           O3  \\\n",
       "count  8946.000000  8.946000e+03  8946.000000  8946.000000  8946.000000   \n",
       "mean     11.994634  2.019068e+07    40.861055    24.335345     0.025025   \n",
       "std       7.208651  3.446862e+02    27.457152    18.836681     0.013961   \n",
       "min       0.000000  2.019010e+07     0.000000     0.000000     0.000000   \n",
       "25%       6.000000  2.019041e+07    23.000000    13.000000     0.014000   \n",
       "50%      12.000000  2.019070e+07    35.000000    20.000000     0.024000   \n",
       "75%      18.000000  2.019100e+07    51.000000    30.000000     0.035000   \n",
       "max      24.000000  2.019123e+07   240.000000   155.000000     0.088000   \n",
       "\n",
       "               NO2           CO          SO2  \n",
       "count  8946.000000  8946.000000  8946.000000  \n",
       "mean      0.026945     0.512464     0.003857  \n",
       "std       0.013103     0.231364     0.001419  \n",
       "min       0.000000     0.000000     0.000000  \n",
       "25%       0.017000     0.400000     0.003000  \n",
       "50%       0.025000     0.500000     0.004000  \n",
       "75%       0.036000     0.600000     0.005000  \n",
       "max       0.080000     1.900000     0.012000  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-4269adb4-446c-4603-b6e7-480326e853c1\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Unnamed: 0</th>\n",
       "      <th>MSRDT_DE</th>\n",
       "      <th>PM10</th>\n",
       "      <th>PM25</th>\n",
       "      <th>O3</th>\n",
       "      <th>NO2</th>\n",
       "      <th>CO</th>\n",
       "      <th>SO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8.946000e+03</td>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8946.000000</td>\n",
       "      <td>8946.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.994634</td>\n",
       "      <td>2.019068e+07</td>\n",
       "      <td>40.861055</td>\n",
       "      <td>24.335345</td>\n",
       "      <td>0.025025</td>\n",
       "      <td>0.026945</td>\n",
       "      <td>0.512464</td>\n",
       "      <td>0.003857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.208651</td>\n",
       "      <td>3.446862e+02</td>\n",
       "      <td>27.457152</td>\n",
       "      <td>18.836681</td>\n",
       "      <td>0.013961</td>\n",
       "      <td>0.013103</td>\n",
       "      <td>0.231364</td>\n",
       "      <td>0.001419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.019010e+07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>2.019041e+07</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>0.014000</td>\n",
       "      <td>0.017000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.003000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>2.019070e+07</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>0.024000</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.004000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>2.019100e+07</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>0.035000</td>\n",
       "      <td>0.036000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.005000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>24.000000</td>\n",
       "      <td>2.019123e+07</td>\n",
       "      <td>240.000000</td>\n",
       "      <td>155.000000</td>\n",
       "      <td>0.088000</td>\n",
       "      <td>0.080000</td>\n",
       "      <td>1.900000</td>\n",
       "      <td>0.012000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4269adb4-446c-4603-b6e7-480326e853c1')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-4269adb4-446c-4603-b6e7-480326e853c1 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-4269adb4-446c-4603-b6e7-480326e853c1');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ]
     },
     "metadata": {},
     "execution_count": 3
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "total_df.corr()"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "id": "h_BXOe9yPB4p",
    "outputId": "7d1fb63c-17f1-4b4e-e20a-87464df27654",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 4,
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "            Unnamed: 0  MSRDT_DE      PM10      PM25        O3       NO2  \\\n",
       "Unnamed: 0    1.000000 -0.000936  0.046662  0.035748 -0.047753  0.065022   \n",
       "MSRDT_DE     -0.000936  1.000000 -0.397410 -0.357238 -0.183525 -0.204867   \n",
       "PM10          0.046662 -0.397410  1.000000  0.902611  0.016442  0.576541   \n",
       "PM25          0.035748 -0.357238  0.902611  1.000000  0.024517  0.566740   \n",
       "O3           -0.047753 -0.183525  0.016442  0.024517  1.000000 -0.343200   \n",
       "NO2           0.065022 -0.204867  0.576541  0.566740 -0.343200  1.000000   \n",
       "CO           -0.178870 -0.218219  0.612297  0.625629 -0.252219  0.687978   \n",
       "SO2           0.035999 -0.422283  0.444610  0.440845  0.092254  0.428213   \n",
       "\n",
       "                  CO       SO2  \n",
       "Unnamed: 0 -0.178870  0.035999  \n",
       "MSRDT_DE   -0.218219 -0.422283  \n",
       "PM10        0.612297  0.444610  \n",
       "PM25        0.625629  0.440845  \n",
       "O3         -0.252219  0.092254  \n",
       "NO2         0.687978  0.428213  \n",
       "CO          1.000000  0.427551  \n",
       "SO2         0.427551  1.000000  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-9e0e05c5-9edf-41c2-86b0-cac3aa2b9dce\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Unnamed: 0</th>\n",
       "      <th>MSRDT_DE</th>\n",
       "      <th>PM10</th>\n",
       "      <th>PM25</th>\n",
       "      <th>O3</th>\n",
       "      <th>NO2</th>\n",
       "      <th>CO</th>\n",
       "      <th>SO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.000936</td>\n",
       "      <td>0.046662</td>\n",
       "      <td>0.035748</td>\n",
       "      <td>-0.047753</td>\n",
       "      <td>0.065022</td>\n",
       "      <td>-0.178870</td>\n",
       "      <td>0.035999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSRDT_DE</th>\n",
       "      <td>-0.000936</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.397410</td>\n",
       "      <td>-0.357238</td>\n",
       "      <td>-0.183525</td>\n",
       "      <td>-0.204867</td>\n",
       "      <td>-0.218219</td>\n",
       "      <td>-0.422283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PM10</th>\n",
       "      <td>0.046662</td>\n",
       "      <td>-0.397410</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.902611</td>\n",
       "      <td>0.016442</td>\n",
       "      <td>0.576541</td>\n",
       "      <td>0.612297</td>\n",
       "      <td>0.444610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PM25</th>\n",
       "      <td>0.035748</td>\n",
       "      <td>-0.357238</td>\n",
       "      <td>0.902611</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.024517</td>\n",
       "      <td>0.566740</td>\n",
       "      <td>0.625629</td>\n",
       "      <td>0.440845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>O3</th>\n",
       "      <td>-0.047753</td>\n",
       "      <td>-0.183525</td>\n",
       "      <td>0.016442</td>\n",
       "      <td>0.024517</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.343200</td>\n",
       "      <td>-0.252219</td>\n",
       "      <td>0.092254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NO2</th>\n",
       "      <td>0.065022</td>\n",
       "      <td>-0.204867</td>\n",
       "      <td>0.576541</td>\n",
       "      <td>0.566740</td>\n",
       "      <td>-0.343200</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.687978</td>\n",
       "      <td>0.428213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CO</th>\n",
       "      <td>-0.178870</td>\n",
       "      <td>-0.218219</td>\n",
       "      <td>0.612297</td>\n",
       "      <td>0.625629</td>\n",
       "      <td>-0.252219</td>\n",
       "      <td>0.687978</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.427551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SO2</th>\n",
       "      <td>0.035999</td>\n",
       "      <td>-0.422283</td>\n",
       "      <td>0.444610</td>\n",
       "      <td>0.440845</td>\n",
       "      <td>0.092254</td>\n",
       "      <td>0.428213</td>\n",
       "      <td>0.427551</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9e0e05c5-9edf-41c2-86b0-cac3aa2b9dce')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-9e0e05c5-9edf-41c2-86b0-cac3aa2b9dce button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-9e0e05c5-9edf-41c2-86b0-cac3aa2b9dce');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ]
     },
     "metadata": {},
     "execution_count": 4
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 서로 상관계수가 높은 변수들이고, 해당 과제에서 input 값이 명백히 정해져 있는 것이 아니다.\n",
    "\n",
    "### 그렇다면 간단하게 각 자치구 별로 PM10을 시계열 모델로 예측하여, 그 이후 날짜를 입력받으면 해당 날짜의 미세먼지 농도를 예측하는 방향으로 접근하자"
   ],
   "metadata": {
    "id": "JOEDxP_ATZGA",
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "total_df.drop(['PM25','O3','NO2','CO','SO2'], axis=1, inplace=True)"
   ],
   "metadata": {
    "id": "nfPFJpU2UGlg",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 5,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "total_df.sort_values(['MSRSTE_NM','MSRDT_DE'], inplace=True)"
   ],
   "metadata": {
    "id": "H6IybVD6Rmqk",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 6,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "region_list =list(total_df['MSRSTE_NM'].unique())"
   ],
   "metadata": {
    "id": "uhXnWj2oEMJu",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 7,
   "outputs": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "region_dict = {\n",
    "\t'종로구' : 'Jongno',\n",
    "\t'중구' : 'Jung',\n",
    "\t'용산구' : 'Yongsan',\n",
    "\t'성동구' : 'Seongdong',\n",
    "\t'광진구' : 'Gwangjin',\n",
    "\t'동대문구' : 'Dongdaemun',\n",
    "\t'중랑구' : 'Jungnang',\n",
    "\t'성북구' : 'Seongbuk',\n",
    "\t'강북구' : 'Gangbuk',\n",
    "\t'도봉구' : 'Dobong',\n",
    "\t'노원구' : 'Nowon',\n",
    "\t'은평구' : 'Eunpyeong',\n",
    "\t'서대문구' : 'Seodaemun',\n",
    "\t'마포구' : 'Mapo',\n",
    "\t'양천구' : 'Yangcheon',\n",
    "\t'강서구' : 'Gangseo',\n",
    "\t'구로구' : 'Guro',\n",
    "\t'금천구' : 'Geumcheon',\n",
    "\t'영등포구' : 'Yeongdeungpo',\n",
    "\t'동작구' : 'Dongjak',\n",
    "\t'관악구' : 'Gwanak',\n",
    "\t'서초구' : 'Seocho',\n",
    "\t'강남구' : 'Gangnam',\n",
    "\t'송파구' : 'Songpa',\n",
    "\t'강동구' : 'Gangdong'\n",
    "}"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "def train_val_split(time, series, time_step):\n",
    "\n",
    "    time_train = time[:time_step]\n",
    "    series_train = series[:time_step]\n",
    "    time_valid = time[time_step:]\n",
    "    series_valid = series[time_step:]\n",
    "\n",
    "    return time_train, series_train, time_valid, series_valid\n",
    "\n",
    "def windowed_dataset(series, window_size, batch_size, shuffle_buffer):\n",
    "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
    "    ds = ds.window(window_size + 1, shift=1, drop_remainder=True)\n",
    "    ds = ds.flat_map(lambda w: w.batch(window_size + 1))\n",
    "    ds = ds.shuffle(shuffle_buffer)\n",
    "    ds = ds.map(lambda w: (w[:-1], w[-1]))\n",
    "    ds = ds.batch(batch_size).prefetch(1)\n",
    "    return ds"
   ],
   "metadata": {
    "id": "q4GDZew7Chy_",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 8,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "window_size = 15\n",
    "batch_size = 100\n",
    "shuffle_buffer_size = 30\n",
    "\n",
    "earlystopping = EarlyStopping(\n",
    "    monitor='val_loss',\n",
    "    patience=20\n",
    "    )\n",
    "\n",
    "for i in region_list:\n",
    "  temp_df = total_df.loc[total_df['MSRSTE_NM']==i]\n",
    "  temp_time = np.array(temp_df.reset_index().index)\n",
    "  temp_series = np.array(temp_df['PM10'])\n",
    "  split_time = int(len(temp_df)*0.8)\n",
    "\n",
    "  time_train, series_train, time_valid, series_valid = train_val_split(temp_time, temp_series, split_time)\n",
    "\n",
    "  train_set = windowed_dataset(series_train, window_size=window_size, batch_size=batch_size, shuffle_buffer=shuffle_buffer_size)\n",
    "  valid_set = windowed_dataset(series_valid, window_size=window_size, batch_size=batch_size, shuffle_buffer=shuffle_buffer_size)  \n",
    "  tf.keras.backend.clear_session()\n",
    "  tf.random.set_seed(51)\n",
    "  np.random.seed(51)\n",
    "\n",
    "  model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Conv1D(filters=60, kernel_size=5,\n",
    "                        strides=1, padding=\"causal\",\n",
    "                        activation=\"relu\",\n",
    "                        input_shape=[None, 1]),\n",
    "    tf.keras.layers.LSTM(90, return_sequences=True),\n",
    "    tf.keras.layers.LSTM(90),\n",
    "    tf.keras.layers.Dense(30, activation=\"relu\"),\n",
    "    tf.keras.layers.Dense(10, activation=\"relu\"),\n",
    "    tf.keras.layers.Dense(1),\n",
    "    tf.keras.layers.Lambda(lambda x: x * 400)\n",
    "  ])\n",
    "\n",
    "\n",
    "  optimizer = tf.keras.optimizers.SGD(learning_rate=1e-5, momentum=0.9)\n",
    "  model.compile(loss=tf.keras.losses.Huber(),\n",
    "                optimizer=optimizer,\n",
    "                metrics=[\"mae\"])\n",
    "  history = model.fit(train_set,\n",
    "                      validation_data = valid_set,\n",
    "                      epochs=150,\n",
    "                      callbacks=[earlystopping])\n",
    "  \n",
    "  model.save(year+'_'+region_dict[i]+'.h5')"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_7zgy2LwE17n",
    "outputId": "62066e8b-0d26-44e9-f793-176f2641159b",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 9,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Epoch 1/150\n",
      "3/3 [==============================] - 11s 1s/step - loss: 79.8716 - mae: 80.3715 - val_loss: 19.5055 - val_mae: 19.9908\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 57ms/step - loss: 24.1303 - mae: 24.6262 - val_loss: 11.6086 - val_mae: 12.1085\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 69ms/step - loss: 16.0192 - mae: 16.5105 - val_loss: 22.4660 - val_mae: 22.9660\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 43ms/step - loss: 18.5743 - mae: 19.0726 - val_loss: 12.5493 - val_mae: 13.0479\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 64ms/step - loss: 15.8888 - mae: 16.3816 - val_loss: 12.1902 - val_mae: 12.6847\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 62ms/step - loss: 13.8033 - mae: 14.2951 - val_loss: 12.2554 - val_mae: 12.7388\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 48ms/step - loss: 14.2117 - mae: 14.7025 - val_loss: 11.8166 - val_mae: 12.3025\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 13.1814 - mae: 13.6762 - val_loss: 11.9903 - val_mae: 12.4733\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 48ms/step - loss: 13.6440 - mae: 14.1332 - val_loss: 11.4731 - val_mae: 11.9725\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 62ms/step - loss: 13.4084 - mae: 13.9021 - val_loss: 11.4722 - val_mae: 11.9713\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 49ms/step - loss: 13.0583 - mae: 13.5490 - val_loss: 11.5083 - val_mae: 12.0067\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 64ms/step - loss: 13.1885 - mae: 13.6797 - val_loss: 11.6211 - val_mae: 12.1202\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 12.8485 - mae: 13.3437 - val_loss: 11.5108 - val_mae: 12.0058\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 56ms/step - loss: 12.6855 - mae: 13.1755 - val_loss: 11.4700 - val_mae: 11.9642\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 67ms/step - loss: 12.4757 - mae: 12.9649 - val_loss: 11.2454 - val_mae: 11.7299\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 12.5081 - mae: 12.9980 - val_loss: 11.1820 - val_mae: 11.6716\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.3367 - mae: 12.8266 - val_loss: 11.1281 - val_mae: 11.6164\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.2688 - mae: 12.7589 - val_loss: 11.1347 - val_mae: 11.6334\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.2526 - mae: 12.7367 - val_loss: 11.0386 - val_mae: 11.5148\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.2553 - mae: 12.7447 - val_loss: 11.2074 - val_mae: 11.6946\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.2030 - mae: 12.6919 - val_loss: 11.2219 - val_mae: 11.7168\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.0497 - mae: 12.5336 - val_loss: 11.3510 - val_mae: 11.8386\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.0656 - mae: 12.5462 - val_loss: 11.0805 - val_mae: 11.5794\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 11.9048 - mae: 12.3908 - val_loss: 11.1422 - val_mae: 11.6367\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.8249 - mae: 12.3124 - val_loss: 11.0483 - val_mae: 11.5347\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.9056 - mae: 12.3934 - val_loss: 11.2540 - val_mae: 11.7540\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.7213 - mae: 12.2076 - val_loss: 11.1783 - val_mae: 11.6599\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.8597 - mae: 12.3502 - val_loss: 11.3238 - val_mae: 11.8182\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.7650 - mae: 12.2469 - val_loss: 11.1842 - val_mae: 11.6741\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.8279 - mae: 12.3149 - val_loss: 11.3743 - val_mae: 11.8733\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 11.7995 - mae: 12.2866 - val_loss: 11.4188 - val_mae: 11.9068\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.1612 - mae: 12.6516 - val_loss: 11.4929 - val_mae: 11.9828\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.2934 - mae: 12.7805 - val_loss: 11.4633 - val_mae: 11.9590\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.6940 - mae: 12.1778 - val_loss: 11.2770 - val_mae: 11.7574\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7159 - mae: 12.2018 - val_loss: 10.7461 - val_mae: 11.2361\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.5074 - mae: 11.9968 - val_loss: 10.7846 - val_mae: 11.2811\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.4044 - mae: 11.8867 - val_loss: 11.1675 - val_mae: 11.6440\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.5259 - mae: 12.0134 - val_loss: 11.3464 - val_mae: 11.8403\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.4830 - mae: 11.9689 - val_loss: 11.1496 - val_mae: 11.6247\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3363 - mae: 11.8244 - val_loss: 11.5842 - val_mae: 12.0812\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.5771 - mae: 12.0679 - val_loss: 11.6449 - val_mae: 12.1301\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.4778 - mae: 11.9626 - val_loss: 11.6433 - val_mae: 12.1333\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.4792 - mae: 11.9632 - val_loss: 11.3346 - val_mae: 11.8048\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.7177 - mae: 12.2016 - val_loss: 11.5998 - val_mae: 12.0897\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5385 - mae: 12.0274 - val_loss: 11.4119 - val_mae: 11.8922\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.2191 - mae: 12.7060 - val_loss: 11.8218 - val_mae: 12.3218\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.0523 - mae: 13.5470 - val_loss: 11.5000 - val_mae: 11.9838\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.3051 - mae: 12.7956 - val_loss: 11.5038 - val_mae: 11.9962\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.9733 - mae: 12.4635 - val_loss: 10.9105 - val_mae: 11.4009\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.1318 - mae: 11.6227 - val_loss: 10.8009 - val_mae: 11.2829\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.4662 - mae: 11.9560 - val_loss: 10.7005 - val_mae: 11.1803\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.0452 - mae: 11.5349 - val_loss: 10.8031 - val_mae: 11.3002\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.1841 - mae: 11.6711 - val_loss: 10.8319 - val_mae: 11.3257\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.0368 - mae: 11.5260 - val_loss: 10.7240 - val_mae: 11.1984\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.0649 - mae: 11.5536 - val_loss: 11.1079 - val_mae: 11.5986\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.0483 - mae: 11.5300 - val_loss: 10.9115 - val_mae: 11.4031\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.9385 - mae: 11.4298 - val_loss: 11.4606 - val_mae: 11.9582\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.2961 - mae: 11.7855 - val_loss: 10.8716 - val_mae: 11.3637\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.0082 - mae: 11.4956 - val_loss: 10.8140 - val_mae: 11.3066\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.9768 - mae: 11.4610 - val_loss: 11.1154 - val_mae: 11.6147\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 11.0051 - mae: 11.4911 - val_loss: 11.3443 - val_mae: 11.8377\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1641 - mae: 11.6515 - val_loss: 10.8181 - val_mae: 11.2773\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.9530 - mae: 11.4303 - val_loss: 11.0323 - val_mae: 11.5233\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.2500 - mae: 11.7380 - val_loss: 11.7588 - val_mae: 12.2487\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.3438 - mae: 11.8353 - val_loss: 10.7548 - val_mae: 11.2459\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.6601 - mae: 11.1518 - val_loss: 10.9219 - val_mae: 11.4182\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.8898 - mae: 11.3806 - val_loss: 10.7885 - val_mae: 11.2812\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.5843 - mae: 11.0691 - val_loss: 10.9924 - val_mae: 11.4906\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.8814 - mae: 11.3703 - val_loss: 10.6306 - val_mae: 11.1208\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.6392 - mae: 11.1119 - val_loss: 10.6694 - val_mae: 11.1516\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 10.9134 - mae: 11.3977 - val_loss: 10.8568 - val_mae: 11.3405\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.7588 - mae: 11.2410 - val_loss: 11.2821 - val_mae: 11.7821\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.1281 - mae: 11.6149 - val_loss: 11.0240 - val_mae: 11.5072\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.6274 - mae: 11.1061 - val_loss: 11.4280 - val_mae: 11.9276\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5630 - mae: 12.0549 - val_loss: 11.2188 - val_mae: 11.7061\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.8251 - mae: 11.3077 - val_loss: 11.1586 - val_mae: 11.6586\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.2794 - mae: 11.7663 - val_loss: 10.9029 - val_mae: 11.4011\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.5482 - mae: 11.0287 - val_loss: 11.0795 - val_mae: 11.5795\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 10.6121 - mae: 11.0951 - val_loss: 10.7528 - val_mae: 11.2308\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.4563 - mae: 10.9450 - val_loss: 10.4962 - val_mae: 10.9802\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.4332 - mae: 10.9189 - val_loss: 10.8662 - val_mae: 11.3644\n",
      "Epoch 82/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.3775 - mae: 10.8615 - val_loss: 10.8720 - val_mae: 11.3560\n",
      "Epoch 83/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.1802 - mae: 10.6624 - val_loss: 10.8578 - val_mae: 11.3535\n",
      "Epoch 84/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.2522 - mae: 10.7360 - val_loss: 10.9978 - val_mae: 11.4833\n",
      "Epoch 85/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.2334 - mae: 10.7190 - val_loss: 10.8801 - val_mae: 11.3579\n",
      "Epoch 86/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.0359 - mae: 10.5204 - val_loss: 10.7411 - val_mae: 11.2213\n",
      "Epoch 87/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.0562 - mae: 10.5356 - val_loss: 10.6282 - val_mae: 11.1174\n",
      "Epoch 88/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 9.9520 - mae: 10.4263 - val_loss: 10.8080 - val_mae: 11.2946\n",
      "Epoch 89/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 10.0140 - mae: 10.4925 - val_loss: 10.7338 - val_mae: 11.2273\n",
      "Epoch 90/150\n",
      "3/3 [==============================] - 0s 41ms/step - loss: 10.3714 - mae: 10.8572 - val_loss: 11.5056 - val_mae: 12.0054\n",
      "Epoch 91/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 10.5212 - mae: 11.0053 - val_loss: 11.6680 - val_mae: 12.1656\n",
      "Epoch 92/150\n",
      "3/3 [==============================] - 0s 41ms/step - loss: 11.1378 - mae: 11.6265 - val_loss: 10.5971 - val_mae: 11.0816\n",
      "Epoch 93/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 10.9819 - mae: 11.4677 - val_loss: 11.5445 - val_mae: 12.0415\n",
      "Epoch 94/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.4321 - mae: 11.9241 - val_loss: 11.2523 - val_mae: 11.7478\n",
      "Epoch 95/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.8181 - mae: 11.3070 - val_loss: 10.9802 - val_mae: 11.4786\n",
      "Epoch 96/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.2529 - mae: 11.7438 - val_loss: 10.7552 - val_mae: 11.2323\n",
      "Epoch 97/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.2196 - mae: 11.7076 - val_loss: 10.7935 - val_mae: 11.2742\n",
      "Epoch 98/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.4174 - mae: 11.9086 - val_loss: 10.6985 - val_mae: 11.1899\n",
      "Epoch 99/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0619 - mae: 11.5472 - val_loss: 10.6432 - val_mae: 11.1360\n",
      "Epoch 100/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.2981 - mae: 11.7865 - val_loss: 10.6175 - val_mae: 11.0948\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 483ms/step - loss: 80.3013 - mae: 80.7998 - val_loss: 27.8230 - val_mae: 28.3230\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 27.0582 - mae: 27.5567 - val_loss: 15.9457 - val_mae: 16.4366\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 15.6834 - mae: 16.1748 - val_loss: 28.8731 - val_mae: 29.3730\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 19.8363 - mae: 20.3313 - val_loss: 16.3355 - val_mae: 16.8355\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.9677 - mae: 15.4583 - val_loss: 16.9561 - val_mae: 17.4493\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.8825 - mae: 15.3738 - val_loss: 15.6381 - val_mae: 16.1290\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 15.6322 - mae: 16.1283 - val_loss: 15.0506 - val_mae: 15.5380\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.6267 - mae: 14.1149 - val_loss: 16.0498 - val_mae: 16.5489\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.2592 - mae: 13.7423 - val_loss: 14.8878 - val_mae: 15.3819\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4452 - mae: 13.9399 - val_loss: 15.4595 - val_mae: 15.9558\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.5847 - mae: 14.0791 - val_loss: 14.9850 - val_mae: 15.4733\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.9838 - mae: 14.4730 - val_loss: 14.9519 - val_mae: 15.4378\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 12.9614 - mae: 13.4497 - val_loss: 15.4176 - val_mae: 15.9116\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 12.8551 - mae: 13.3452 - val_loss: 15.0276 - val_mae: 15.5153\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.4554 - mae: 12.9436 - val_loss: 15.5407 - val_mae: 16.0407\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.5589 - mae: 13.0471 - val_loss: 15.0659 - val_mae: 15.5576\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.6455 - mae: 13.1334 - val_loss: 15.2737 - val_mae: 15.7737\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.2979 - mae: 12.7858 - val_loss: 15.0640 - val_mae: 15.5473\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.3489 - mae: 12.8351 - val_loss: 15.0852 - val_mae: 15.5824\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.1237 - mae: 12.6146 - val_loss: 15.0278 - val_mae: 15.5115\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.2614 - mae: 12.7468 - val_loss: 15.1974 - val_mae: 15.6956\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.9470 - mae: 12.4404 - val_loss: 15.2075 - val_mae: 15.7029\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.9179 - mae: 12.4075 - val_loss: 15.0656 - val_mae: 15.5595\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.7894 - mae: 12.2804 - val_loss: 14.8871 - val_mae: 15.3644\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7663 - mae: 12.2534 - val_loss: 15.1177 - val_mae: 15.6173\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.7082 - mae: 12.1995 - val_loss: 15.0328 - val_mae: 15.5233\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 11.6688 - mae: 12.1557 - val_loss: 15.1383 - val_mae: 15.6312\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.6443 - mae: 12.1333 - val_loss: 14.8325 - val_mae: 15.3256\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.8393 - mae: 12.3280 - val_loss: 15.2142 - val_mae: 15.7135\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.7160 - mae: 12.2037 - val_loss: 14.7523 - val_mae: 15.2426\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.6674 - mae: 12.1600 - val_loss: 15.3504 - val_mae: 15.8432\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.6006 - mae: 12.0833 - val_loss: 14.7703 - val_mae: 15.2663\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.6805 - mae: 12.1719 - val_loss: 15.6198 - val_mae: 16.1122\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.7391 - mae: 12.2253 - val_loss: 14.6955 - val_mae: 15.1827\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.6921 - mae: 12.1818 - val_loss: 15.5057 - val_mae: 16.0000\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.8755 - mae: 12.3678 - val_loss: 14.8035 - val_mae: 15.2951\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.0327 - mae: 12.5270 - val_loss: 15.5721 - val_mae: 16.0663\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.2049 - mae: 12.6970 - val_loss: 14.9771 - val_mae: 15.4695\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 14.0911 - mae: 14.5856 - val_loss: 15.3673 - val_mae: 15.8571\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.6129 - mae: 14.1065 - val_loss: 15.2753 - val_mae: 15.7675\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.2763 - mae: 13.7679 - val_loss: 15.0152 - val_mae: 15.5085\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8148 - mae: 12.3001 - val_loss: 15.7529 - val_mae: 16.2522\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.1432 - mae: 12.6363 - val_loss: 15.0659 - val_mae: 15.5566\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1493 - mae: 11.6373 - val_loss: 15.3174 - val_mae: 15.8168\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.6662 - mae: 12.1585 - val_loss: 15.1918 - val_mae: 15.6799\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.9929 - mae: 11.4778 - val_loss: 14.9768 - val_mae: 15.4684\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 11.0546 - mae: 11.5426 - val_loss: 15.6069 - val_mae: 16.0986\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.0770 - mae: 11.5641 - val_loss: 14.8286 - val_mae: 15.3100\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.8652 - mae: 11.3541 - val_loss: 15.7431 - val_mae: 16.2360\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.0276 - mae: 11.5069 - val_loss: 14.7913 - val_mae: 15.2911\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 22ms/step - loss: 10.8549 - mae: 11.3388 - val_loss: 15.5099 - val_mae: 16.0085\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.9067 - mae: 11.3940 - val_loss: 14.8754 - val_mae: 15.3751\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.8859 - mae: 11.3643 - val_loss: 15.0845 - val_mae: 15.5695\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.7319 - mae: 11.2193 - val_loss: 15.2920 - val_mae: 15.7791\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 385ms/step - loss: 85.3088 - mae: 85.8066 - val_loss: 17.9236 - val_mae: 18.4226\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 26.3505 - mae: 26.8479 - val_loss: 13.4875 - val_mae: 13.9671\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 17.9093 - mae: 18.3999 - val_loss: 23.4691 - val_mae: 23.9691\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 21.5515 - mae: 22.0457 - val_loss: 13.9494 - val_mae: 14.4299\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 18.2763 - mae: 18.7710 - val_loss: 12.3929 - val_mae: 12.8844\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 16.7078 - mae: 17.1963 - val_loss: 12.0962 - val_mae: 12.5934\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 17.4370 - mae: 17.9348 - val_loss: 12.9854 - val_mae: 13.4640\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 15.4400 - mae: 15.9287 - val_loss: 12.0335 - val_mae: 12.5233\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 15.4864 - mae: 15.9690 - val_loss: 13.1972 - val_mae: 13.6972\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.9876 - mae: 15.4830 - val_loss: 12.0765 - val_mae: 12.5477\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.1610 - mae: 15.6528 - val_loss: 13.2511 - val_mae: 13.7459\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 15.2578 - mae: 15.7530 - val_loss: 11.8105 - val_mae: 12.3010\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.8584 - mae: 15.3530 - val_loss: 11.9439 - val_mae: 12.4331\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 14.4723 - mae: 14.9627 - val_loss: 11.7596 - val_mae: 12.2460\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.1898 - mae: 14.6789 - val_loss: 11.6511 - val_mae: 12.1403\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.0798 - mae: 14.5710 - val_loss: 11.6059 - val_mae: 12.0948\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.9895 - mae: 14.4822 - val_loss: 11.5515 - val_mae: 12.0436\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.1032 - mae: 14.5916 - val_loss: 11.5724 - val_mae: 12.0649\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.9582 - mae: 14.4495 - val_loss: 11.5398 - val_mae: 12.0378\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.8936 - mae: 14.3804 - val_loss: 11.4831 - val_mae: 11.9718\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.7727 - mae: 14.2643 - val_loss: 11.4509 - val_mae: 11.9340\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.6150 - mae: 14.1042 - val_loss: 11.4292 - val_mae: 11.9086\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.5633 - mae: 14.0558 - val_loss: 11.4680 - val_mae: 11.9649\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4523 - mae: 13.9435 - val_loss: 11.4101 - val_mae: 11.8918\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.5139 - mae: 14.0017 - val_loss: 11.4397 - val_mae: 11.9262\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.5165 - mae: 14.0093 - val_loss: 11.4992 - val_mae: 11.9840\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.2569 - mae: 13.7455 - val_loss: 12.2258 - val_mae: 12.7052\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.6646 - mae: 14.1593 - val_loss: 11.4911 - val_mae: 11.9810\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.1870 - mae: 13.6773 - val_loss: 11.8399 - val_mae: 12.3235\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.4318 - mae: 13.9263 - val_loss: 11.3208 - val_mae: 11.8054\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.2703 - mae: 13.7596 - val_loss: 11.2095 - val_mae: 11.6955\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3169 - mae: 13.8042 - val_loss: 11.4896 - val_mae: 11.9743\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0761 - mae: 13.5661 - val_loss: 11.1513 - val_mae: 11.6304\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8378 - mae: 13.3218 - val_loss: 11.4406 - val_mae: 11.9148\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.0506 - mae: 13.5388 - val_loss: 11.3743 - val_mae: 11.8559\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.8992 - mae: 13.3819 - val_loss: 11.3031 - val_mae: 11.7914\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3394 - mae: 13.8326 - val_loss: 12.4762 - val_mae: 12.9572\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.8918 - mae: 13.3817 - val_loss: 11.8272 - val_mae: 12.3222\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.1149 - mae: 13.6032 - val_loss: 12.5986 - val_mae: 13.0789\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.0165 - mae: 13.5052 - val_loss: 11.2742 - val_mae: 11.7731\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.6893 - mae: 13.1815 - val_loss: 11.3211 - val_mae: 11.8156\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.5979 - mae: 13.0849 - val_loss: 12.0479 - val_mae: 12.5297\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.7028 - mae: 13.1894 - val_loss: 11.2059 - val_mae: 11.7012\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.3974 - mae: 12.8873 - val_loss: 11.3896 - val_mae: 11.8832\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.3273 - mae: 12.8181 - val_loss: 11.4317 - val_mae: 11.9263\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2233 - mae: 12.7032 - val_loss: 11.3336 - val_mae: 11.8215\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.4810 - mae: 12.9730 - val_loss: 12.9303 - val_mae: 13.4226\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.8076 - mae: 13.2969 - val_loss: 11.2308 - val_mae: 11.7145\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.2300 - mae: 12.7214 - val_loss: 11.4539 - val_mae: 11.9412\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.1620 - mae: 12.6543 - val_loss: 11.4478 - val_mae: 11.9407\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.0865 - mae: 12.5791 - val_loss: 11.4041 - val_mae: 11.9014\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2606 - mae: 12.7475 - val_loss: 11.7707 - val_mae: 12.2527\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.2086 - mae: 12.6993 - val_loss: 11.4291 - val_mae: 11.9287\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 372ms/step - loss: 85.2701 - mae: 85.7695 - val_loss: 26.3836 - val_mae: 26.8835\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 27.7605 - mae: 28.2559 - val_loss: 15.3029 - val_mae: 15.7934\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 18.9142 - mae: 19.4071 - val_loss: 29.7689 - val_mae: 30.2649\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 23.8253 - mae: 24.3186 - val_loss: 14.2810 - val_mae: 14.7664\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 18.9061 - mae: 19.4032 - val_loss: 15.2618 - val_mae: 15.7611\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 17.6791 - mae: 18.1725 - val_loss: 14.4078 - val_mae: 14.9052\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 19.1009 - mae: 19.5974 - val_loss: 14.6684 - val_mae: 15.1651\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 16.3487 - mae: 16.8460 - val_loss: 14.9124 - val_mae: 15.3988\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 16.5064 - mae: 16.9921 - val_loss: 14.2219 - val_mae: 14.7190\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.1143 - mae: 16.6069 - val_loss: 14.7059 - val_mae: 15.1907\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 16.3915 - mae: 16.8806 - val_loss: 14.0445 - val_mae: 14.5270\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 18.0674 - mae: 18.5655 - val_loss: 14.4097 - val_mae: 14.9016\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 15.7521 - mae: 16.2457 - val_loss: 14.6127 - val_mae: 15.1047\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 15.9070 - mae: 16.4012 - val_loss: 13.9175 - val_mae: 14.4002\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 15.1288 - mae: 15.6245 - val_loss: 14.9627 - val_mae: 15.4503\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.4909 - mae: 15.9712 - val_loss: 13.8008 - val_mae: 14.2940\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 15.8979 - mae: 16.3957 - val_loss: 13.9538 - val_mae: 14.4477\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 15.6375 - mae: 16.1315 - val_loss: 13.7724 - val_mae: 14.2681\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.2816 - mae: 16.7768 - val_loss: 13.6638 - val_mae: 14.1603\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 15.1802 - mae: 15.6705 - val_loss: 14.1467 - val_mae: 14.6440\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 15.5749 - mae: 16.0645 - val_loss: 13.5692 - val_mae: 14.0688\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.5348 - mae: 15.0272 - val_loss: 14.3267 - val_mae: 14.8241\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.8523 - mae: 15.3462 - val_loss: 13.5621 - val_mae: 14.0403\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.3036 - mae: 14.7923 - val_loss: 14.0455 - val_mae: 14.5366\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.5257 - mae: 15.0174 - val_loss: 14.0640 - val_mae: 14.5565\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.2420 - mae: 14.7278 - val_loss: 13.7706 - val_mae: 14.2656\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 14.3832 - mae: 14.8742 - val_loss: 14.0721 - val_mae: 14.5612\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1340 - mae: 14.6233 - val_loss: 13.5307 - val_mae: 14.0222\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.4869 - mae: 14.9800 - val_loss: 14.7182 - val_mae: 15.2133\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.1733 - mae: 14.6552 - val_loss: 13.4337 - val_mae: 13.9297\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.0812 - mae: 14.5734 - val_loss: 14.2689 - val_mae: 14.7604\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.7999 - mae: 14.2871 - val_loss: 13.4416 - val_mae: 13.9356\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.8048 - mae: 14.2989 - val_loss: 14.6957 - val_mae: 15.1933\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.8955 - mae: 14.3815 - val_loss: 13.3453 - val_mae: 13.8418\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.5672 - mae: 14.0538 - val_loss: 14.1489 - val_mae: 14.6398\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.4858 - mae: 13.9732 - val_loss: 13.3955 - val_mae: 13.8906\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.6275 - mae: 14.1167 - val_loss: 14.4644 - val_mae: 14.9637\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.3810 - mae: 13.8686 - val_loss: 13.3775 - val_mae: 13.8647\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.5633 - mae: 14.0568 - val_loss: 15.1909 - val_mae: 15.6894\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.6434 - mae: 14.1337 - val_loss: 13.6195 - val_mae: 14.1075\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 13.6976 - mae: 14.1887 - val_loss: 14.9266 - val_mae: 15.4206\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.4330 - mae: 13.9187 - val_loss: 13.2185 - val_mae: 13.7055\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.1152 - mae: 13.6043 - val_loss: 15.3603 - val_mae: 15.8596\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.4209 - mae: 13.9064 - val_loss: 13.1481 - val_mae: 13.6361\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.9448 - mae: 13.4292 - val_loss: 14.0687 - val_mae: 14.5612\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.9983 - mae: 13.4900 - val_loss: 13.2466 - val_mae: 13.7330\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.7430 - mae: 13.2348 - val_loss: 13.4741 - val_mae: 13.9675\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.8534 - mae: 13.3475 - val_loss: 14.2193 - val_mae: 14.7181\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8924 - mae: 13.3830 - val_loss: 13.3362 - val_mae: 13.8280\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.6222 - mae: 13.1109 - val_loss: 13.0248 - val_mae: 13.5138\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.0373 - mae: 13.5275 - val_loss: 15.5679 - val_mae: 16.0679\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.2496 - mae: 13.7354 - val_loss: 13.1264 - val_mae: 13.6111\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.5498 - mae: 13.0371 - val_loss: 13.0021 - val_mae: 13.4877\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0287 - mae: 13.5224 - val_loss: 16.1500 - val_mae: 16.6500\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.5184 - mae: 14.0131 - val_loss: 13.4826 - val_mae: 13.9826\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 12.9608 - mae: 13.4512 - val_loss: 15.0505 - val_mae: 15.5491\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.9167 - mae: 13.4049 - val_loss: 12.9274 - val_mae: 13.4225\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.4347 - mae: 12.9259 - val_loss: 12.8131 - val_mae: 13.2955\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.4238 - mae: 13.9156 - val_loss: 16.4719 - val_mae: 16.9646\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.7467 - mae: 14.2416 - val_loss: 13.1075 - val_mae: 13.5923\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8918 - mae: 13.3808 - val_loss: 13.3744 - val_mae: 13.8740\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0147 - mae: 13.5088 - val_loss: 13.9577 - val_mae: 14.4536\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.6132 - mae: 13.1029 - val_loss: 12.8632 - val_mae: 13.3535\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0169 - mae: 13.4993 - val_loss: 15.8009 - val_mae: 16.2972\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.1062 - mae: 13.5944 - val_loss: 12.8895 - val_mae: 13.3886\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.3546 - mae: 12.8465 - val_loss: 13.1139 - val_mae: 13.6075\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.8316 - mae: 13.3239 - val_loss: 14.5435 - val_mae: 15.0390\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.5483 - mae: 13.0345 - val_loss: 14.5531 - val_mae: 15.0302\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.6836 - mae: 14.1758 - val_loss: 16.3936 - val_mae: 16.8664\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.5430 - mae: 14.0332 - val_loss: 12.7738 - val_mae: 13.2595\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.1542 - mae: 12.6394 - val_loss: 13.0154 - val_mae: 13.5078\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.9045 - mae: 13.3944 - val_loss: 14.3073 - val_mae: 14.8006\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.9303 - mae: 13.4157 - val_loss: 13.5054 - val_mae: 13.9919\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.5578 - mae: 14.0475 - val_loss: 15.5670 - val_mae: 16.0615\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8223 - mae: 13.3098 - val_loss: 13.0042 - val_mae: 13.4979\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.5400 - mae: 13.0301 - val_loss: 13.9672 - val_mae: 14.4631\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.1105 - mae: 12.5976 - val_loss: 12.8298 - val_mae: 13.3177\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1144 - mae: 12.6022 - val_loss: 13.7438 - val_mae: 14.2408\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.0068 - mae: 12.4960 - val_loss: 12.8438 - val_mae: 13.3360\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.8342 - mae: 12.3222 - val_loss: 12.7394 - val_mae: 13.2353\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0546 - mae: 12.5404 - val_loss: 13.6326 - val_mae: 14.1210\n",
      "Epoch 82/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.9876 - mae: 12.4789 - val_loss: 12.9128 - val_mae: 13.4104\n",
      "Epoch 83/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.7450 - mae: 12.2322 - val_loss: 12.8458 - val_mae: 13.3456\n",
      "Epoch 84/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.1960 - mae: 12.6897 - val_loss: 13.5557 - val_mae: 14.0475\n",
      "Epoch 85/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2245 - mae: 12.7186 - val_loss: 14.8247 - val_mae: 15.3190\n",
      "Epoch 86/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.4005 - mae: 12.8897 - val_loss: 12.8592 - val_mae: 13.3418\n",
      "Epoch 87/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.7563 - mae: 12.2478 - val_loss: 13.4473 - val_mae: 13.9352\n",
      "Epoch 88/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0017 - mae: 13.4881 - val_loss: 14.1301 - val_mae: 14.6170\n",
      "Epoch 89/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.9988 - mae: 12.4904 - val_loss: 13.4521 - val_mae: 13.9370\n",
      "Epoch 90/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.6874 - mae: 12.1772 - val_loss: 12.6012 - val_mae: 13.0945\n",
      "Epoch 91/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.0386 - mae: 12.5227 - val_loss: 13.2716 - val_mae: 13.7654\n",
      "Epoch 92/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.2857 - mae: 12.7748 - val_loss: 14.5724 - val_mae: 15.0702\n",
      "Epoch 93/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.0875 - mae: 12.5783 - val_loss: 12.8299 - val_mae: 13.3061\n",
      "Epoch 94/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.6118 - mae: 12.0989 - val_loss: 12.8834 - val_mae: 13.3756\n",
      "Epoch 95/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.0554 - mae: 12.5440 - val_loss: 13.2152 - val_mae: 13.7065\n",
      "Epoch 96/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.4112 - mae: 12.8994 - val_loss: 14.6321 - val_mae: 15.1272\n",
      "Epoch 97/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5556 - mae: 13.0414 - val_loss: 13.9685 - val_mae: 14.4629\n",
      "Epoch 98/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.8558 - mae: 12.3481 - val_loss: 12.9068 - val_mae: 13.4032\n",
      "Epoch 99/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8657 - mae: 12.3535 - val_loss: 13.5842 - val_mae: 14.0706\n",
      "Epoch 100/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.6700 - mae: 13.1624 - val_loss: 14.8206 - val_mae: 15.3154\n",
      "Epoch 101/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.5518 - mae: 13.0430 - val_loss: 13.5658 - val_mae: 14.0580\n",
      "Epoch 102/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.5799 - mae: 12.0738 - val_loss: 12.5654 - val_mae: 13.0473\n",
      "Epoch 103/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.9216 - mae: 12.4157 - val_loss: 13.9113 - val_mae: 14.3943\n",
      "Epoch 104/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8070 - mae: 12.2948 - val_loss: 13.2713 - val_mae: 13.7687\n",
      "Epoch 105/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.4821 - mae: 11.9716 - val_loss: 12.8471 - val_mae: 13.3426\n",
      "Epoch 106/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7802 - mae: 12.2685 - val_loss: 12.8417 - val_mae: 13.3285\n",
      "Epoch 107/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6733 - mae: 13.1660 - val_loss: 16.6272 - val_mae: 17.1272\n",
      "Epoch 108/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.5626 - mae: 13.0560 - val_loss: 12.6952 - val_mae: 13.1917\n",
      "Epoch 109/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.4650 - mae: 11.9556 - val_loss: 12.6111 - val_mae: 13.0943\n",
      "Epoch 110/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7610 - mae: 12.2527 - val_loss: 12.7583 - val_mae: 13.2535\n",
      "Epoch 111/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.4413 - mae: 11.9281 - val_loss: 13.7577 - val_mae: 14.2577\n",
      "Epoch 112/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3736 - mae: 11.8565 - val_loss: 13.1816 - val_mae: 13.6797\n",
      "Epoch 113/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 11.1999 - mae: 11.6830 - val_loss: 12.7264 - val_mae: 13.2056\n",
      "Epoch 114/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.2595 - mae: 11.7466 - val_loss: 12.7106 - val_mae: 13.2023\n",
      "Epoch 115/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.9287 - mae: 12.4214 - val_loss: 13.2560 - val_mae: 13.7407\n",
      "Epoch 116/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.3779 - mae: 11.8700 - val_loss: 13.8993 - val_mae: 14.3993\n",
      "Epoch 117/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.6364 - mae: 12.1286 - val_loss: 14.9371 - val_mae: 15.4277\n",
      "Epoch 118/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.6031 - mae: 12.0852 - val_loss: 13.1328 - val_mae: 13.6192\n",
      "Epoch 119/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.3671 - mae: 13.8559 - val_loss: 15.0770 - val_mae: 15.5635\n",
      "Epoch 120/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.6956 - mae: 14.1857 - val_loss: 15.9890 - val_mae: 16.4802\n",
      "Epoch 121/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.9876 - mae: 14.4812 - val_loss: 14.3120 - val_mae: 14.8120\n",
      "Epoch 122/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.9126 - mae: 12.3995 - val_loss: 12.7261 - val_mae: 13.2241\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 389ms/step - loss: 85.2238 - mae: 85.7224 - val_loss: 23.9363 - val_mae: 24.4351\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 26.4612 - mae: 26.9585 - val_loss: 11.3787 - val_mae: 11.8581\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 21.9136 - mae: 22.4100 - val_loss: 10.8247 - val_mae: 11.3169\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 17.9724 - mae: 18.4599 - val_loss: 14.6603 - val_mae: 15.1547\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 19.6180 - mae: 20.1144 - val_loss: 13.1887 - val_mae: 13.6813\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 15.9510 - mae: 16.4461 - val_loss: 12.9508 - val_mae: 13.4470\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 16.0447 - mae: 16.5343 - val_loss: 13.0017 - val_mae: 13.4958\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.6457 - mae: 15.1410 - val_loss: 13.8061 - val_mae: 14.3023\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 14.5386 - mae: 15.0298 - val_loss: 12.2515 - val_mae: 12.7487\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.4408 - mae: 14.9372 - val_loss: 14.6080 - val_mae: 15.1052\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.0660 - mae: 14.5604 - val_loss: 12.0299 - val_mae: 12.5151\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.4024 - mae: 13.9003 - val_loss: 10.6874 - val_mae: 11.1801\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5046 - mae: 13.9935 - val_loss: 10.8465 - val_mae: 11.3285\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.1754 - mae: 13.6650 - val_loss: 10.5918 - val_mae: 11.0783\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0987 - mae: 13.5915 - val_loss: 11.5695 - val_mae: 12.0455\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.7941 - mae: 13.2820 - val_loss: 10.9865 - val_mae: 11.4770\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.5709 - mae: 13.0518 - val_loss: 10.2872 - val_mae: 10.7834\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6397 - mae: 13.1342 - val_loss: 10.9834 - val_mae: 11.4725\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.5028 - mae: 12.9924 - val_loss: 10.0381 - val_mae: 10.5228\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.2950 - mae: 12.7830 - val_loss: 11.7262 - val_mae: 12.2132\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.3770 - mae: 12.8690 - val_loss: 10.8954 - val_mae: 11.3823\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.6672 - mae: 13.1546 - val_loss: 12.3477 - val_mae: 12.8371\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.6015 - mae: 13.0976 - val_loss: 10.2853 - val_mae: 10.7687\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.3355 - mae: 12.8257 - val_loss: 11.2355 - val_mae: 11.7199\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.5579 - mae: 13.0441 - val_loss: 9.9813 - val_mae: 10.4740\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 12.3834 - mae: 12.8787 - val_loss: 10.3207 - val_mae: 10.8114\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7849 - mae: 12.2260 - val_loss: 10.0510 - val_mae: 10.5466\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.9650 - mae: 12.4596 - val_loss: 9.9300 - val_mae: 10.4148\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.6875 - mae: 12.1568 - val_loss: 9.7337 - val_mae: 10.2147\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.4183 - mae: 11.9063 - val_loss: 10.0683 - val_mae: 10.5489\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3650 - mae: 11.7898 - val_loss: 11.6796 - val_mae: 12.1746\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.1428 - mae: 12.6340 - val_loss: 10.9497 - val_mae: 11.4378\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.7078 - mae: 13.1966 - val_loss: 10.8195 - val_mae: 11.3076\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.2052 - mae: 11.6960 - val_loss: 11.4025 - val_mae: 11.8925\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.3443 - mae: 12.8382 - val_loss: 11.6022 - val_mae: 12.0962\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.1544 - mae: 12.6153 - val_loss: 10.0202 - val_mae: 10.5081\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.8983 - mae: 12.3893 - val_loss: 10.0081 - val_mae: 10.5044\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1989 - mae: 11.6899 - val_loss: 10.0638 - val_mae: 10.5511\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.0438 - mae: 11.5128 - val_loss: 10.3622 - val_mae: 10.8513\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.1217 - mae: 11.5702 - val_loss: 9.9861 - val_mae: 10.4716\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.7885 - mae: 11.2704 - val_loss: 9.9455 - val_mae: 10.4378\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.9023 - mae: 11.3667 - val_loss: 10.0307 - val_mae: 10.5246\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 10.6790 - mae: 11.1349 - val_loss: 11.5277 - val_mae: 12.0130\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.2387 - mae: 11.7243 - val_loss: 10.9761 - val_mae: 11.4660\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.6796 - mae: 12.1669 - val_loss: 12.0275 - val_mae: 12.5195\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.4750 - mae: 11.9029 - val_loss: 11.2850 - val_mae: 11.7846\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.5580 - mae: 12.0316 - val_loss: 12.4567 - val_mae: 12.9567\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.6733 - mae: 12.1614 - val_loss: 12.3192 - val_mae: 12.8185\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2980 - mae: 12.7447 - val_loss: 13.5429 - val_mae: 14.0297\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 3s 372ms/step - loss: 83.8256 - mae: 84.3240 - val_loss: 15.4440 - val_mae: 15.9435\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 26.0268 - mae: 26.5257 - val_loss: 13.9536 - val_mae: 14.4452\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 17.2810 - mae: 17.7745 - val_loss: 22.6842 - val_mae: 23.1842\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 22.7158 - mae: 23.2055 - val_loss: 13.7533 - val_mae: 14.2527\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 18.9133 - mae: 19.4100 - val_loss: 10.6655 - val_mae: 11.1611\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.9071 - mae: 16.3964 - val_loss: 10.6711 - val_mae: 11.1598\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 17.2734 - mae: 17.7650 - val_loss: 13.6151 - val_mae: 14.1029\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.8386 - mae: 15.3244 - val_loss: 11.0968 - val_mae: 11.5734\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 15.9704 - mae: 16.4649 - val_loss: 11.3802 - val_mae: 11.8745\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 16.7494 - mae: 17.2450 - val_loss: 12.0188 - val_mae: 12.5060\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.1046 - mae: 15.5960 - val_loss: 10.5278 - val_mae: 11.0236\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.7340 - mae: 15.2212 - val_loss: 12.1242 - val_mae: 12.6106\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.0827 - mae: 14.5770 - val_loss: 10.3959 - val_mae: 10.8923\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.3343 - mae: 14.8312 - val_loss: 11.0134 - val_mae: 11.5044\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.7300 - mae: 15.2251 - val_loss: 10.2210 - val_mae: 10.6982\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.2265 - mae: 14.7203 - val_loss: 10.0915 - val_mae: 10.5669\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.2935 - mae: 14.7878 - val_loss: 10.4938 - val_mae: 10.9693\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.6272 - mae: 14.1198 - val_loss: 10.2328 - val_mae: 10.7300\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.7872 - mae: 14.2811 - val_loss: 10.2460 - val_mae: 10.7253\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.4164 - mae: 13.9118 - val_loss: 10.1437 - val_mae: 10.6356\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 23ms/step - loss: 13.5185 - mae: 14.0120 - val_loss: 9.8610 - val_mae: 10.3440\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.2258 - mae: 13.7187 - val_loss: 9.8840 - val_mae: 10.3801\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.1962 - mae: 13.6872 - val_loss: 9.7618 - val_mae: 10.2535\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.1331 - mae: 13.6224 - val_loss: 9.7822 - val_mae: 10.2741\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0247 - mae: 13.5147 - val_loss: 9.7725 - val_mae: 10.2574\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.0447 - mae: 13.5316 - val_loss: 9.6644 - val_mae: 10.1535\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.7990 - mae: 13.2929 - val_loss: 10.0009 - val_mae: 10.4915\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.9366 - mae: 13.4240 - val_loss: 9.6211 - val_mae: 10.0966\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.6591 - mae: 13.1540 - val_loss: 10.1247 - val_mae: 10.6155\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 12.8127 - mae: 13.2990 - val_loss: 9.6874 - val_mae: 10.1684\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.5952 - mae: 13.0881 - val_loss: 9.6151 - val_mae: 10.0951\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.5466 - mae: 13.0404 - val_loss: 9.6985 - val_mae: 10.1905\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.3841 - mae: 12.8710 - val_loss: 9.5830 - val_mae: 10.0773\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.3143 - mae: 12.7940 - val_loss: 9.5769 - val_mae: 10.0625\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.4694 - mae: 12.9582 - val_loss: 10.4077 - val_mae: 10.8990\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.9283 - mae: 13.4180 - val_loss: 9.9538 - val_mae: 10.4366\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.7615 - mae: 13.2559 - val_loss: 10.3455 - val_mae: 10.8368\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.3205 - mae: 12.8057 - val_loss: 10.5562 - val_mae: 11.0560\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.7125 - mae: 13.2082 - val_loss: 10.4972 - val_mae: 10.9917\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2540 - mae: 12.7412 - val_loss: 9.7646 - val_mae: 10.2281\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.2312 - mae: 12.7184 - val_loss: 9.8135 - val_mae: 10.3030\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.1450 - mae: 12.6335 - val_loss: 9.8597 - val_mae: 10.3497\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.9728 - mae: 12.4569 - val_loss: 9.6240 - val_mae: 10.1184\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.8291 - mae: 12.3125 - val_loss: 10.0870 - val_mae: 10.5584\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.8776 - mae: 12.3564 - val_loss: 9.7066 - val_mae: 10.1902\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 24ms/step - loss: 11.7003 - mae: 12.1861 - val_loss: 9.7103 - val_mae: 10.2008\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8897 - mae: 12.3777 - val_loss: 9.8847 - val_mae: 10.3597\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.9242 - mae: 12.4134 - val_loss: 10.5847 - val_mae: 11.0812\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.2385 - mae: 12.7298 - val_loss: 9.8857 - val_mae: 10.3652\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 11.8005 - mae: 12.2888 - val_loss: 9.8912 - val_mae: 10.3817\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.3105 - mae: 12.8009 - val_loss: 12.8041 - val_mae: 13.2959\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.2530 - mae: 13.7407 - val_loss: 11.0439 - val_mae: 11.5334\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.4573 - mae: 12.9494 - val_loss: 11.2779 - val_mae: 11.7667\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.4303 - mae: 12.9246 - val_loss: 9.9005 - val_mae: 10.3975\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 397ms/step - loss: 84.5819 - mae: 85.0813 - val_loss: 20.2198 - val_mae: 20.7115\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 27.7805 - mae: 28.2768 - val_loss: 13.3211 - val_mae: 13.8127\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 19.7961 - mae: 20.2931 - val_loss: 21.7150 - val_mae: 22.2139\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 21.3166 - mae: 21.8056 - val_loss: 12.2628 - val_mae: 12.7491\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 17.6633 - mae: 18.1580 - val_loss: 12.6356 - val_mae: 13.1328\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 16.2624 - mae: 16.7562 - val_loss: 13.5083 - val_mae: 13.9922\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 17.1571 - mae: 17.6498 - val_loss: 14.3109 - val_mae: 14.8031\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 16.6578 - mae: 17.1487 - val_loss: 12.3175 - val_mae: 12.8038\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 16.1297 - mae: 16.6189 - val_loss: 15.1622 - val_mae: 15.6622\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.3228 - mae: 15.8177 - val_loss: 11.9340 - val_mae: 12.4291\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 15.1428 - mae: 15.6346 - val_loss: 14.5032 - val_mae: 14.9996\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 16.3146 - mae: 16.8111 - val_loss: 12.6037 - val_mae: 13.0904\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 15.7932 - mae: 16.2845 - val_loss: 12.2580 - val_mae: 12.7499\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 15.6537 - mae: 16.1497 - val_loss: 14.3156 - val_mae: 14.8156\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.7259 - mae: 15.2134 - val_loss: 12.4075 - val_mae: 12.8973\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.4613 - mae: 14.9518 - val_loss: 13.4117 - val_mae: 13.9093\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.2901 - mae: 14.7790 - val_loss: 12.0318 - val_mae: 12.5151\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.1214 - mae: 14.6142 - val_loss: 11.7482 - val_mae: 12.2234\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.0468 - mae: 14.5413 - val_loss: 11.5953 - val_mae: 12.0654\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.9948 - mae: 14.4852 - val_loss: 11.3934 - val_mae: 11.8762\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.8882 - mae: 14.3794 - val_loss: 11.4760 - val_mae: 11.9405\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.8966 - mae: 14.3858 - val_loss: 11.4607 - val_mae: 11.9515\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.8931 - mae: 14.3868 - val_loss: 11.8077 - val_mae: 12.2892\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.7323 - mae: 14.2226 - val_loss: 11.4209 - val_mae: 11.9155\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.5730 - mae: 14.0658 - val_loss: 11.4590 - val_mae: 11.9402\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.5752 - mae: 14.0599 - val_loss: 11.6593 - val_mae: 12.1574\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.6772 - mae: 14.1685 - val_loss: 11.9083 - val_mae: 12.3907\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9254 - mae: 14.4199 - val_loss: 11.8323 - val_mae: 12.3252\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.1606 - mae: 14.6507 - val_loss: 13.0906 - val_mae: 13.5841\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.6755 - mae: 16.1679 - val_loss: 11.9800 - val_mae: 12.4637\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.5907 - mae: 15.0857 - val_loss: 11.3724 - val_mae: 11.8563\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.8531 - mae: 15.3491 - val_loss: 11.5329 - val_mae: 12.0093\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.7104 - mae: 14.2000 - val_loss: 11.2730 - val_mae: 11.7691\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.5668 - mae: 15.0614 - val_loss: 11.4211 - val_mae: 11.8998\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.6701 - mae: 14.1611 - val_loss: 11.4184 - val_mae: 11.9035\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.8779 - mae: 14.3705 - val_loss: 11.3543 - val_mae: 11.8462\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3838 - mae: 13.8755 - val_loss: 11.6290 - val_mae: 12.0983\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.3606 - mae: 13.8528 - val_loss: 11.5728 - val_mae: 12.0667\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.8481 - mae: 13.3339 - val_loss: 12.1006 - val_mae: 12.5990\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.0495 - mae: 13.5416 - val_loss: 11.6891 - val_mae: 12.1764\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.7624 - mae: 13.2475 - val_loss: 12.1505 - val_mae: 12.6478\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9586 - mae: 13.4474 - val_loss: 11.8614 - val_mae: 12.3462\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.8846 - mae: 13.3730 - val_loss: 12.4053 - val_mae: 12.8917\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.8039 - mae: 13.3007 - val_loss: 11.8050 - val_mae: 12.2762\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.6927 - mae: 13.1818 - val_loss: 13.0673 - val_mae: 13.5673\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.1705 - mae: 13.6642 - val_loss: 11.6251 - val_mae: 12.1040\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.0082 - mae: 13.5002 - val_loss: 13.5225 - val_mae: 14.0225\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.2417 - mae: 14.7351 - val_loss: 11.5501 - val_mae: 12.0432\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.5142 - mae: 14.0050 - val_loss: 12.3551 - val_mae: 12.8445\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.8295 - mae: 14.3269 - val_loss: 11.4498 - val_mae: 11.9409\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0766 - mae: 13.5624 - val_loss: 12.5465 - val_mae: 13.0388\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.0304 - mae: 14.5265 - val_loss: 11.5050 - val_mae: 11.9912\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1315 - mae: 14.6232 - val_loss: 12.6065 - val_mae: 13.1059\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 541ms/step - loss: 78.2943 - mae: 78.7896 - val_loss: 12.8219 - val_mae: 13.3138\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 28.2772 - mae: 28.7760 - val_loss: 12.0260 - val_mae: 12.5224\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 16.0337 - mae: 16.5316 - val_loss: 16.0597 - val_mae: 16.5563\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 18.2006 - mae: 18.6919 - val_loss: 10.2042 - val_mae: 10.6924\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 19.7984 - mae: 20.2977 - val_loss: 10.6190 - val_mae: 11.1010\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1861 - mae: 14.6784 - val_loss: 9.5775 - val_mae: 10.0547\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.1150 - mae: 14.6035 - val_loss: 9.3962 - val_mae: 9.8730\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.9272 - mae: 14.4198 - val_loss: 10.1655 - val_mae: 10.6649\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.3370 - mae: 12.8277 - val_loss: 9.0540 - val_mae: 9.5512\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2714 - mae: 12.7586 - val_loss: 9.1786 - val_mae: 9.6784\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 12.1085 - mae: 12.5959 - val_loss: 9.9062 - val_mae: 10.4024\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8708 - mae: 12.3596 - val_loss: 9.2366 - val_mae: 9.7245\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.7991 - mae: 12.2882 - val_loss: 9.3847 - val_mae: 9.8821\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7702 - mae: 12.2661 - val_loss: 9.6770 - val_mae: 10.1716\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.4880 - mae: 11.9740 - val_loss: 9.0269 - val_mae: 9.5227\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.4969 - mae: 11.9872 - val_loss: 9.3795 - val_mae: 9.8792\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5003 - mae: 11.9902 - val_loss: 9.1365 - val_mae: 9.6252\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.1553 - mae: 11.6440 - val_loss: 8.8110 - val_mae: 9.2974\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.1464 - mae: 11.6272 - val_loss: 8.8946 - val_mae: 9.3884\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0492 - mae: 11.5357 - val_loss: 8.7205 - val_mae: 9.2160\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9434 - mae: 11.4274 - val_loss: 8.7057 - val_mae: 9.2013\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.9862 - mae: 11.4777 - val_loss: 8.6956 - val_mae: 9.1870\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.7556 - mae: 11.2379 - val_loss: 8.6041 - val_mae: 9.0849\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.8039 - mae: 11.2900 - val_loss: 8.6460 - val_mae: 9.1364\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.7727 - mae: 11.2605 - val_loss: 8.6235 - val_mae: 9.1017\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.6861 - mae: 11.1752 - val_loss: 8.6050 - val_mae: 9.1027\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.6441 - mae: 11.1289 - val_loss: 8.5697 - val_mae: 9.0355\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.5281 - mae: 11.0224 - val_loss: 8.4935 - val_mae: 8.9918\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.5222 - mae: 11.0081 - val_loss: 8.4259 - val_mae: 8.9038\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.4180 - mae: 10.9065 - val_loss: 8.3803 - val_mae: 8.8697\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.3830 - mae: 10.8720 - val_loss: 8.4914 - val_mae: 8.9726\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 10.4761 - mae: 10.9658 - val_loss: 8.5450 - val_mae: 9.0241\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.5140 - mae: 10.9992 - val_loss: 9.0190 - val_mae: 9.5123\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.8674 - mae: 11.3586 - val_loss: 9.2966 - val_mae: 9.7881\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.0267 - mae: 11.5115 - val_loss: 9.3082 - val_mae: 9.7901\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.2603 - mae: 11.7497 - val_loss: 8.5219 - val_mae: 9.0036\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.9457 - mae: 11.4360 - val_loss: 8.4404 - val_mae: 8.9251\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.8264 - mae: 11.3113 - val_loss: 8.5080 - val_mae: 9.0024\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.3409 - mae: 10.8255 - val_loss: 8.5811 - val_mae: 9.0589\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.5171 - mae: 11.0060 - val_loss: 8.5584 - val_mae: 9.0542\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.1785 - mae: 10.6683 - val_loss: 8.5993 - val_mae: 9.0732\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.2834 - mae: 10.7665 - val_loss: 8.5355 - val_mae: 9.0269\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.1284 - mae: 10.6167 - val_loss: 8.5108 - val_mae: 9.0049\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.1426 - mae: 10.6312 - val_loss: 8.5401 - val_mae: 9.0370\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.1458 - mae: 10.6366 - val_loss: 8.4714 - val_mae: 8.9584\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.1025 - mae: 10.5953 - val_loss: 8.4611 - val_mae: 8.9536\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.1411 - mae: 10.6301 - val_loss: 8.4328 - val_mae: 8.9232\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.0078 - mae: 10.4970 - val_loss: 8.4365 - val_mae: 8.9306\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.1084 - mae: 10.5952 - val_loss: 8.4174 - val_mae: 8.9012\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.9972 - mae: 10.4836 - val_loss: 8.4814 - val_mae: 8.9738\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 435ms/step - loss: 80.4344 - mae: 80.9344 - val_loss: 12.1454 - val_mae: 12.6229\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 27.6857 - mae: 28.1839 - val_loss: 9.9428 - val_mae: 10.4293\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.6642 - mae: 16.1607 - val_loss: 19.5826 - val_mae: 20.0826\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 19.4099 - mae: 19.9045 - val_loss: 12.9783 - val_mae: 13.4640\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.0316 - mae: 16.5290 - val_loss: 10.3390 - val_mae: 10.8166\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 15.1548 - mae: 15.6438 - val_loss: 11.2292 - val_mae: 11.7189\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 16.9131 - mae: 17.4115 - val_loss: 13.8962 - val_mae: 14.3866\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.6573 - mae: 14.1472 - val_loss: 10.0028 - val_mae: 10.4883\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.6916 - mae: 14.1804 - val_loss: 11.7148 - val_mae: 12.2073\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.7288 - mae: 14.2224 - val_loss: 9.9943 - val_mae: 10.4789\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.9487 - mae: 14.4373 - val_loss: 10.8011 - val_mae: 11.2853\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 15.0349 - mae: 15.5280 - val_loss: 11.8331 - val_mae: 12.3277\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3632 - mae: 13.8578 - val_loss: 9.6917 - val_mae: 10.1858\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.1872 - mae: 13.6753 - val_loss: 10.8836 - val_mae: 11.3798\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.3976 - mae: 12.8783 - val_loss: 9.5487 - val_mae: 10.0109\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.7708 - mae: 13.2618 - val_loss: 10.3449 - val_mae: 10.8352\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.3294 - mae: 12.8180 - val_loss: 9.3688 - val_mae: 9.8405\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.4530 - mae: 12.9359 - val_loss: 9.8657 - val_mae: 10.3624\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.0718 - mae: 12.5535 - val_loss: 9.3164 - val_mae: 9.8003\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.1935 - mae: 12.6813 - val_loss: 9.7864 - val_mae: 10.2812\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.0056 - mae: 12.4915 - val_loss: 9.1949 - val_mae: 9.6776\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.0856 - mae: 12.5680 - val_loss: 9.2660 - val_mae: 9.7576\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.7809 - mae: 12.2698 - val_loss: 9.0472 - val_mae: 9.5315\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.7566 - mae: 12.2450 - val_loss: 9.1099 - val_mae: 9.5908\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.6957 - mae: 12.1827 - val_loss: 8.9989 - val_mae: 9.4693\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.7587 - mae: 12.2502 - val_loss: 9.0725 - val_mae: 9.5621\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5348 - mae: 12.0246 - val_loss: 9.0438 - val_mae: 9.5313\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.6338 - mae: 12.1193 - val_loss: 8.9525 - val_mae: 9.4359\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.4728 - mae: 11.9561 - val_loss: 9.1087 - val_mae: 9.5841\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.5278 - mae: 12.0208 - val_loss: 8.9355 - val_mae: 9.4188\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.4295 - mae: 11.9135 - val_loss: 8.8665 - val_mae: 9.3373\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.3394 - mae: 11.8277 - val_loss: 8.9084 - val_mae: 9.3918\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.3143 - mae: 11.7998 - val_loss: 9.2891 - val_mae: 9.7778\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3917 - mae: 11.8837 - val_loss: 9.0287 - val_mae: 9.5228\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.2591 - mae: 11.7492 - val_loss: 9.4008 - val_mae: 9.8934\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5220 - mae: 12.0119 - val_loss: 9.1963 - val_mae: 9.6953\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.4727 - mae: 11.9622 - val_loss: 9.9795 - val_mae: 10.4643\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.3960 - mae: 11.8855 - val_loss: 9.5876 - val_mae: 10.0799\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5218 - mae: 12.0133 - val_loss: 10.7737 - val_mae: 11.2632\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7316 - mae: 12.2174 - val_loss: 9.7045 - val_mae: 10.2003\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.6374 - mae: 12.1185 - val_loss: 10.2568 - val_mae: 10.7537\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.3993 - mae: 11.8896 - val_loss: 9.3661 - val_mae: 9.8578\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3647 - mae: 11.8541 - val_loss: 10.8673 - val_mae: 11.3669\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.6240 - mae: 12.1115 - val_loss: 9.5693 - val_mae: 10.0600\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.4658 - mae: 11.9530 - val_loss: 10.8477 - val_mae: 11.3477\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.3044 - mae: 11.7894 - val_loss: 9.4821 - val_mae: 9.9732\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.4215 - mae: 11.9116 - val_loss: 10.6785 - val_mae: 11.1785\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.2461 - mae: 11.7292 - val_loss: 8.9351 - val_mae: 9.4297\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.8397 - mae: 11.3189 - val_loss: 10.7181 - val_mae: 11.2181\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.3244 - mae: 11.8110 - val_loss: 9.4025 - val_mae: 9.8921\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.3302 - mae: 11.8157 - val_loss: 11.6668 - val_mae: 12.1571\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 422ms/step - loss: 80.8979 - mae: 81.3951 - val_loss: 14.4543 - val_mae: 14.9459\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 26.2950 - mae: 26.7918 - val_loss: 12.5366 - val_mae: 13.0340\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.8982 - mae: 16.3941 - val_loss: 25.8663 - val_mae: 26.3663\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 23.3045 - mae: 23.7999 - val_loss: 15.8462 - val_mae: 16.3439\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 17.0594 - mae: 17.5552 - val_loss: 11.3694 - val_mae: 11.8582\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.9628 - mae: 17.4582 - val_loss: 12.1363 - val_mae: 12.6262\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 16.6100 - mae: 17.1058 - val_loss: 15.3086 - val_mae: 15.8063\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.7524 - mae: 15.2471 - val_loss: 11.1762 - val_mae: 11.6677\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.1056 - mae: 14.5997 - val_loss: 14.3325 - val_mae: 14.8205\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.8265 - mae: 15.3226 - val_loss: 11.5949 - val_mae: 12.0890\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.9125 - mae: 15.4040 - val_loss: 12.1397 - val_mae: 12.6340\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.8141 - mae: 16.3109 - val_loss: 13.8098 - val_mae: 14.3097\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1254 - mae: 14.6230 - val_loss: 10.6620 - val_mae: 11.1413\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 14.1341 - mae: 14.6294 - val_loss: 12.8698 - val_mae: 13.3641\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3532 - mae: 13.8474 - val_loss: 10.4939 - val_mae: 10.9860\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.4646 - mae: 13.9518 - val_loss: 11.5659 - val_mae: 12.0629\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.5969 - mae: 14.0854 - val_loss: 10.4866 - val_mae: 10.9622\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.4049 - mae: 13.8880 - val_loss: 11.2790 - val_mae: 11.7686\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.5855 - mae: 15.0775 - val_loss: 10.7699 - val_mae: 11.2668\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.8818 - mae: 14.3723 - val_loss: 10.5784 - val_mae: 11.0774\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.5755 - mae: 15.0715 - val_loss: 11.3630 - val_mae: 11.8572\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.3054 - mae: 13.7931 - val_loss: 10.1574 - val_mae: 10.6382\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.1961 - mae: 13.6867 - val_loss: 10.4928 - val_mae: 10.9912\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.8418 - mae: 13.3310 - val_loss: 10.2846 - val_mae: 10.7774\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.3223 - mae: 13.8106 - val_loss: 9.9905 - val_mae: 10.4566\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.0237 - mae: 13.5155 - val_loss: 10.2873 - val_mae: 10.7836\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.7349 - mae: 14.2270 - val_loss: 9.9342 - val_mae: 10.4077\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.1601 - mae: 13.6545 - val_loss: 9.9918 - val_mae: 10.4713\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.4680 - mae: 13.9597 - val_loss: 9.9130 - val_mae: 10.3829\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0003 - mae: 13.4916 - val_loss: 9.9125 - val_mae: 10.3845\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.1261 - mae: 13.6171 - val_loss: 9.8939 - val_mae: 10.3643\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.4330 - mae: 12.9244 - val_loss: 9.9127 - val_mae: 10.3913\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.0278 - mae: 13.5202 - val_loss: 9.9944 - val_mae: 10.4875\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5773 - mae: 13.0695 - val_loss: 9.9226 - val_mae: 10.4008\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.2105 - mae: 13.6976 - val_loss: 10.0300 - val_mae: 10.5186\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4376 - mae: 13.9305 - val_loss: 9.8780 - val_mae: 10.3532\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.6198 - mae: 15.1176 - val_loss: 10.0168 - val_mae: 10.5028\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.2264 - mae: 13.7182 - val_loss: 10.1319 - val_mae: 10.6239\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.2157 - mae: 13.7104 - val_loss: 9.8308 - val_mae: 10.3032\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.4037 - mae: 12.8974 - val_loss: 9.9290 - val_mae: 10.4180\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.4267 - mae: 12.9157 - val_loss: 9.8376 - val_mae: 10.3259\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8265 - mae: 12.3116 - val_loss: 9.8534 - val_mae: 10.3448\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.9241 - mae: 12.4190 - val_loss: 10.1175 - val_mae: 10.5967\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.7899 - mae: 12.2838 - val_loss: 9.8316 - val_mae: 10.3072\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.9018 - mae: 12.3900 - val_loss: 10.4909 - val_mae: 10.9778\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.8708 - mae: 12.3558 - val_loss: 10.0109 - val_mae: 10.4997\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.2244 - mae: 12.7169 - val_loss: 11.0195 - val_mae: 11.5127\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.1676 - mae: 12.6541 - val_loss: 10.1641 - val_mae: 10.6540\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.8573 - mae: 13.3514 - val_loss: 10.9899 - val_mae: 11.4807\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.8183 - mae: 13.3045 - val_loss: 9.7649 - val_mae: 10.2404\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.4017 - mae: 13.8891 - val_loss: 10.3293 - val_mae: 10.8081\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.6966 - mae: 13.1861 - val_loss: 9.9373 - val_mae: 10.4372\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.7555 - mae: 13.2487 - val_loss: 9.8061 - val_mae: 10.2909\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.5475 - mae: 12.0329 - val_loss: 10.0172 - val_mae: 10.5139\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.6971 - mae: 12.1872 - val_loss: 9.9251 - val_mae: 10.4247\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1910 - mae: 11.6790 - val_loss: 10.2670 - val_mae: 10.7455\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.2145 - mae: 11.7039 - val_loss: 10.0363 - val_mae: 10.5363\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.1099 - mae: 11.5938 - val_loss: 9.8416 - val_mae: 10.3166\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.3237 - mae: 11.8140 - val_loss: 10.2364 - val_mae: 10.7273\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.0876 - mae: 11.5750 - val_loss: 10.1188 - val_mae: 10.6179\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0083 - mae: 11.4933 - val_loss: 10.3518 - val_mae: 10.8319\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9843 - mae: 11.4721 - val_loss: 10.0428 - val_mae: 10.5426\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9099 - mae: 11.3993 - val_loss: 10.4629 - val_mae: 10.9467\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9493 - mae: 11.4387 - val_loss: 10.1406 - val_mae: 10.6405\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.8518 - mae: 11.3407 - val_loss: 10.3558 - val_mae: 10.8382\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.8227 - mae: 11.3116 - val_loss: 10.0849 - val_mae: 10.5729\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.7938 - mae: 11.2802 - val_loss: 10.6748 - val_mae: 11.1555\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.8432 - mae: 11.3259 - val_loss: 9.8895 - val_mae: 10.3671\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9409 - mae: 11.4269 - val_loss: 11.4549 - val_mae: 11.9422\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.2783 - mae: 11.7661 - val_loss: 9.8965 - val_mae: 10.3798\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 411ms/step - loss: 80.1567 - mae: 80.6560 - val_loss: 26.5513 - val_mae: 27.0513\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 24.7326 - mae: 25.2324 - val_loss: 12.0462 - val_mae: 12.5427\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 17.8057 - mae: 18.3006 - val_loss: 18.1981 - val_mae: 18.6897\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 17.0414 - mae: 17.5376 - val_loss: 13.1385 - val_mae: 13.6188\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.0864 - mae: 16.5804 - val_loss: 12.4170 - val_mae: 12.9029\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.7263 - mae: 15.2172 - val_loss: 12.5711 - val_mae: 13.0585\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.3894 - mae: 14.8853 - val_loss: 11.8592 - val_mae: 12.3526\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.7041 - mae: 15.1949 - val_loss: 11.9508 - val_mae: 12.4491\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.3922 - mae: 14.8839 - val_loss: 11.1236 - val_mae: 11.5966\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.2022 - mae: 13.6926 - val_loss: 11.5061 - val_mae: 11.9865\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 13.3563 - mae: 13.8471 - val_loss: 11.6097 - val_mae: 12.0972\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8304 - mae: 13.3210 - val_loss: 11.0677 - val_mae: 11.5622\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6085 - mae: 13.0989 - val_loss: 11.1371 - val_mae: 11.6309\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.3612 - mae: 12.8455 - val_loss: 10.8635 - val_mae: 11.3542\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.2138 - mae: 12.7021 - val_loss: 10.9676 - val_mae: 11.4676\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.1465 - mae: 12.6409 - val_loss: 10.8290 - val_mae: 11.3148\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8950 - mae: 12.3798 - val_loss: 10.9614 - val_mae: 11.4614\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8591 - mae: 12.3457 - val_loss: 10.9647 - val_mae: 11.4529\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8602 - mae: 12.3372 - val_loss: 10.7751 - val_mae: 11.2750\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.6774 - mae: 12.1677 - val_loss: 11.0723 - val_mae: 11.5662\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.6122 - mae: 12.0936 - val_loss: 10.8665 - val_mae: 11.3592\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.5620 - mae: 12.0350 - val_loss: 11.0254 - val_mae: 11.5218\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5829 - mae: 12.0621 - val_loss: 10.6415 - val_mae: 11.1326\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.2971 - mae: 11.7836 - val_loss: 11.2441 - val_mae: 11.7333\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3820 - mae: 11.8627 - val_loss: 10.3576 - val_mae: 10.8472\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.2425 - mae: 11.7166 - val_loss: 11.5807 - val_mae: 12.0749\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.6607 - mae: 12.1506 - val_loss: 10.4492 - val_mae: 10.9380\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.8351 - mae: 12.3024 - val_loss: 12.0326 - val_mae: 12.5326\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.7563 - mae: 12.2320 - val_loss: 10.6540 - val_mae: 11.1345\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8253 - mae: 12.3161 - val_loss: 12.5764 - val_mae: 13.0705\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.4465 - mae: 12.9321 - val_loss: 11.7767 - val_mae: 12.2765\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6066 - mae: 13.0898 - val_loss: 12.8638 - val_mae: 13.3613\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 12.7968 - mae: 13.2900 - val_loss: 11.0245 - val_mae: 11.5126\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.4718 - mae: 13.9605 - val_loss: 10.6481 - val_mae: 11.1344\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.6230 - mae: 12.1153 - val_loss: 10.2401 - val_mae: 10.7199\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.7733 - mae: 12.2602 - val_loss: 10.8550 - val_mae: 11.3550\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9788 - mae: 11.4669 - val_loss: 10.6436 - val_mae: 11.1377\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9078 - mae: 11.3983 - val_loss: 10.7507 - val_mae: 11.2496\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.6436 - mae: 11.1131 - val_loss: 10.4446 - val_mae: 10.9221\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.6099 - mae: 11.0733 - val_loss: 10.6295 - val_mae: 11.1207\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.5858 - mae: 11.0742 - val_loss: 10.1902 - val_mae: 10.6646\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 10.6748 - mae: 11.1438 - val_loss: 11.2252 - val_mae: 11.7217\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.7656 - mae: 11.2430 - val_loss: 10.1604 - val_mae: 10.6236\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.4486 - mae: 10.9091 - val_loss: 11.5671 - val_mae: 12.0615\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.5870 - mae: 11.0529 - val_loss: 10.2783 - val_mae: 10.7683\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.5183 - mae: 10.9926 - val_loss: 11.3798 - val_mae: 11.8770\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.5074 - mae: 10.9821 - val_loss: 10.1543 - val_mae: 10.6388\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.2557 - mae: 10.7310 - val_loss: 10.7349 - val_mae: 11.2318\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.4357 - mae: 10.9124 - val_loss: 10.4563 - val_mae: 10.9376\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.3287 - mae: 10.8066 - val_loss: 10.2508 - val_mae: 10.7231\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.6304 - mae: 11.1079 - val_loss: 12.4635 - val_mae: 12.9560\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0295 - mae: 11.4923 - val_loss: 11.8420 - val_mae: 12.3392\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.2979 - mae: 11.7682 - val_loss: 11.7098 - val_mae: 12.2044\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.5827 - mae: 11.0577 - val_loss: 10.1002 - val_mae: 10.5707\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.1388 - mae: 10.6162 - val_loss: 10.0762 - val_mae: 10.5525\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.2618 - mae: 10.7327 - val_loss: 12.0794 - val_mae: 12.5794\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.7127 - mae: 11.1755 - val_loss: 10.5020 - val_mae: 10.9933\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.2624 - mae: 10.7454 - val_loss: 10.9078 - val_mae: 11.4078\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.0667 - mae: 10.5357 - val_loss: 10.1457 - val_mae: 10.6248\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.0801 - mae: 10.5619 - val_loss: 11.3396 - val_mae: 11.8379\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.2904 - mae: 10.7735 - val_loss: 10.0989 - val_mae: 10.5776\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.9010 - mae: 10.3865 - val_loss: 10.6700 - val_mae: 11.1676\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.0041 - mae: 10.4664 - val_loss: 10.5217 - val_mae: 11.0034\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.9930 - mae: 10.4596 - val_loss: 10.0970 - val_mae: 10.5784\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.8248 - mae: 10.2849 - val_loss: 10.2526 - val_mae: 10.7305\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 9.7925 - mae: 10.2536 - val_loss: 10.2067 - val_mae: 10.6895\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 9.7503 - mae: 10.2148 - val_loss: 10.6829 - val_mae: 11.1788\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 9.7446 - mae: 10.2010 - val_loss: 10.0278 - val_mae: 10.4954\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.8468 - mae: 10.3295 - val_loss: 11.2144 - val_mae: 11.7130\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.2057 - mae: 10.6885 - val_loss: 10.3764 - val_mae: 10.8636\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.9469 - mae: 10.4210 - val_loss: 10.8275 - val_mae: 11.3206\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.4462 - mae: 10.9137 - val_loss: 12.3269 - val_mae: 12.8165\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.5064 - mae: 10.9893 - val_loss: 10.3402 - val_mae: 10.8352\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.0751 - mae: 10.5605 - val_loss: 10.5526 - val_mae: 11.0344\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 9.9670 - mae: 10.4438 - val_loss: 10.8661 - val_mae: 11.3639\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.0049 - mae: 10.4752 - val_loss: 10.5261 - val_mae: 11.0219\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.8800 - mae: 10.3468 - val_loss: 10.7259 - val_mae: 11.2174\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 9.7839 - mae: 10.2643 - val_loss: 10.6400 - val_mae: 11.1253\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.7450 - mae: 10.2106 - val_loss: 10.0950 - val_mae: 10.5644\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 9.6133 - mae: 10.0934 - val_loss: 10.2153 - val_mae: 10.7075\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.5645 - mae: 10.0343 - val_loss: 10.8376 - val_mae: 11.3258\n",
      "Epoch 82/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.8049 - mae: 10.2913 - val_loss: 10.1215 - val_mae: 10.5896\n",
      "Epoch 83/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 9.6003 - mae: 10.0816 - val_loss: 10.0943 - val_mae: 10.5760\n",
      "Epoch 84/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 9.6563 - mae: 10.1176 - val_loss: 11.5565 - val_mae: 12.0563\n",
      "Epoch 85/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.1355 - mae: 10.6213 - val_loss: 10.1949 - val_mae: 10.6666\n",
      "Epoch 86/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.6240 - mae: 10.1067 - val_loss: 10.0918 - val_mae: 10.5725\n",
      "Epoch 87/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.5462 - mae: 10.0066 - val_loss: 10.4612 - val_mae: 10.9422\n",
      "Epoch 88/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 9.4930 - mae: 9.9719 - val_loss: 10.5916 - val_mae: 11.0808\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 3s 378ms/step - loss: 81.7892 - mae: 82.2867 - val_loss: 17.8551 - val_mae: 18.3522\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 25.6226 - mae: 26.1201 - val_loss: 12.1565 - val_mae: 12.6302\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 17.3200 - mae: 17.8170 - val_loss: 21.3018 - val_mae: 21.8018\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 18.7424 - mae: 19.2385 - val_loss: 12.7966 - val_mae: 13.2846\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 17.9013 - mae: 18.3953 - val_loss: 12.0391 - val_mae: 12.5365\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 15.3109 - mae: 15.8038 - val_loss: 13.8368 - val_mae: 14.3314\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.5686 - mae: 16.0605 - val_loss: 12.4175 - val_mae: 12.8980\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 14.8390 - mae: 15.3347 - val_loss: 11.3907 - val_mae: 11.8858\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 25ms/step - loss: 13.5256 - mae: 14.0155 - val_loss: 11.7039 - val_mae: 12.1902\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.5116 - mae: 13.9974 - val_loss: 11.4881 - val_mae: 11.9696\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.4318 - mae: 13.9157 - val_loss: 11.3600 - val_mae: 11.8571\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.1208 - mae: 13.6097 - val_loss: 11.3208 - val_mae: 11.8093\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1081 - mae: 13.5950 - val_loss: 11.2937 - val_mae: 11.7895\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.0869 - mae: 13.5762 - val_loss: 11.2674 - val_mae: 11.7640\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8740 - mae: 13.3649 - val_loss: 11.1974 - val_mae: 11.6865\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.8595 - mae: 13.3481 - val_loss: 11.1842 - val_mae: 11.6842\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6945 - mae: 13.1843 - val_loss: 11.1019 - val_mae: 11.5941\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6471 - mae: 13.1330 - val_loss: 11.0699 - val_mae: 11.5631\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.5554 - mae: 13.0458 - val_loss: 11.0029 - val_mae: 11.4831\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.4595 - mae: 12.9486 - val_loss: 10.9993 - val_mae: 11.4937\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.4231 - mae: 12.9096 - val_loss: 10.9158 - val_mae: 11.4070\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.3095 - mae: 12.7984 - val_loss: 10.8936 - val_mae: 11.3852\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2426 - mae: 12.7230 - val_loss: 10.8969 - val_mae: 11.3860\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.1533 - mae: 12.6402 - val_loss: 10.8598 - val_mae: 11.3569\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.1610 - mae: 12.6409 - val_loss: 10.8537 - val_mae: 11.3524\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.0108 - mae: 12.4993 - val_loss: 10.8330 - val_mae: 11.3284\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.9851 - mae: 12.4688 - val_loss: 10.8214 - val_mae: 11.3206\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8903 - mae: 12.3752 - val_loss: 10.7925 - val_mae: 11.2787\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8444 - mae: 12.3271 - val_loss: 10.7461 - val_mae: 11.2360\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8079 - mae: 12.2899 - val_loss: 10.7261 - val_mae: 11.2067\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.7416 - mae: 12.2280 - val_loss: 10.7438 - val_mae: 11.2282\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.9046 - mae: 12.3944 - val_loss: 10.7382 - val_mae: 11.2227\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.7216 - mae: 12.2024 - val_loss: 10.7576 - val_mae: 11.2451\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.8129 - mae: 12.3037 - val_loss: 10.7282 - val_mae: 11.2104\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.6095 - mae: 12.0892 - val_loss: 10.7356 - val_mae: 11.2260\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.5109 - mae: 11.9977 - val_loss: 10.7119 - val_mae: 11.2031\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.5554 - mae: 12.0408 - val_loss: 10.7154 - val_mae: 11.2071\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.4346 - mae: 11.9238 - val_loss: 10.7645 - val_mae: 11.2624\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.4044 - mae: 11.8826 - val_loss: 10.7420 - val_mae: 11.2195\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.5616 - mae: 12.0526 - val_loss: 10.7643 - val_mae: 11.2557\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3367 - mae: 11.8290 - val_loss: 10.7168 - val_mae: 11.2145\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.4566 - mae: 11.9412 - val_loss: 10.6802 - val_mae: 11.1643\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.3055 - mae: 11.7949 - val_loss: 10.7091 - val_mae: 11.2073\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.2317 - mae: 11.7105 - val_loss: 10.6898 - val_mae: 11.1730\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.2955 - mae: 11.7776 - val_loss: 10.7106 - val_mae: 11.2079\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1311 - mae: 11.6104 - val_loss: 10.7061 - val_mae: 11.2021\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.1502 - mae: 11.6333 - val_loss: 10.6786 - val_mae: 11.1617\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.1197 - mae: 11.6075 - val_loss: 10.6867 - val_mae: 11.1797\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0367 - mae: 11.5234 - val_loss: 10.6683 - val_mae: 11.1513\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.0764 - mae: 11.5642 - val_loss: 10.7264 - val_mae: 11.2174\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.9907 - mae: 11.4747 - val_loss: 10.6494 - val_mae: 11.1337\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.2585 - mae: 11.7433 - val_loss: 10.7211 - val_mae: 11.2090\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.0443 - mae: 11.5314 - val_loss: 11.0059 - val_mae: 11.5012\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.2616 - mae: 11.7522 - val_loss: 11.3600 - val_mae: 11.8305\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.5598 - mae: 12.0465 - val_loss: 11.4429 - val_mae: 11.9421\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.3310 - mae: 11.8246 - val_loss: 11.8244 - val_mae: 12.3226\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.6135 - mae: 12.0986 - val_loss: 11.1591 - val_mae: 11.6452\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.1626 - mae: 11.6564 - val_loss: 10.6351 - val_mae: 11.1163\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.1654 - mae: 11.6494 - val_loss: 10.6257 - val_mae: 11.1103\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.8426 - mae: 11.3240 - val_loss: 10.7093 - val_mae: 11.2049\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9781 - mae: 11.4665 - val_loss: 10.6762 - val_mae: 11.1647\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.2496 - mae: 11.7340 - val_loss: 10.6916 - val_mae: 11.1872\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.9760 - mae: 11.4589 - val_loss: 10.5994 - val_mae: 11.0827\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.7523 - mae: 11.2344 - val_loss: 10.7047 - val_mae: 11.1970\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.7714 - mae: 11.2591 - val_loss: 10.6420 - val_mae: 11.1309\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.8767 - mae: 11.3619 - val_loss: 10.6233 - val_mae: 11.1063\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.7670 - mae: 11.2497 - val_loss: 10.6271 - val_mae: 11.1140\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.8433 - mae: 11.3289 - val_loss: 10.6029 - val_mae: 11.0968\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0972 - mae: 11.5833 - val_loss: 11.3417 - val_mae: 11.8223\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.4371 - mae: 11.9296 - val_loss: 10.7428 - val_mae: 11.2418\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.7739 - mae: 11.2584 - val_loss: 10.5497 - val_mae: 11.0377\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.0576 - mae: 11.5439 - val_loss: 10.6883 - val_mae: 11.1795\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.7845 - mae: 11.2715 - val_loss: 11.0308 - val_mae: 11.5231\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.1364 - mae: 11.6294 - val_loss: 10.8988 - val_mae: 11.3882\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.3329 - mae: 11.8230 - val_loss: 10.6473 - val_mae: 11.1115\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.7797 - mae: 11.2663 - val_loss: 11.2506 - val_mae: 11.7464\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.1108 - mae: 11.5970 - val_loss: 10.9636 - val_mae: 11.4567\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.7985 - mae: 11.2844 - val_loss: 10.6276 - val_mae: 11.1110\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 10.5427 - mae: 11.0257 - val_loss: 10.7566 - val_mae: 11.2515\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.7199 - mae: 11.2037 - val_loss: 10.8506 - val_mae: 11.3504\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.9588 - mae: 11.4471 - val_loss: 10.8557 - val_mae: 11.3541\n",
      "Epoch 82/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.6096 - mae: 11.0935 - val_loss: 10.8060 - val_mae: 11.2984\n",
      "Epoch 83/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.6076 - mae: 11.0971 - val_loss: 10.6613 - val_mae: 11.1539\n",
      "Epoch 84/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 10.6986 - mae: 11.1842 - val_loss: 11.2570 - val_mae: 11.7309\n",
      "Epoch 85/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.7508 - mae: 11.2411 - val_loss: 10.9973 - val_mae: 11.4908\n",
      "Epoch 86/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.8315 - mae: 11.3210 - val_loss: 10.9083 - val_mae: 11.4074\n",
      "Epoch 87/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 10.9036 - mae: 11.3909 - val_loss: 10.6861 - val_mae: 11.1666\n",
      "Epoch 88/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.4827 - mae: 10.9671 - val_loss: 11.0206 - val_mae: 11.5110\n",
      "Epoch 89/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.6970 - mae: 11.1885 - val_loss: 11.1823 - val_mae: 11.6720\n",
      "Epoch 90/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.6004 - mae: 11.0903 - val_loss: 10.6128 - val_mae: 11.0969\n",
      "Epoch 91/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.3358 - mae: 10.8227 - val_loss: 10.6120 - val_mae: 11.0939\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 390ms/step - loss: 86.0871 - mae: 86.5861 - val_loss: 17.0213 - val_mae: 17.5161\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 28.4658 - mae: 28.9610 - val_loss: 11.3486 - val_mae: 11.8465\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 23.3709 - mae: 23.8684 - val_loss: 10.4765 - val_mae: 10.9691\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 20.2963 - mae: 20.7861 - val_loss: 11.8491 - val_mae: 12.3347\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 18.6455 - mae: 19.1412 - val_loss: 11.8467 - val_mae: 12.3384\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 16.7582 - mae: 17.2492 - val_loss: 12.0694 - val_mae: 12.5602\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 15.9615 - mae: 16.4544 - val_loss: 13.1150 - val_mae: 13.6150\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.4421 - mae: 15.9384 - val_loss: 11.4796 - val_mae: 11.9709\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.2061 - mae: 15.6981 - val_loss: 12.1078 - val_mae: 12.6078\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 15.2552 - mae: 15.7523 - val_loss: 12.3312 - val_mae: 12.8239\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.4111 - mae: 14.8991 - val_loss: 11.2513 - val_mae: 11.7493\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.2140 - mae: 14.7018 - val_loss: 11.6032 - val_mae: 12.0897\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.9510 - mae: 14.4420 - val_loss: 9.8846 - val_mae: 10.3661\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.7037 - mae: 14.1932 - val_loss: 10.2411 - val_mae: 10.7319\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.5099 - mae: 13.9762 - val_loss: 9.6180 - val_mae: 10.0941\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.5338 - mae: 14.0162 - val_loss: 11.6792 - val_mae: 12.1716\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.5815 - mae: 14.0550 - val_loss: 9.6796 - val_mae: 10.1595\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.1445 - mae: 13.6080 - val_loss: 9.8517 - val_mae: 10.3362\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.9522 - mae: 13.4177 - val_loss: 9.3703 - val_mae: 9.8630\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9393 - mae: 13.4055 - val_loss: 10.4382 - val_mae: 10.9309\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.7345 - mae: 13.2194 - val_loss: 9.2058 - val_mae: 9.6929\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.9782 - mae: 14.4715 - val_loss: 9.2501 - val_mae: 9.7422\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.9832 - mae: 13.4723 - val_loss: 14.0006 - val_mae: 14.4988\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.3195 - mae: 14.7804 - val_loss: 9.4528 - val_mae: 9.9432\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 16.4586 - mae: 16.9581 - val_loss: 10.6376 - val_mae: 11.1294\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.7244 - mae: 13.2097 - val_loss: 11.9578 - val_mae: 12.4578\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.6389 - mae: 15.1256 - val_loss: 20.2725 - val_mae: 20.7723\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 17.4646 - mae: 17.9621 - val_loss: 9.8485 - val_mae: 10.3485\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.6456 - mae: 14.1341 - val_loss: 10.7010 - val_mae: 11.1896\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 15.9937 - mae: 16.4851 - val_loss: 11.1340 - val_mae: 11.6144\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.1181 - mae: 14.6070 - val_loss: 9.4610 - val_mae: 9.9583\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.3337 - mae: 14.8281 - val_loss: 12.0468 - val_mae: 12.5319\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.3887 - mae: 14.8848 - val_loss: 9.0510 - val_mae: 9.5391\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.2124 - mae: 14.7053 - val_loss: 9.2426 - val_mae: 9.7157\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.5304 - mae: 13.9906 - val_loss: 9.3486 - val_mae: 9.8321\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.6081 - mae: 14.1032 - val_loss: 9.2327 - val_mae: 9.7186\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.8241 - mae: 14.3195 - val_loss: 9.1247 - val_mae: 9.6092\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.2912 - mae: 13.7746 - val_loss: 10.4525 - val_mae: 10.9416\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.6067 - mae: 14.1004 - val_loss: 9.9568 - val_mae: 10.4499\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.3145 - mae: 13.8021 - val_loss: 11.5944 - val_mae: 12.0874\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 40ms/step - loss: 12.5709 - mae: 13.0405 - val_loss: 9.7928 - val_mae: 10.2909\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.1098 - mae: 13.5974 - val_loss: 9.3991 - val_mae: 9.8843\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 11.9855 - mae: 12.4718 - val_loss: 9.1261 - val_mae: 9.6187\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 61ms/step - loss: 11.9822 - mae: 12.4583 - val_loss: 10.0162 - val_mae: 10.5129\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 58ms/step - loss: 12.0153 - mae: 12.4884 - val_loss: 9.0402 - val_mae: 9.5385\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 95ms/step - loss: 12.1943 - mae: 12.6851 - val_loss: 9.0771 - val_mae: 9.5717\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 45ms/step - loss: 11.8892 - mae: 12.3642 - val_loss: 10.1348 - val_mae: 10.6348\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 47ms/step - loss: 11.7813 - mae: 12.2637 - val_loss: 8.9858 - val_mae: 9.4643\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 46ms/step - loss: 12.5144 - mae: 13.0056 - val_loss: 9.1052 - val_mae: 9.5969\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 11.8208 - mae: 12.3108 - val_loss: 11.9102 - val_mae: 12.4065\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 82ms/step - loss: 12.5158 - mae: 12.9791 - val_loss: 10.2135 - val_mae: 10.7097\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 48ms/step - loss: 14.2186 - mae: 14.7148 - val_loss: 9.3425 - val_mae: 9.8315\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.6786 - mae: 12.1511 - val_loss: 9.4450 - val_mae: 9.9371\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 41ms/step - loss: 12.3463 - mae: 12.8385 - val_loss: 12.6067 - val_mae: 13.0969\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 15.0400 - mae: 15.5359 - val_loss: 9.8077 - val_mae: 10.2904\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 95ms/step - loss: 13.2984 - mae: 13.7926 - val_loss: 9.0081 - val_mae: 9.4896\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 76ms/step - loss: 11.4794 - mae: 11.9566 - val_loss: 8.9669 - val_mae: 9.4417\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 57ms/step - loss: 11.8667 - mae: 12.3276 - val_loss: 9.1659 - val_mae: 9.6574\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 11.9903 - mae: 12.4813 - val_loss: 10.1447 - val_mae: 10.6356\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 70ms/step - loss: 11.6275 - mae: 12.0918 - val_loss: 8.7762 - val_mae: 9.2615\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 13.2951 - mae: 13.7916 - val_loss: 8.5685 - val_mae: 9.0479\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 11.9694 - mae: 12.4567 - val_loss: 8.9077 - val_mae: 9.3997\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 65ms/step - loss: 11.5112 - mae: 11.9996 - val_loss: 13.3077 - val_mae: 13.8042\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5934 - mae: 14.0887 - val_loss: 9.3002 - val_mae: 9.7967\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.4476 - mae: 12.9423 - val_loss: 8.8943 - val_mae: 9.3801\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.1951 - mae: 11.6470 - val_loss: 9.0312 - val_mae: 9.5202\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.2585 - mae: 11.7144 - val_loss: 9.0038 - val_mae: 9.4917\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.0026 - mae: 11.4593 - val_loss: 10.0984 - val_mae: 10.5884\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.2003 - mae: 11.6643 - val_loss: 10.3090 - val_mae: 10.8036\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.8524 - mae: 13.3466 - val_loss: 9.1416 - val_mae: 9.6299\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0543 - mae: 11.5480 - val_loss: 8.8407 - val_mae: 9.3137\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.2754 - mae: 11.7401 - val_loss: 9.1161 - val_mae: 9.6025\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.9703 - mae: 11.4436 - val_loss: 11.6394 - val_mae: 12.1358\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.2604 - mae: 12.7423 - val_loss: 9.5050 - val_mae: 9.9883\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.2356 - mae: 13.7338 - val_loss: 8.8504 - val_mae: 9.3441\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9208 - mae: 11.3828 - val_loss: 9.0307 - val_mae: 9.5198\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0013 - mae: 11.4890 - val_loss: 8.9190 - val_mae: 9.4064\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.8687 - mae: 11.3588 - val_loss: 8.9763 - val_mae: 9.4593\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.8056 - mae: 11.2664 - val_loss: 9.2807 - val_mae: 9.7644\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0471 - mae: 11.5386 - val_loss: 9.2296 - val_mae: 9.7224\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.9714 - mae: 11.4447 - val_loss: 11.5938 - val_mae: 12.0936\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 530ms/step - loss: 83.8550 - mae: 84.3542 - val_loss: 16.1655 - val_mae: 16.6639\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 27.1577 - mae: 27.6574 - val_loss: 11.5437 - val_mae: 12.0247\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 16.9712 - mae: 17.4606 - val_loss: 22.9395 - val_mae: 23.4395\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 22.2924 - mae: 22.7832 - val_loss: 12.8113 - val_mae: 13.2998\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 15.9290 - mae: 16.4257 - val_loss: 12.5363 - val_mae: 13.0235\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 17.2503 - mae: 17.7480 - val_loss: 12.7357 - val_mae: 13.2357\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 18.6590 - mae: 19.1537 - val_loss: 15.4425 - val_mae: 15.9373\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 15.0687 - mae: 15.5660 - val_loss: 10.9887 - val_mae: 11.4786\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.8741 - mae: 15.3649 - val_loss: 12.6737 - val_mae: 13.1684\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 15.0708 - mae: 15.5620 - val_loss: 11.1075 - val_mae: 11.5932\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.5505 - mae: 16.0458 - val_loss: 11.5091 - val_mae: 12.0040\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.8272 - mae: 16.3229 - val_loss: 12.5462 - val_mae: 13.0379\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.5996 - mae: 15.0933 - val_loss: 10.9023 - val_mae: 11.3954\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.3791 - mae: 14.8720 - val_loss: 11.9609 - val_mae: 12.4512\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.9364 - mae: 14.4286 - val_loss: 10.9936 - val_mae: 11.4924\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9277 - mae: 14.4214 - val_loss: 11.4208 - val_mae: 11.9072\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.8397 - mae: 14.3325 - val_loss: 11.0392 - val_mae: 11.5316\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7572 - mae: 14.2497 - val_loss: 10.7281 - val_mae: 11.2278\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7012 - mae: 14.1912 - val_loss: 10.5381 - val_mae: 11.0276\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.6959 - mae: 14.1874 - val_loss: 10.6386 - val_mae: 11.1382\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.6357 - mae: 14.1297 - val_loss: 10.2414 - val_mae: 10.7285\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.5415 - mae: 14.0347 - val_loss: 10.1928 - val_mae: 10.6716\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5665 - mae: 14.0592 - val_loss: 10.2550 - val_mae: 10.7444\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 13.4088 - mae: 13.9047 - val_loss: 10.0052 - val_mae: 10.4987\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.4651 - mae: 13.9555 - val_loss: 10.0880 - val_mae: 10.5693\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.4439 - mae: 13.9313 - val_loss: 10.0817 - val_mae: 10.5574\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.2222 - mae: 13.7167 - val_loss: 9.9160 - val_mae: 10.4051\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.2566 - mae: 13.7500 - val_loss: 10.1975 - val_mae: 10.6882\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1836 - mae: 13.6743 - val_loss: 9.9415 - val_mae: 10.4269\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.2285 - mae: 13.7178 - val_loss: 10.1117 - val_mae: 10.6038\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.0568 - mae: 13.5445 - val_loss: 9.9149 - val_mae: 10.4074\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.9952 - mae: 13.4848 - val_loss: 9.9374 - val_mae: 10.4170\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.0072 - mae: 13.5022 - val_loss: 9.8890 - val_mae: 10.3745\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.8182 - mae: 13.3089 - val_loss: 9.8508 - val_mae: 10.3376\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.8483 - mae: 13.3351 - val_loss: 9.8815 - val_mae: 10.3686\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.7295 - mae: 13.2185 - val_loss: 9.9913 - val_mae: 10.4744\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.0078 - mae: 13.4994 - val_loss: 9.9777 - val_mae: 10.4761\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 12.5752 - mae: 13.0651 - val_loss: 10.6041 - val_mae: 11.0888\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0722 - mae: 13.5671 - val_loss: 10.3145 - val_mae: 10.8044\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.6719 - mae: 13.1616 - val_loss: 10.9725 - val_mae: 11.4650\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.0883 - mae: 13.5758 - val_loss: 10.4617 - val_mae: 10.9534\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6560 - mae: 13.1439 - val_loss: 10.6306 - val_mae: 11.1116\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.8460 - mae: 13.3351 - val_loss: 10.6688 - val_mae: 11.1645\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.7090 - mae: 13.2017 - val_loss: 10.6506 - val_mae: 11.1345\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6269 - mae: 13.1128 - val_loss: 10.3054 - val_mae: 10.8009\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.3268 - mae: 12.8130 - val_loss: 10.3651 - val_mae: 10.8602\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 45ms/step - loss: 12.4855 - mae: 12.9757 - val_loss: 10.6346 - val_mae: 11.1340\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.4553 - mae: 12.9398 - val_loss: 10.0406 - val_mae: 10.5296\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0839 - mae: 12.5694 - val_loss: 10.2539 - val_mae: 10.7455\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.1782 - mae: 12.6669 - val_loss: 10.2780 - val_mae: 10.7567\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 12.2820 - mae: 12.7688 - val_loss: 10.5909 - val_mae: 11.0905\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 75ms/step - loss: 12.3111 - mae: 12.7973 - val_loss: 10.0959 - val_mae: 10.5866\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 60ms/step - loss: 12.1292 - mae: 12.6149 - val_loss: 9.9212 - val_mae: 10.4071\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 53ms/step - loss: 12.4818 - mae: 12.9691 - val_loss: 10.6751 - val_mae: 11.1751\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 386ms/step - loss: 84.2571 - mae: 84.7530 - val_loss: 26.3621 - val_mae: 26.8620\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 27.4095 - mae: 27.9079 - val_loss: 13.9475 - val_mae: 14.4416\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 20.9941 - mae: 21.4929 - val_loss: 22.0042 - val_mae: 22.4802\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 19.8885 - mae: 20.3843 - val_loss: 15.4326 - val_mae: 15.9326\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 16.8090 - mae: 17.3033 - val_loss: 14.6858 - val_mae: 15.1786\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 16.1878 - mae: 16.6842 - val_loss: 14.2364 - val_mae: 14.7289\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 16.3824 - mae: 16.8773 - val_loss: 14.0933 - val_mae: 14.5850\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 16.2822 - mae: 16.7723 - val_loss: 13.7000 - val_mae: 14.1840\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 15.4725 - mae: 15.9649 - val_loss: 14.1593 - val_mae: 14.6479\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.9547 - mae: 15.4478 - val_loss: 13.3662 - val_mae: 13.8457\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.8730 - mae: 15.3673 - val_loss: 13.8874 - val_mae: 14.3801\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.0212 - mae: 15.5113 - val_loss: 13.2621 - val_mae: 13.7525\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.7112 - mae: 15.2091 - val_loss: 13.5227 - val_mae: 14.0102\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.9314 - mae: 15.4218 - val_loss: 13.2586 - val_mae: 13.7562\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.6860 - mae: 15.1805 - val_loss: 13.2146 - val_mae: 13.7138\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.8405 - mae: 15.3326 - val_loss: 13.2349 - val_mae: 13.7279\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 14.3466 - mae: 14.8387 - val_loss: 13.2495 - val_mae: 13.7479\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.5304 - mae: 15.0277 - val_loss: 13.3539 - val_mae: 13.8340\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.1869 - mae: 14.6775 - val_loss: 13.3363 - val_mae: 13.8281\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.2673 - mae: 14.7591 - val_loss: 13.2630 - val_mae: 13.7444\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.8807 - mae: 14.3742 - val_loss: 13.2211 - val_mae: 13.7128\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.0317 - mae: 14.5237 - val_loss: 13.1945 - val_mae: 13.6886\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.6529 - mae: 14.1401 - val_loss: 13.1547 - val_mae: 13.6440\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.7883 - mae: 14.2764 - val_loss: 13.1184 - val_mae: 13.6099\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.6088 - mae: 14.0984 - val_loss: 13.1476 - val_mae: 13.6393\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.0042 - mae: 14.4921 - val_loss: 13.0695 - val_mae: 13.5647\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.5434 - mae: 14.0316 - val_loss: 13.0151 - val_mae: 13.4969\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.8062 - mae: 14.2934 - val_loss: 12.9978 - val_mae: 13.4811\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4310 - mae: 13.9199 - val_loss: 13.0019 - val_mae: 13.4922\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.8251 - mae: 14.3118 - val_loss: 13.0851 - val_mae: 13.5667\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9235 - mae: 14.4140 - val_loss: 13.2009 - val_mae: 13.6763\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.8217 - mae: 15.3197 - val_loss: 12.9751 - val_mae: 13.4623\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.8074 - mae: 15.2975 - val_loss: 13.0704 - val_mae: 13.5703\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.7584 - mae: 15.2509 - val_loss: 13.0798 - val_mae: 13.5798\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.8045 - mae: 14.3003 - val_loss: 12.9794 - val_mae: 13.4586\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.9054 - mae: 14.3961 - val_loss: 12.8566 - val_mae: 13.3407\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.7040 - mae: 14.1945 - val_loss: 13.0839 - val_mae: 13.5698\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.0038 - mae: 14.4978 - val_loss: 13.0799 - val_mae: 13.5623\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.4322 - mae: 13.9230 - val_loss: 13.2179 - val_mae: 13.7154\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.5106 - mae: 15.0094 - val_loss: 12.9806 - val_mae: 13.4722\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.8663 - mae: 14.3606 - val_loss: 12.7672 - val_mae: 13.2610\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.1059 - mae: 15.6007 - val_loss: 12.8577 - val_mae: 13.3466\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 43ms/step - loss: 13.4767 - mae: 13.9677 - val_loss: 13.1886 - val_mae: 13.6671\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.2805 - mae: 13.7680 - val_loss: 12.9874 - val_mae: 13.4814\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.8134 - mae: 13.3011 - val_loss: 13.0186 - val_mae: 13.5017\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.9476 - mae: 13.4413 - val_loss: 12.8188 - val_mae: 13.3040\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6523 - mae: 13.1354 - val_loss: 13.0163 - val_mae: 13.5017\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.8248 - mae: 13.3168 - val_loss: 13.0863 - val_mae: 13.5777\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5129 - mae: 12.9939 - val_loss: 13.1857 - val_mae: 13.6736\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 61ms/step - loss: 12.4968 - mae: 12.9836 - val_loss: 13.1231 - val_mae: 13.6162\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 12.4327 - mae: 12.9230 - val_loss: 13.3433 - val_mae: 13.8396\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.3997 - mae: 12.8882 - val_loss: 13.1190 - val_mae: 13.6067\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 87ms/step - loss: 12.2936 - mae: 12.7808 - val_loss: 12.9816 - val_mae: 13.4582\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 79ms/step - loss: 12.4961 - mae: 12.9797 - val_loss: 13.4782 - val_mae: 13.9780\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 90ms/step - loss: 12.4030 - mae: 12.8904 - val_loss: 13.2061 - val_mae: 13.7033\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 87ms/step - loss: 12.6963 - mae: 13.1833 - val_loss: 14.0945 - val_mae: 14.5868\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 86ms/step - loss: 12.9411 - mae: 13.4316 - val_loss: 14.0159 - val_mae: 14.4845\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.3163 - mae: 13.8075 - val_loss: 14.0952 - val_mae: 14.5897\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 45ms/step - loss: 13.7355 - mae: 14.2291 - val_loss: 14.9745 - val_mae: 15.4654\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 79ms/step - loss: 16.9548 - mae: 17.4463 - val_loss: 12.8350 - val_mae: 13.3268\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 64ms/step - loss: 14.0817 - mae: 14.5742 - val_loss: 13.0509 - val_mae: 13.5367\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 8s 873ms/step - loss: 87.0036 - mae: 87.5019 - val_loss: 13.1761 - val_mae: 13.6704\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 30.6670 - mae: 31.1655 - val_loss: 17.2394 - val_mae: 17.7380\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 22.3966 - mae: 22.8928 - val_loss: 12.2970 - val_mae: 12.7970\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 20.2213 - mae: 20.7180 - val_loss: 10.5913 - val_mae: 11.0796\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 19.1976 - mae: 19.6902 - val_loss: 12.0416 - val_mae: 12.5410\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 16.2283 - mae: 16.7155 - val_loss: 11.4880 - val_mae: 11.9679\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 17.0733 - mae: 17.5683 - val_loss: 11.9189 - val_mae: 12.4063\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 16.4146 - mae: 16.9124 - val_loss: 11.2106 - val_mae: 11.7059\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 16.6944 - mae: 17.1877 - val_loss: 12.3223 - val_mae: 12.8173\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.4244 - mae: 15.9158 - val_loss: 10.1809 - val_mae: 10.6661\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.6870 - mae: 16.1782 - val_loss: 11.3794 - val_mae: 11.8791\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 15.0739 - mae: 15.5652 - val_loss: 9.9701 - val_mae: 10.4495\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 15.3610 - mae: 15.8517 - val_loss: 11.1366 - val_mae: 11.6348\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.8479 - mae: 15.3429 - val_loss: 9.8705 - val_mae: 10.3612\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.8836 - mae: 15.3761 - val_loss: 10.1577 - val_mae: 10.6346\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.6364 - mae: 15.1296 - val_loss: 9.7827 - val_mae: 10.2736\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.5306 - mae: 15.0230 - val_loss: 10.0075 - val_mae: 10.4928\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.4347 - mae: 14.9264 - val_loss: 9.6476 - val_mae: 10.1198\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.3115 - mae: 14.8022 - val_loss: 9.6182 - val_mae: 10.0993\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.2551 - mae: 14.7490 - val_loss: 9.7318 - val_mae: 10.2131\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1187 - mae: 14.6096 - val_loss: 9.4933 - val_mae: 9.9693\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.1320 - mae: 14.6178 - val_loss: 9.5282 - val_mae: 10.0080\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9832 - mae: 14.4714 - val_loss: 9.6509 - val_mae: 10.1375\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.8838 - mae: 14.3737 - val_loss: 9.4491 - val_mae: 9.9360\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.0008 - mae: 14.4863 - val_loss: 9.4353 - val_mae: 9.9142\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9201 - mae: 14.4039 - val_loss: 9.4743 - val_mae: 9.9560\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.8716 - mae: 14.3644 - val_loss: 9.6031 - val_mae: 10.0923\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.7324 - mae: 14.2205 - val_loss: 9.4571 - val_mae: 9.9317\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.6539 - mae: 14.1360 - val_loss: 9.7448 - val_mae: 10.2376\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.6588 - mae: 14.1455 - val_loss: 9.6696 - val_mae: 10.1621\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.4970 - mae: 13.9808 - val_loss: 9.6850 - val_mae: 10.1816\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.4778 - mae: 13.9710 - val_loss: 9.5450 - val_mae: 10.0281\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.4514 - mae: 13.9376 - val_loss: 9.6204 - val_mae: 10.1074\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.5707 - mae: 14.0605 - val_loss: 9.3917 - val_mae: 9.8696\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3083 - mae: 13.7979 - val_loss: 9.4251 - val_mae: 9.9036\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.1611 - mae: 13.6517 - val_loss: 9.5636 - val_mae: 10.0458\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.3755 - mae: 13.8667 - val_loss: 9.6321 - val_mae: 10.1297\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 46ms/step - loss: 13.1366 - mae: 13.6225 - val_loss: 9.4437 - val_mae: 9.9309\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.1886 - mae: 13.6762 - val_loss: 10.1867 - val_mae: 10.6818\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 13.4347 - mae: 13.9255 - val_loss: 10.0464 - val_mae: 10.5452\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3687 - mae: 13.8574 - val_loss: 10.1117 - val_mae: 10.5967\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3858 - mae: 13.8764 - val_loss: 9.9589 - val_mae: 10.4509\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.0290 - mae: 13.5175 - val_loss: 9.7685 - val_mae: 10.2610\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.1436 - mae: 13.6281 - val_loss: 10.1095 - val_mae: 10.5988\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.9730 - mae: 13.4606 - val_loss: 9.4250 - val_mae: 9.9034\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.8591 - mae: 13.3458 - val_loss: 9.4508 - val_mae: 9.9362\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 12.8287 - mae: 13.3176 - val_loss: 10.4500 - val_mae: 10.9333\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.1785 - mae: 13.6653 - val_loss: 9.4213 - val_mae: 9.8997\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.5627 - mae: 13.0498 - val_loss: 9.7117 - val_mae: 10.2031\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5884 - mae: 13.0726 - val_loss: 9.5253 - val_mae: 10.0111\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6197 - mae: 13.1139 - val_loss: 10.3480 - val_mae: 10.8372\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1092 - mae: 13.5963 - val_loss: 9.6587 - val_mae: 10.1513\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.4351 - mae: 12.9219 - val_loss: 9.6670 - val_mae: 10.1636\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5855 - mae: 13.0737 - val_loss: 11.1180 - val_mae: 11.6099\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 433ms/step - loss: 84.8386 - mae: 85.3374 - val_loss: 17.5100 - val_mae: 18.0046\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 27.6315 - mae: 28.1238 - val_loss: 16.1241 - val_mae: 16.6177\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 20.9580 - mae: 21.4517 - val_loss: 12.7103 - val_mae: 13.2033\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 17.1301 - mae: 17.6248 - val_loss: 15.2608 - val_mae: 15.7447\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 18.8832 - mae: 19.3790 - val_loss: 11.2499 - val_mae: 11.7421\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.9396 - mae: 15.4314 - val_loss: 13.6890 - val_mae: 14.1870\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.9646 - mae: 16.4518 - val_loss: 11.4392 - val_mae: 11.9284\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.6932 - mae: 15.1836 - val_loss: 11.7021 - val_mae: 12.1857\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.7952 - mae: 15.2875 - val_loss: 11.5060 - val_mae: 12.0022\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 15.3580 - mae: 15.8539 - val_loss: 11.3819 - val_mae: 11.8729\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.2166 - mae: 15.7080 - val_loss: 11.3229 - val_mae: 11.8135\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 16.1396 - mae: 16.6355 - val_loss: 11.8304 - val_mae: 12.3158\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 14.3379 - mae: 14.8341 - val_loss: 12.2547 - val_mae: 12.7536\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.5588 - mae: 15.0491 - val_loss: 11.8248 - val_mae: 12.3152\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.7828 - mae: 14.2770 - val_loss: 11.6610 - val_mae: 12.1440\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.2115 - mae: 14.7037 - val_loss: 11.7111 - val_mae: 12.2023\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 14.2364 - mae: 14.7323 - val_loss: 11.1788 - val_mae: 11.6643\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.0272 - mae: 14.5171 - val_loss: 11.2523 - val_mae: 11.7507\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.2535 - mae: 14.7469 - val_loss: 11.3474 - val_mae: 11.8386\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.1537 - mae: 14.6473 - val_loss: 11.0954 - val_mae: 11.5754\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.3983 - mae: 14.8914 - val_loss: 11.5301 - val_mae: 12.0260\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.8631 - mae: 14.3573 - val_loss: 11.0764 - val_mae: 11.5514\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.0284 - mae: 14.5199 - val_loss: 11.5346 - val_mae: 12.0281\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.6481 - mae: 14.1381 - val_loss: 11.0597 - val_mae: 11.5364\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 13.7182 - mae: 14.2097 - val_loss: 11.3061 - val_mae: 11.7960\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.3059 - mae: 13.7933 - val_loss: 11.0435 - val_mae: 11.5193\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.4243 - mae: 13.9182 - val_loss: 11.0022 - val_mae: 11.4969\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.1677 - mae: 13.6550 - val_loss: 10.8750 - val_mae: 11.3606\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.2050 - mae: 13.6974 - val_loss: 10.7787 - val_mae: 11.2602\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1901 - mae: 13.6832 - val_loss: 10.7460 - val_mae: 11.2370\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1580 - mae: 13.6483 - val_loss: 10.7138 - val_mae: 11.2053\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.4669 - mae: 13.9517 - val_loss: 10.8466 - val_mae: 11.3366\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.1742 - mae: 13.6646 - val_loss: 10.5680 - val_mae: 11.0457\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9462 - mae: 14.4352 - val_loss: 11.1670 - val_mae: 11.6576\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.4885 - mae: 13.9799 - val_loss: 10.6577 - val_mae: 11.1554\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.0282 - mae: 14.5164 - val_loss: 11.1058 - val_mae: 11.5951\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3755 - mae: 13.8672 - val_loss: 10.6121 - val_mae: 11.1047\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7782 - mae: 14.2734 - val_loss: 10.6091 - val_mae: 11.1056\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3083 - mae: 13.8011 - val_loss: 10.4718 - val_mae: 10.9487\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.6022 - mae: 15.0929 - val_loss: 10.6718 - val_mae: 11.1666\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.9848 - mae: 14.4770 - val_loss: 10.6634 - val_mae: 11.1546\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 15.1319 - mae: 15.6235 - val_loss: 10.5923 - val_mae: 11.0803\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3344 - mae: 13.8208 - val_loss: 10.6992 - val_mae: 11.1989\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.7918 - mae: 14.2801 - val_loss: 10.5084 - val_mae: 11.0000\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.0114 - mae: 13.4995 - val_loss: 10.6013 - val_mae: 11.1003\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5871 - mae: 14.0811 - val_loss: 10.5977 - val_mae: 11.0977\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0203 - mae: 13.5096 - val_loss: 10.5087 - val_mae: 11.0053\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 13.4975 - mae: 13.9852 - val_loss: 10.7186 - val_mae: 11.2186\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.7087 - mae: 13.2005 - val_loss: 10.8118 - val_mae: 11.3037\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.8295 - mae: 13.3202 - val_loss: 10.3870 - val_mae: 10.8713\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5919 - mae: 13.0819 - val_loss: 10.3657 - val_mae: 10.8423\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.3575 - mae: 13.8465 - val_loss: 10.6628 - val_mae: 11.1541\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9776 - mae: 13.4674 - val_loss: 10.9102 - val_mae: 11.4033\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 14.3849 - mae: 14.8731 - val_loss: 11.4426 - val_mae: 11.9405\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.3708 - mae: 14.8643 - val_loss: 10.8217 - val_mae: 11.3111\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.2392 - mae: 15.7337 - val_loss: 10.6797 - val_mae: 11.1742\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.1825 - mae: 14.6772 - val_loss: 10.9899 - val_mae: 11.4887\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.9301 - mae: 14.4213 - val_loss: 11.4921 - val_mae: 11.9715\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1685 - mae: 13.6605 - val_loss: 12.1686 - val_mae: 12.6538\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.9245 - mae: 14.4204 - val_loss: 11.2484 - val_mae: 11.7325\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5845 - mae: 13.0785 - val_loss: 11.2229 - val_mae: 11.7207\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.5840 - mae: 13.0797 - val_loss: 10.8329 - val_mae: 11.3096\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.3207 - mae: 12.8095 - val_loss: 10.8160 - val_mae: 11.3152\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2008 - mae: 12.6917 - val_loss: 10.4663 - val_mae: 10.9538\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.0147 - mae: 12.5009 - val_loss: 11.2380 - val_mae: 11.7310\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.5058 - mae: 12.9951 - val_loss: 10.6233 - val_mae: 11.1098\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.8821 - mae: 12.3705 - val_loss: 11.1141 - val_mae: 11.5981\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.0514 - mae: 12.5429 - val_loss: 11.1992 - val_mae: 11.6817\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1383 - mae: 12.6309 - val_loss: 10.9301 - val_mae: 11.4149\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.9247 - mae: 12.4111 - val_loss: 10.8036 - val_mae: 11.2821\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0600 - mae: 12.5479 - val_loss: 10.5661 - val_mae: 11.0573\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 388ms/step - loss: 82.4602 - mae: 82.9584 - val_loss: 13.4843 - val_mae: 13.9840\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 27.9645 - mae: 28.4595 - val_loss: 14.1168 - val_mae: 14.6043\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 19.3653 - mae: 19.8601 - val_loss: 15.8926 - val_mae: 16.3824\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 19.2380 - mae: 19.7291 - val_loss: 9.9353 - val_mae: 10.4286\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 16.5072 - mae: 16.9982 - val_loss: 9.5675 - val_mae: 10.0541\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 15.2474 - mae: 15.7442 - val_loss: 10.6296 - val_mae: 11.1227\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 16.0759 - mae: 16.5662 - val_loss: 12.1322 - val_mae: 12.6249\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.7001 - mae: 15.1894 - val_loss: 9.3439 - val_mae: 9.8316\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.4148 - mae: 14.9054 - val_loss: 11.0409 - val_mae: 11.5320\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.2973 - mae: 13.7876 - val_loss: 9.0462 - val_mae: 9.5223\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5887 - mae: 14.0747 - val_loss: 10.0082 - val_mae: 10.4913\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.0202 - mae: 13.5063 - val_loss: 9.2752 - val_mae: 9.7692\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.2738 - mae: 13.7553 - val_loss: 9.0454 - val_mae: 9.5286\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.9288 - mae: 13.4193 - val_loss: 9.5268 - val_mae: 10.0115\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.0368 - mae: 13.5273 - val_loss: 8.6494 - val_mae: 9.1225\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.0558 - mae: 13.5434 - val_loss: 10.0889 - val_mae: 10.5785\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4496 - mae: 13.9438 - val_loss: 8.5422 - val_mae: 9.0371\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4949 - mae: 13.9900 - val_loss: 9.9108 - val_mae: 10.4104\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 14.4766 - mae: 14.9738 - val_loss: 8.6410 - val_mae: 9.1231\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.0560 - mae: 14.5498 - val_loss: 8.4535 - val_mae: 8.9374\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.9932 - mae: 14.4833 - val_loss: 9.6862 - val_mae: 10.1835\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6131 - mae: 13.1036 - val_loss: 8.6426 - val_mae: 9.1282\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.7457 - mae: 13.2336 - val_loss: 8.8056 - val_mae: 9.2934\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.4135 - mae: 12.9090 - val_loss: 8.2950 - val_mae: 8.7779\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.3668 - mae: 12.8523 - val_loss: 8.2589 - val_mae: 8.7413\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.3357 - mae: 12.8288 - val_loss: 8.2507 - val_mae: 8.7261\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2778 - mae: 12.7660 - val_loss: 8.2609 - val_mae: 8.7441\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1054 - mae: 12.5917 - val_loss: 8.2674 - val_mae: 8.7437\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.0900 - mae: 12.5822 - val_loss: 8.2172 - val_mae: 8.6986\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.9571 - mae: 12.4454 - val_loss: 8.2312 - val_mae: 8.7029\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.9329 - mae: 12.4254 - val_loss: 8.1730 - val_mae: 8.6578\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8692 - mae: 12.3593 - val_loss: 8.1860 - val_mae: 8.6606\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 11.8754 - mae: 12.3651 - val_loss: 8.1760 - val_mae: 8.6562\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.8193 - mae: 12.3048 - val_loss: 8.1703 - val_mae: 8.6575\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.7699 - mae: 12.2568 - val_loss: 8.2153 - val_mae: 8.6961\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.7607 - mae: 12.2482 - val_loss: 8.3278 - val_mae: 8.8035\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.9166 - mae: 12.4122 - val_loss: 8.3358 - val_mae: 8.8205\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7243 - mae: 12.2167 - val_loss: 9.1924 - val_mae: 9.6818\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.1968 - mae: 12.6845 - val_loss: 9.3469 - val_mae: 9.8392\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.5673 - mae: 13.0577 - val_loss: 10.2502 - val_mae: 10.7446\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.8603 - mae: 13.3454 - val_loss: 9.0959 - val_mae: 9.5917\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.9076 - mae: 13.3977 - val_loss: 9.2910 - val_mae: 9.7724\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.8657 - mae: 14.3587 - val_loss: 8.2609 - val_mae: 8.7346\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.3580 - mae: 13.8464 - val_loss: 8.3515 - val_mae: 8.8393\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.0197 - mae: 13.5126 - val_loss: 8.3005 - val_mae: 8.7870\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.9055 - mae: 13.4013 - val_loss: 8.4598 - val_mae: 8.9428\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.5308 - mae: 14.0244 - val_loss: 8.1985 - val_mae: 8.6688\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.6094 - mae: 13.0976 - val_loss: 8.1970 - val_mae: 8.6795\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1935 - mae: 13.6879 - val_loss: 8.5363 - val_mae: 9.0278\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.9720 - mae: 13.4593 - val_loss: 8.1569 - val_mae: 8.6451\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4622 - mae: 13.9594 - val_loss: 8.2090 - val_mae: 8.6757\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.3068 - mae: 12.7959 - val_loss: 8.1292 - val_mae: 8.6166\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2116 - mae: 12.6987 - val_loss: 8.1704 - val_mae: 8.6543\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.6573 - mae: 12.1485 - val_loss: 8.1137 - val_mae: 8.5912\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8255 - mae: 12.3178 - val_loss: 8.1356 - val_mae: 8.6237\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.2536 - mae: 11.7369 - val_loss: 8.1243 - val_mae: 8.6040\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.2940 - mae: 11.7842 - val_loss: 8.2064 - val_mae: 8.6981\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 11.1353 - mae: 11.6225 - val_loss: 8.1177 - val_mae: 8.5970\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.2741 - mae: 11.7561 - val_loss: 8.3441 - val_mae: 8.8310\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.1060 - mae: 11.5929 - val_loss: 8.2418 - val_mae: 8.7325\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.0357 - mae: 11.5156 - val_loss: 8.3733 - val_mae: 8.8644\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.9992 - mae: 11.4832 - val_loss: 8.2400 - val_mae: 8.7252\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9066 - mae: 11.3929 - val_loss: 8.2221 - val_mae: 8.7086\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9637 - mae: 11.4529 - val_loss: 8.4031 - val_mae: 8.8882\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.0493 - mae: 11.5430 - val_loss: 8.4700 - val_mae: 8.9550\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9402 - mae: 11.4293 - val_loss: 8.2060 - val_mae: 8.6941\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 42ms/step - loss: 10.9256 - mae: 11.4118 - val_loss: 8.4017 - val_mae: 8.8851\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9968 - mae: 11.4848 - val_loss: 8.2200 - val_mae: 8.6864\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.9919 - mae: 11.4771 - val_loss: 8.7099 - val_mae: 9.1956\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0008 - mae: 11.4894 - val_loss: 8.4480 - val_mae: 8.9279\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.7520 - mae: 11.2409 - val_loss: 8.4642 - val_mae: 8.9612\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.0341 - mae: 11.5263 - val_loss: 8.6325 - val_mae: 9.1116\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.7394 - mae: 11.2248 - val_loss: 8.3581 - val_mae: 8.8493\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9424 - mae: 11.4321 - val_loss: 8.4032 - val_mae: 8.8762\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 433ms/step - loss: 85.4802 - mae: 85.9778 - val_loss: 22.6563 - val_mae: 23.1457\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 26.5525 - mae: 27.0504 - val_loss: 13.5146 - val_mae: 14.0037\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 18.2863 - mae: 18.7816 - val_loss: 28.0054 - val_mae: 28.5054\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 22.8390 - mae: 23.3338 - val_loss: 12.9676 - val_mae: 13.4607\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 17.0733 - mae: 17.5693 - val_loss: 16.3390 - val_mae: 16.8259\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 18.3000 - mae: 18.7935 - val_loss: 14.0695 - val_mae: 14.5544\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 18.1836 - mae: 18.6797 - val_loss: 12.7052 - val_mae: 13.1948\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 18.3241 - mae: 18.8189 - val_loss: 12.8441 - val_mae: 13.3341\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 19.1765 - mae: 19.6697 - val_loss: 14.1367 - val_mae: 14.6157\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.7483 - mae: 16.2411 - val_loss: 13.5456 - val_mae: 14.0226\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 16.2833 - mae: 16.7765 - val_loss: 12.1867 - val_mae: 12.6709\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 16.1765 - mae: 16.6657 - val_loss: 12.2865 - val_mae: 12.7748\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 16.0421 - mae: 16.5362 - val_loss: 12.1733 - val_mae: 12.6665\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 16.0209 - mae: 16.5151 - val_loss: 12.2026 - val_mae: 12.6919\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.2015 - mae: 15.6924 - val_loss: 12.2024 - val_mae: 12.6892\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 15.2243 - mae: 15.7174 - val_loss: 12.0282 - val_mae: 12.5151\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.9949 - mae: 15.4896 - val_loss: 11.8648 - val_mae: 12.3608\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.8605 - mae: 15.3536 - val_loss: 11.7357 - val_mae: 12.2171\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.8228 - mae: 15.3157 - val_loss: 11.7476 - val_mae: 12.2397\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.6995 - mae: 15.1904 - val_loss: 11.5898 - val_mae: 12.0710\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.6346 - mae: 15.1267 - val_loss: 11.8097 - val_mae: 12.2940\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.5090 - mae: 15.0000 - val_loss: 11.5980 - val_mae: 12.0896\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.4156 - mae: 14.9120 - val_loss: 11.6407 - val_mae: 12.1261\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.2181 - mae: 14.7056 - val_loss: 11.5441 - val_mae: 12.0295\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 14.1843 - mae: 14.6760 - val_loss: 11.7968 - val_mae: 12.2959\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.1486 - mae: 14.6368 - val_loss: 11.6978 - val_mae: 12.1912\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 13.9793 - mae: 14.4666 - val_loss: 11.5990 - val_mae: 12.0859\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.9895 - mae: 14.4820 - val_loss: 11.7482 - val_mae: 12.2395\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.7786 - mae: 14.2633 - val_loss: 11.5862 - val_mae: 12.0805\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.6995 - mae: 14.1869 - val_loss: 12.5750 - val_mae: 13.0677\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 13.9068 - mae: 14.3973 - val_loss: 12.3320 - val_mae: 12.8262\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.9291 - mae: 14.4191 - val_loss: 11.6960 - val_mae: 12.1787\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 13.6811 - mae: 14.1670 - val_loss: 11.6507 - val_mae: 12.1430\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.1611 - mae: 14.6529 - val_loss: 11.9709 - val_mae: 12.4524\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 13.6924 - mae: 14.1732 - val_loss: 12.0011 - val_mae: 12.4966\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7357 - mae: 14.2266 - val_loss: 13.1801 - val_mae: 13.6641\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.0674 - mae: 14.5501 - val_loss: 12.7827 - val_mae: 13.2784\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.9321 - mae: 14.4208 - val_loss: 11.4625 - val_mae: 11.9499\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.2373 - mae: 13.7277 - val_loss: 11.5737 - val_mae: 12.0588\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1803 - mae: 13.6702 - val_loss: 11.6603 - val_mae: 12.1492\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.5787 - mae: 14.0685 - val_loss: 12.7831 - val_mae: 13.2665\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.6829 - mae: 14.1705 - val_loss: 11.6905 - val_mae: 12.1804\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.1937 - mae: 13.6797 - val_loss: 11.9395 - val_mae: 12.4286\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 13.2864 - mae: 13.7807 - val_loss: 11.7248 - val_mae: 12.2156\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.9355 - mae: 13.4160 - val_loss: 12.0524 - val_mae: 12.5501\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.3529 - mae: 13.8427 - val_loss: 11.4015 - val_mae: 11.8892\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.9418 - mae: 13.4308 - val_loss: 11.4600 - val_mae: 11.9568\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.7760 - mae: 13.2559 - val_loss: 11.5102 - val_mae: 11.9888\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9489 - mae: 13.4419 - val_loss: 12.2888 - val_mae: 12.7763\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1638 - mae: 13.6493 - val_loss: 12.8929 - val_mae: 13.3846\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.3450 - mae: 13.8364 - val_loss: 11.5213 - val_mae: 12.0193\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1339 - mae: 13.6230 - val_loss: 11.4799 - val_mae: 11.9653\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.5830 - mae: 13.0699 - val_loss: 12.5941 - val_mae: 13.0864\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3802 - mae: 13.8655 - val_loss: 13.2900 - val_mae: 13.7884\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.5865 - mae: 14.0750 - val_loss: 12.4923 - val_mae: 12.9830\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.9689 - mae: 13.4570 - val_loss: 12.6710 - val_mae: 13.1452\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4122 - mae: 13.9074 - val_loss: 11.5314 - val_mae: 12.0314\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.6694 - mae: 13.1619 - val_loss: 11.6533 - val_mae: 12.1423\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.4513 - mae: 13.9446 - val_loss: 12.8749 - val_mae: 13.3611\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.2816 - mae: 13.7697 - val_loss: 12.7061 - val_mae: 13.2014\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1629 - mae: 13.6533 - val_loss: 11.5178 - val_mae: 12.0030\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 12.9760 - mae: 13.4662 - val_loss: 11.4471 - val_mae: 11.9351\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.3802 - mae: 12.8688 - val_loss: 11.8157 - val_mae: 12.3069\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.2925 - mae: 13.7809 - val_loss: 13.6034 - val_mae: 14.0940\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4428 - mae: 13.9329 - val_loss: 12.2384 - val_mae: 12.7320\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.7511 - mae: 13.2403 - val_loss: 11.5303 - val_mae: 12.0170\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 538ms/step - loss: 86.0520 - mae: 86.5511 - val_loss: 10.6401 - val_mae: 11.1384\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 30.1201 - mae: 30.6159 - val_loss: 11.7824 - val_mae: 12.2704\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 20.4666 - mae: 20.9624 - val_loss: 13.0038 - val_mae: 13.4964\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 20.6216 - mae: 21.1160 - val_loss: 10.9608 - val_mae: 11.4595\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 16.3770 - mae: 16.8708 - val_loss: 14.0830 - val_mae: 14.5764\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 17.2408 - mae: 17.7309 - val_loss: 13.9052 - val_mae: 14.3990\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 18.1822 - mae: 18.6758 - val_loss: 9.3878 - val_mae: 9.8682\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 19.2729 - mae: 19.7682 - val_loss: 9.9337 - val_mae: 10.4243\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 18.9229 - mae: 19.4181 - val_loss: 14.2140 - val_mae: 14.7080\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.7759 - mae: 15.2726 - val_loss: 9.0896 - val_mae: 9.5640\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 16.9602 - mae: 17.4530 - val_loss: 12.4739 - val_mae: 12.9700\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 14.8302 - mae: 15.3250 - val_loss: 8.8155 - val_mae: 9.3090\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 15.5617 - mae: 16.0492 - val_loss: 11.6917 - val_mae: 12.1779\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 16.2211 - mae: 16.7162 - val_loss: 9.5503 - val_mae: 10.0384\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 15.3439 - mae: 15.8354 - val_loss: 8.9395 - val_mae: 9.4220\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 15.1958 - mae: 15.6906 - val_loss: 11.3151 - val_mae: 11.8128\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 14.4172 - mae: 14.9105 - val_loss: 8.6481 - val_mae: 9.1428\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.2095 - mae: 14.6970 - val_loss: 9.3379 - val_mae: 9.8164\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.9579 - mae: 14.4502 - val_loss: 9.1039 - val_mae: 9.5989\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.0347 - mae: 14.5206 - val_loss: 8.7255 - val_mae: 9.2254\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.8424 - mae: 14.3378 - val_loss: 9.0371 - val_mae: 9.5226\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.2939 - mae: 14.7888 - val_loss: 8.3591 - val_mae: 8.8364\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.0208 - mae: 14.5055 - val_loss: 9.0226 - val_mae: 9.5094\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.4264 - mae: 14.9220 - val_loss: 8.6729 - val_mae: 9.1663\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.8303 - mae: 14.3245 - val_loss: 8.8559 - val_mae: 9.3443\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.4006 - mae: 14.8910 - val_loss: 8.7419 - val_mae: 9.2310\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.5453 - mae: 14.0373 - val_loss: 8.4857 - val_mae: 8.9829\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.8857 - mae: 14.3795 - val_loss: 8.3952 - val_mae: 8.8707\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.5339 - mae: 14.0263 - val_loss: 8.9739 - val_mae: 9.4576\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.0892 - mae: 14.5843 - val_loss: 8.3565 - val_mae: 8.8375\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.8734 - mae: 14.3679 - val_loss: 9.3857 - val_mae: 9.8857\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 14.2036 - mae: 14.6998 - val_loss: 8.3887 - val_mae: 8.8790\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 13.6042 - mae: 14.0925 - val_loss: 9.3250 - val_mae: 9.8173\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.5289 - mae: 15.0276 - val_loss: 8.4187 - val_mae: 8.9165\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.7392 - mae: 14.2265 - val_loss: 9.1981 - val_mae: 9.6926\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 14.0094 - mae: 14.5053 - val_loss: 8.3838 - val_mae: 8.8664\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.2628 - mae: 13.7508 - val_loss: 8.8467 - val_mae: 9.3407\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.5309 - mae: 14.0242 - val_loss: 8.3547 - val_mae: 8.8429\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9220 - mae: 13.4105 - val_loss: 9.2562 - val_mae: 9.7462\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.6332 - mae: 14.1292 - val_loss: 8.5057 - val_mae: 8.9932\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.4557 - mae: 13.9477 - val_loss: 10.1209 - val_mae: 10.6046\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.8157 - mae: 15.3089 - val_loss: 8.3502 - val_mae: 8.8312\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.7678 - mae: 14.2556 - val_loss: 8.6516 - val_mae: 9.1264\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 14.1331 - mae: 14.6276 - val_loss: 8.3994 - val_mae: 8.8850\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.2431 - mae: 13.7369 - val_loss: 8.8892 - val_mae: 9.3809\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.8619 - mae: 14.3528 - val_loss: 8.3989 - val_mae: 8.8954\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.3380 - mae: 13.8279 - val_loss: 9.7360 - val_mae: 10.2279\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.1100 - mae: 14.6025 - val_loss: 8.2823 - val_mae: 8.7595\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.6728 - mae: 14.1656 - val_loss: 8.6223 - val_mae: 9.1100\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.6573 - mae: 14.1504 - val_loss: 8.3322 - val_mae: 8.8171\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.8492 - mae: 13.3410 - val_loss: 8.4024 - val_mae: 8.8843\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.2344 - mae: 13.7294 - val_loss: 8.3946 - val_mae: 8.8831\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6669 - mae: 13.1584 - val_loss: 9.1918 - val_mae: 9.6912\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.2039 - mae: 13.6976 - val_loss: 8.5698 - val_mae: 9.0461\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.6812 - mae: 14.1736 - val_loss: 9.2901 - val_mae: 9.7849\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 14.3857 - mae: 14.8783 - val_loss: 8.4294 - val_mae: 8.9275\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.2304 - mae: 13.7214 - val_loss: 8.3411 - val_mae: 8.8209\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4144 - mae: 13.9119 - val_loss: 8.2682 - val_mae: 8.7634\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.1832 - mae: 12.6708 - val_loss: 8.2767 - val_mae: 8.7500\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.5160 - mae: 13.0147 - val_loss: 8.3290 - val_mae: 8.8279\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.4289 - mae: 12.9155 - val_loss: 8.8018 - val_mae: 9.2875\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.6568 - mae: 13.1489 - val_loss: 8.6705 - val_mae: 9.1352\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.1146 - mae: 12.6032 - val_loss: 8.5978 - val_mae: 9.0870\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.9738 - mae: 12.4654 - val_loss: 8.5065 - val_mae: 8.9968\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.6221 - mae: 12.1096 - val_loss: 8.5897 - val_mae: 9.0720\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.9065 - mae: 12.3944 - val_loss: 8.9597 - val_mae: 9.4440\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8466 - mae: 12.3302 - val_loss: 8.5893 - val_mae: 9.0767\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.8904 - mae: 12.3855 - val_loss: 8.4397 - val_mae: 8.9266\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7842 - mae: 12.2696 - val_loss: 8.3230 - val_mae: 8.8212\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.3591 - mae: 11.8457 - val_loss: 8.2976 - val_mae: 8.7829\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.2635 - mae: 11.7595 - val_loss: 8.4285 - val_mae: 8.9160\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.3245 - mae: 11.8107 - val_loss: 8.5328 - val_mae: 9.0236\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.5100 - mae: 12.0073 - val_loss: 8.3274 - val_mae: 8.8177\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.3473 - mae: 11.8415 - val_loss: 8.4336 - val_mae: 8.9092\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 26ms/step - loss: 11.6427 - mae: 12.1340 - val_loss: 8.4771 - val_mae: 8.9669\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.1763 - mae: 11.6615 - val_loss: 8.2977 - val_mae: 8.7891\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 10.9029 - mae: 11.3961 - val_loss: 8.3185 - val_mae: 8.8182\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.1475 - mae: 11.6350 - val_loss: 8.3323 - val_mae: 8.8169\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 444ms/step - loss: 77.3664 - mae: 77.8663 - val_loss: 16.0426 - val_mae: 16.5396\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 23.5739 - mae: 24.0706 - val_loss: 11.7737 - val_mae: 12.2641\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.8680 - mae: 15.3613 - val_loss: 15.3416 - val_mae: 15.8328\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 15.2436 - mae: 15.7350 - val_loss: 10.5712 - val_mae: 11.0555\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.7833 - mae: 13.2750 - val_loss: 10.4915 - val_mae: 10.9709\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.0395 - mae: 13.5333 - val_loss: 11.0554 - val_mae: 11.5413\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9221 - mae: 13.4104 - val_loss: 10.2004 - val_mae: 10.6868\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.9707 - mae: 13.4630 - val_loss: 10.7451 - val_mae: 11.2398\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.1140 - mae: 13.6065 - val_loss: 10.1654 - val_mae: 10.6447\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.8731 - mae: 13.3653 - val_loss: 10.1236 - val_mae: 10.6196\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1442 - mae: 13.6395 - val_loss: 10.1319 - val_mae: 10.6194\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.6590 - mae: 13.1429 - val_loss: 9.7550 - val_mae: 10.2504\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3453 - mae: 13.8360 - val_loss: 10.6433 - val_mae: 11.1265\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.9655 - mae: 13.4535 - val_loss: 10.1640 - val_mae: 10.6349\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.0855 - mae: 14.5829 - val_loss: 10.1212 - val_mae: 10.6122\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7526 - mae: 14.2407 - val_loss: 9.5609 - val_mae: 10.0512\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.9070 - mae: 14.3992 - val_loss: 9.4586 - val_mae: 9.9497\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.6925 - mae: 13.1866 - val_loss: 9.3919 - val_mae: 9.8802\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.0621 - mae: 13.5593 - val_loss: 9.4102 - val_mae: 9.8887\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.3221 - mae: 12.8148 - val_loss: 9.3539 - val_mae: 9.8319\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.8970 - mae: 13.3931 - val_loss: 9.3516 - val_mae: 9.8336\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 11.9619 - mae: 12.4570 - val_loss: 9.3042 - val_mae: 9.7860\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.3774 - mae: 12.8696 - val_loss: 9.6854 - val_mae: 10.1757\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0133 - mae: 12.5032 - val_loss: 9.2320 - val_mae: 9.7195\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.4591 - mae: 12.9549 - val_loss: 9.7870 - val_mae: 10.2787\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0795 - mae: 12.5664 - val_loss: 9.1782 - val_mae: 9.6720\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.5623 - mae: 13.0577 - val_loss: 10.2836 - val_mae: 10.7663\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0295 - mae: 12.5157 - val_loss: 9.1830 - val_mae: 9.6700\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5265 - mae: 13.0223 - val_loss: 10.1911 - val_mae: 10.6775\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.9661 - mae: 12.4562 - val_loss: 9.1956 - val_mae: 9.6785\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.2263 - mae: 12.7200 - val_loss: 9.8651 - val_mae: 10.3571\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.9977 - mae: 12.4890 - val_loss: 9.2026 - val_mae: 9.6928\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5610 - mae: 13.0510 - val_loss: 9.7071 - val_mae: 10.1959\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 12.0037 - mae: 12.4961 - val_loss: 9.2791 - val_mae: 9.7632\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.2278 - mae: 12.7205 - val_loss: 9.4691 - val_mae: 9.9542\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0566 - mae: 12.5486 - val_loss: 9.1756 - val_mae: 9.6707\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6382 - mae: 13.1291 - val_loss: 9.6677 - val_mae: 10.1548\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 94ms/step - loss: 12.1704 - mae: 12.6627 - val_loss: 9.1793 - val_mae: 9.6633\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 72ms/step - loss: 12.5263 - mae: 13.0173 - val_loss: 9.5913 - val_mae: 10.0755\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 67ms/step - loss: 11.7015 - mae: 12.1927 - val_loss: 9.1735 - val_mae: 9.6547\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7802 - mae: 12.2672 - val_loss: 9.6004 - val_mae: 10.0791\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.4151 - mae: 11.9061 - val_loss: 9.1770 - val_mae: 9.6614\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.2434 - mae: 11.7336 - val_loss: 9.7177 - val_mae: 10.2051\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 85ms/step - loss: 11.0622 - mae: 11.5508 - val_loss: 9.1454 - val_mae: 9.6377\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 94ms/step - loss: 10.9015 - mae: 11.3838 - val_loss: 9.6002 - val_mae: 10.0857\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 76ms/step - loss: 10.6278 - mae: 11.1177 - val_loss: 9.0915 - val_mae: 9.5709\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.5919 - mae: 11.0756 - val_loss: 9.7697 - val_mae: 10.2573\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.6200 - mae: 11.1083 - val_loss: 9.1427 - val_mae: 9.6278\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.4709 - mae: 10.9526 - val_loss: 10.0810 - val_mae: 10.5654\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 41ms/step - loss: 10.6426 - mae: 11.1327 - val_loss: 9.1045 - val_mae: 9.5863\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 87ms/step - loss: 10.6879 - mae: 11.1777 - val_loss: 11.4004 - val_mae: 11.8970\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 122ms/step - loss: 11.0114 - mae: 11.5033 - val_loss: 9.3329 - val_mae: 9.8270\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 61ms/step - loss: 10.8837 - mae: 11.3789 - val_loss: 10.9055 - val_mae: 11.3951\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.9235 - mae: 11.4119 - val_loss: 9.1182 - val_mae: 9.6048\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 10.2704 - mae: 10.7536 - val_loss: 9.6124 - val_mae: 10.1003\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.3500 - mae: 10.8347 - val_loss: 9.3412 - val_mae: 9.8210\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.1784 - mae: 10.6638 - val_loss: 9.2493 - val_mae: 9.7351\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.1296 - mae: 10.6101 - val_loss: 9.3590 - val_mae: 9.8396\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.2316 - mae: 10.7192 - val_loss: 9.5621 - val_mae: 10.0430\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.2083 - mae: 10.6985 - val_loss: 9.2646 - val_mae: 9.7390\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.0500 - mae: 10.5283 - val_loss: 9.4213 - val_mae: 9.9132\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.0960 - mae: 10.5825 - val_loss: 9.5222 - val_mae: 10.0078\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.0674 - mae: 10.5547 - val_loss: 9.1885 - val_mae: 9.6694\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.9858 - mae: 10.4602 - val_loss: 9.4739 - val_mae: 9.9600\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.0038 - mae: 10.4845 - val_loss: 9.2069 - val_mae: 9.6783\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.9341 - mae: 10.4099 - val_loss: 9.7216 - val_mae: 10.2035\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 425ms/step - loss: 83.3656 - mae: 83.8648 - val_loss: 13.7215 - val_mae: 14.1909\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 26.6662 - mae: 27.1642 - val_loss: 10.9052 - val_mae: 11.3915\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 18.4003 - mae: 18.8955 - val_loss: 16.9669 - val_mae: 17.4588\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 19.7237 - mae: 20.2173 - val_loss: 11.7247 - val_mae: 12.2205\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.6902 - mae: 15.1769 - val_loss: 10.8812 - val_mae: 11.3598\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 15.1696 - mae: 15.6617 - val_loss: 13.0528 - val_mae: 13.5454\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 14.7628 - mae: 15.2497 - val_loss: 11.1872 - val_mae: 11.6669\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.6353 - mae: 15.1282 - val_loss: 13.6848 - val_mae: 14.1765\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 15.2603 - mae: 15.7553 - val_loss: 10.7428 - val_mae: 11.2208\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 14.8786 - mae: 15.3654 - val_loss: 12.2193 - val_mae: 12.7081\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 15.8716 - mae: 16.3665 - val_loss: 10.8324 - val_mae: 11.3154\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.6731 - mae: 15.1661 - val_loss: 11.3465 - val_mae: 11.8455\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 16.1382 - mae: 16.6312 - val_loss: 10.8300 - val_mae: 11.3223\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 14.9183 - mae: 15.4151 - val_loss: 10.7070 - val_mae: 11.1964\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 15.5994 - mae: 16.0897 - val_loss: 11.5149 - val_mae: 12.0138\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.2438 - mae: 14.7379 - val_loss: 10.5595 - val_mae: 11.0515\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.2409 - mae: 14.7322 - val_loss: 10.9494 - val_mae: 11.4389\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.6437 - mae: 14.1344 - val_loss: 10.7219 - val_mae: 11.2109\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.1196 - mae: 14.6090 - val_loss: 10.4710 - val_mae: 10.9527\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 13.5983 - mae: 14.0893 - val_loss: 11.0790 - val_mae: 11.5594\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.2084 - mae: 14.7052 - val_loss: 10.3193 - val_mae: 10.8152\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.8050 - mae: 14.2980 - val_loss: 11.0023 - val_mae: 11.5020\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 14.6855 - mae: 15.1815 - val_loss: 10.3055 - val_mae: 10.7986\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.1909 - mae: 14.6837 - val_loss: 10.4052 - val_mae: 10.8932\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 14.8925 - mae: 15.3839 - val_loss: 10.6075 - val_mae: 11.0939\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.8715 - mae: 14.3663 - val_loss: 10.2075 - val_mae: 10.7041\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 14.8272 - mae: 15.3199 - val_loss: 10.4706 - val_mae: 10.9585\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 13.2992 - mae: 13.7937 - val_loss: 10.2594 - val_mae: 10.7518\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 14.4081 - mae: 14.9027 - val_loss: 10.3417 - val_mae: 10.8294\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.1827 - mae: 13.6660 - val_loss: 10.3156 - val_mae: 10.8088\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4612 - mae: 13.9534 - val_loss: 10.3102 - val_mae: 10.7989\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.8930 - mae: 13.3832 - val_loss: 10.2224 - val_mae: 10.7049\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.5859 - mae: 14.0800 - val_loss: 10.0648 - val_mae: 10.5477\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.1382 - mae: 13.6280 - val_loss: 10.4591 - val_mae: 10.9518\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.9800 - mae: 14.4750 - val_loss: 10.2528 - val_mae: 10.7430\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 14.0777 - mae: 14.5648 - val_loss: 11.1762 - val_mae: 11.6728\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 15.8680 - mae: 16.3669 - val_loss: 10.4543 - val_mae: 10.9258\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 14.7611 - mae: 15.2582 - val_loss: 10.1450 - val_mae: 10.6217\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 14.4604 - mae: 14.9554 - val_loss: 10.9133 - val_mae: 11.4118\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 44ms/step - loss: 13.0663 - mae: 13.5551 - val_loss: 10.1107 - val_mae: 10.5900\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.0599 - mae: 13.5467 - val_loss: 10.1757 - val_mae: 10.6575\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.4350 - mae: 12.9206 - val_loss: 10.0074 - val_mae: 10.4949\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.9114 - mae: 13.3966 - val_loss: 9.9915 - val_mae: 10.4772\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5070 - mae: 12.9957 - val_loss: 10.0653 - val_mae: 10.5472\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.4904 - mae: 12.9859 - val_loss: 9.9250 - val_mae: 10.4124\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.3515 - mae: 12.8420 - val_loss: 10.3672 - val_mae: 10.8611\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.9193 - mae: 13.4083 - val_loss: 10.4277 - val_mae: 10.9137\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 12.7094 - mae: 13.1970 - val_loss: 10.8602 - val_mae: 11.3346\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.0567 - mae: 13.5516 - val_loss: 10.3756 - val_mae: 10.8658\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.9802 - mae: 13.4684 - val_loss: 11.4325 - val_mae: 11.9250\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 14.2263 - mae: 14.7186 - val_loss: 10.5223 - val_mae: 11.0100\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 13.4755 - mae: 13.9671 - val_loss: 10.7275 - val_mae: 11.2214\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.6488 - mae: 14.1426 - val_loss: 10.1628 - val_mae: 10.6457\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.6734 - mae: 13.1623 - val_loss: 10.1165 - val_mae: 10.6136\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.9466 - mae: 13.4374 - val_loss: 10.3152 - val_mae: 10.7990\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.1499 - mae: 12.6409 - val_loss: 10.2479 - val_mae: 10.7381\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.2314 - mae: 12.7232 - val_loss: 10.0893 - val_mae: 10.5707\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 12.1949 - mae: 12.6867 - val_loss: 10.5615 - val_mae: 11.0420\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5618 - mae: 13.0517 - val_loss: 10.1560 - val_mae: 10.6371\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.0917 - mae: 12.5793 - val_loss: 10.8073 - val_mae: 11.2875\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 12.5182 - mae: 13.0115 - val_loss: 10.3515 - val_mae: 10.8369\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1893 - mae: 12.6824 - val_loss: 10.7155 - val_mae: 11.1996\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.3658 - mae: 12.8579 - val_loss: 10.5512 - val_mae: 11.0313\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5481 - mae: 13.0382 - val_loss: 12.1551 - val_mae: 12.6443\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.7536 - mae: 14.2492 - val_loss: 10.8837 - val_mae: 11.3722\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 6s 938ms/step - loss: 77.8184 - mae: 78.3175 - val_loss: 15.6479 - val_mae: 16.1416\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 49ms/step - loss: 26.3758 - mae: 26.8710 - val_loss: 11.0106 - val_mae: 11.5012\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 13.5732 - mae: 14.0656 - val_loss: 19.6834 - val_mae: 20.1834\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 96ms/step - loss: 16.1372 - mae: 16.6286 - val_loss: 12.1287 - val_mae: 12.6287\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 70ms/step - loss: 12.4477 - mae: 12.9424 - val_loss: 12.7935 - val_mae: 13.2920\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 64ms/step - loss: 13.3920 - mae: 13.8840 - val_loss: 12.5882 - val_mae: 13.0751\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 12.9858 - mae: 13.4778 - val_loss: 10.7796 - val_mae: 11.2665\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 14.0220 - mae: 14.5141 - val_loss: 12.4577 - val_mae: 12.9486\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 45ms/step - loss: 16.4558 - mae: 16.9543 - val_loss: 12.1115 - val_mae: 12.6096\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 12.4413 - mae: 12.9352 - val_loss: 10.5036 - val_mae: 10.9734\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 83ms/step - loss: 12.3467 - mae: 12.8374 - val_loss: 10.4605 - val_mae: 10.9508\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 69ms/step - loss: 11.9626 - mae: 12.4568 - val_loss: 10.2702 - val_mae: 10.7555\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 74ms/step - loss: 11.9258 - mae: 12.4134 - val_loss: 10.7433 - val_mae: 11.2272\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 52ms/step - loss: 12.6881 - mae: 13.1814 - val_loss: 10.1366 - val_mae: 10.6104\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 78ms/step - loss: 12.2238 - mae: 12.7108 - val_loss: 10.0086 - val_mae: 10.4988\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 49ms/step - loss: 12.5570 - mae: 13.0515 - val_loss: 10.1834 - val_mae: 10.6771\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 98ms/step - loss: 11.9544 - mae: 12.4413 - val_loss: 10.0317 - val_mae: 10.5008\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 46ms/step - loss: 13.0786 - mae: 13.5758 - val_loss: 10.3512 - val_mae: 10.8384\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 66ms/step - loss: 11.9585 - mae: 12.4527 - val_loss: 9.7668 - val_mae: 10.2550\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.5926 - mae: 13.0819 - val_loss: 10.1892 - val_mae: 10.6753\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 117ms/step - loss: 11.3947 - mae: 11.8829 - val_loss: 9.7022 - val_mae: 10.1834\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 74ms/step - loss: 11.9934 - mae: 12.4879 - val_loss: 11.0357 - val_mae: 11.5220\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 11.7297 - mae: 12.2214 - val_loss: 9.6801 - val_mae: 10.1722\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 79ms/step - loss: 11.8447 - mae: 12.3383 - val_loss: 11.2104 - val_mae: 11.7049\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 103ms/step - loss: 11.6771 - mae: 12.1643 - val_loss: 9.9694 - val_mae: 10.4417\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 12.5261 - mae: 13.0239 - val_loss: 10.6049 - val_mae: 11.0976\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 51ms/step - loss: 12.1276 - mae: 12.6178 - val_loss: 9.6862 - val_mae: 10.1779\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 54ms/step - loss: 13.6435 - mae: 14.1403 - val_loss: 9.7051 - val_mae: 10.1984\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.1216 - mae: 12.6173 - val_loss: 9.9368 - val_mae: 10.4277\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.5800 - mae: 13.0712 - val_loss: 9.5507 - val_mae: 10.0297\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 84ms/step - loss: 11.3669 - mae: 11.8567 - val_loss: 9.8797 - val_mae: 10.3664\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 79ms/step - loss: 11.9032 - mae: 12.3967 - val_loss: 9.8634 - val_mae: 10.3485\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 106ms/step - loss: 11.0490 - mae: 11.5378 - val_loss: 9.7892 - val_mae: 10.2772\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 93ms/step - loss: 11.7232 - mae: 12.2170 - val_loss: 10.3018 - val_mae: 10.7974\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 72ms/step - loss: 11.0113 - mae: 11.4967 - val_loss: 9.6554 - val_mae: 10.1438\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 63ms/step - loss: 11.0372 - mae: 11.5317 - val_loss: 10.3380 - val_mae: 10.8303\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 74ms/step - loss: 10.8437 - mae: 11.3244 - val_loss: 9.5873 - val_mae: 10.0801\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 79ms/step - loss: 10.9187 - mae: 11.4106 - val_loss: 10.2084 - val_mae: 10.7008\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 72ms/step - loss: 10.6521 - mae: 11.1416 - val_loss: 9.5718 - val_mae: 10.0635\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 51ms/step - loss: 10.9261 - mae: 11.4183 - val_loss: 10.2536 - val_mae: 10.7485\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 52ms/step - loss: 10.8293 - mae: 11.3135 - val_loss: 9.3608 - val_mae: 9.8449\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.6771 - mae: 12.1693 - val_loss: 10.6207 - val_mae: 11.1080\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 50ms/step - loss: 11.3257 - mae: 11.8088 - val_loss: 9.3530 - val_mae: 9.8268\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 48ms/step - loss: 11.5379 - mae: 12.0308 - val_loss: 10.2013 - val_mae: 10.6949\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 105ms/step - loss: 11.0367 - mae: 11.5224 - val_loss: 9.2982 - val_mae: 9.7828\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 125ms/step - loss: 11.5808 - mae: 12.0738 - val_loss: 10.5655 - val_mae: 11.0546\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 92ms/step - loss: 11.4588 - mae: 11.9446 - val_loss: 9.2831 - val_mae: 9.7758\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 82ms/step - loss: 11.7556 - mae: 12.2449 - val_loss: 10.3037 - val_mae: 10.7905\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 96ms/step - loss: 11.1736 - mae: 11.6608 - val_loss: 9.2592 - val_mae: 9.7336\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 92ms/step - loss: 11.0145 - mae: 11.5039 - val_loss: 10.2290 - val_mae: 10.7179\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 10.8819 - mae: 11.3695 - val_loss: 9.2595 - val_mae: 9.7381\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 82ms/step - loss: 11.3309 - mae: 11.8256 - val_loss: 10.5572 - val_mae: 11.0501\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 94ms/step - loss: 10.9641 - mae: 11.4511 - val_loss: 9.2631 - val_mae: 9.7532\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.7806 - mae: 12.2726 - val_loss: 10.4806 - val_mae: 10.9743\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.4809 - mae: 11.9636 - val_loss: 9.2350 - val_mae: 9.7282\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 12.4717 - mae: 12.9652 - val_loss: 10.0734 - val_mae: 10.5625\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.5552 - mae: 12.0422 - val_loss: 9.1789 - val_mae: 9.6611\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.6536 - mae: 12.1455 - val_loss: 9.9512 - val_mae: 10.4402\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.1324 - mae: 11.6171 - val_loss: 9.1540 - val_mae: 9.6320\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0239 - mae: 12.5147 - val_loss: 9.4357 - val_mae: 9.9273\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.8940 - mae: 11.3844 - val_loss: 9.5602 - val_mae: 10.0520\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 11.5325 - mae: 12.0239 - val_loss: 9.4527 - val_mae: 9.9487\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.4273 - mae: 10.9122 - val_loss: 9.6853 - val_mae: 10.1712\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.0716 - mae: 11.5611 - val_loss: 9.5263 - val_mae: 10.0152\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.0819 - mae: 10.5722 - val_loss: 9.5057 - val_mae: 9.9942\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.3734 - mae: 10.8603 - val_loss: 9.5211 - val_mae: 10.0154\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.8871 - mae: 10.3702 - val_loss: 9.3142 - val_mae: 9.8008\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.9275 - mae: 10.4176 - val_loss: 9.3948 - val_mae: 9.8848\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.7826 - mae: 10.2714 - val_loss: 9.6769 - val_mae: 10.1757\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 9.8104 - mae: 10.3008 - val_loss: 9.3705 - val_mae: 9.8596\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.7197 - mae: 10.2024 - val_loss: 9.4211 - val_mae: 9.9122\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.7318 - mae: 10.2133 - val_loss: 9.4392 - val_mae: 9.9285\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.6639 - mae: 10.1482 - val_loss: 9.3883 - val_mae: 9.8783\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.7004 - mae: 10.1865 - val_loss: 9.5280 - val_mae: 10.0206\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.6933 - mae: 10.1855 - val_loss: 9.6768 - val_mae: 10.1718\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.6556 - mae: 10.1437 - val_loss: 9.5311 - val_mae: 10.0266\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.5877 - mae: 10.0695 - val_loss: 9.3911 - val_mae: 9.8839\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 90ms/step - loss: 9.6688 - mae: 10.1538 - val_loss: 9.7153 - val_mae: 10.2139\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 65ms/step - loss: 9.5709 - mae: 10.0572 - val_loss: 9.4419 - val_mae: 9.9333\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 7s 766ms/step - loss: 78.6555 - mae: 79.1554 - val_loss: 14.8940 - val_mae: 15.3867\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 80ms/step - loss: 27.8969 - mae: 28.3967 - val_loss: 10.0784 - val_mae: 10.5560\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 112ms/step - loss: 14.0287 - mae: 14.5214 - val_loss: 22.7241 - val_mae: 23.2241\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 74ms/step - loss: 18.9262 - mae: 19.4164 - val_loss: 10.9856 - val_mae: 11.4679\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 93ms/step - loss: 13.6214 - mae: 14.1165 - val_loss: 11.8423 - val_mae: 12.3286\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 82ms/step - loss: 14.3974 - mae: 14.8890 - val_loss: 11.6729 - val_mae: 12.1627\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 15.1107 - mae: 15.6027 - val_loss: 10.6891 - val_mae: 11.1824\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 71ms/step - loss: 14.5193 - mae: 15.0084 - val_loss: 10.0044 - val_mae: 10.4906\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 90ms/step - loss: 13.8164 - mae: 14.3077 - val_loss: 11.8290 - val_mae: 12.3188\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 89ms/step - loss: 12.5632 - mae: 13.0488 - val_loss: 10.0895 - val_mae: 10.5805\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 62ms/step - loss: 12.6852 - mae: 13.1693 - val_loss: 11.4696 - val_mae: 11.9541\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 70ms/step - loss: 13.5300 - mae: 14.0241 - val_loss: 10.2470 - val_mae: 10.7315\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.0879 - mae: 13.5784 - val_loss: 10.0583 - val_mae: 10.5463\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.4403 - mae: 13.9292 - val_loss: 11.0071 - val_mae: 11.5021\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0771 - mae: 12.5703 - val_loss: 9.7849 - val_mae: 10.2612\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2315 - mae: 12.7157 - val_loss: 10.3300 - val_mae: 10.8240\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8587 - mae: 12.3498 - val_loss: 9.8362 - val_mae: 10.3232\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.9159 - mae: 12.3955 - val_loss: 9.6132 - val_mae: 10.0861\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7540 - mae: 12.2470 - val_loss: 9.9656 - val_mae: 10.4461\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.7243 - mae: 12.2121 - val_loss: 9.5941 - val_mae: 10.0840\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.5750 - mae: 12.0628 - val_loss: 9.7433 - val_mae: 10.2402\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.4957 - mae: 11.9815 - val_loss: 9.6485 - val_mae: 10.1365\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 11.5457 - mae: 12.0350 - val_loss: 9.7107 - val_mae: 10.2028\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.5003 - mae: 11.9886 - val_loss: 9.9786 - val_mae: 10.4715\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 11.5338 - mae: 12.0188 - val_loss: 9.9384 - val_mae: 10.4382\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.7058 - mae: 12.1990 - val_loss: 10.4922 - val_mae: 10.9761\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.0081 - mae: 12.4920 - val_loss: 9.8671 - val_mae: 10.3671\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.5816 - mae: 13.0727 - val_loss: 9.5049 - val_mae: 9.9941\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.2637 - mae: 12.7489 - val_loss: 9.6047 - val_mae: 10.0970\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.4540 - mae: 13.9499 - val_loss: 9.5319 - val_mae: 10.0191\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.3441 - mae: 12.8322 - val_loss: 9.3057 - val_mae: 9.7852\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 12.5444 - mae: 13.0366 - val_loss: 9.2857 - val_mae: 9.7627\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 11.8844 - mae: 12.3765 - val_loss: 9.2143 - val_mae: 9.7087\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.6734 - mae: 13.1665 - val_loss: 9.2185 - val_mae: 9.7122\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.5798 - mae: 12.0640 - val_loss: 9.2145 - val_mae: 9.7076\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.9893 - mae: 12.4821 - val_loss: 9.4082 - val_mae: 9.9043\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.4105 - mae: 11.8916 - val_loss: 9.1826 - val_mae: 9.6541\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.8319 - mae: 12.3259 - val_loss: 9.3740 - val_mae: 9.8704\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.1672 - mae: 11.6530 - val_loss: 9.1694 - val_mae: 9.6446\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.5812 - mae: 12.0700 - val_loss: 9.7010 - val_mae: 10.1924\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 11.2668 - mae: 11.7525 - val_loss: 9.1184 - val_mae: 9.5981\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.0736 - mae: 12.5668 - val_loss: 9.8729 - val_mae: 10.3637\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.5687 - mae: 12.0562 - val_loss: 9.1617 - val_mae: 9.6360\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.8173 - mae: 12.3074 - val_loss: 9.7484 - val_mae: 10.2362\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.3443 - mae: 11.8289 - val_loss: 9.2226 - val_mae: 9.7066\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.8209 - mae: 12.3163 - val_loss: 9.7980 - val_mae: 10.2898\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.6465 - mae: 12.1355 - val_loss: 9.2553 - val_mae: 9.7392\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 12.4730 - mae: 12.9669 - val_loss: 9.3071 - val_mae: 9.7812\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 50ms/step - loss: 11.9349 - mae: 12.4214 - val_loss: 9.1661 - val_mae: 9.6348\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.7373 - mae: 13.2271 - val_loss: 9.1206 - val_mae: 9.6107\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.7095 - mae: 12.1996 - val_loss: 9.0662 - val_mae: 9.5595\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 12.6157 - mae: 13.1114 - val_loss: 9.2470 - val_mae: 9.7383\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.3612 - mae: 12.8531 - val_loss: 9.0239 - val_mae: 9.5010\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 13.8479 - mae: 14.3441 - val_loss: 9.2986 - val_mae: 9.7914\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.5184 - mae: 12.0088 - val_loss: 9.5700 - val_mae: 10.0426\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1006 - mae: 12.5877 - val_loss: 9.0369 - val_mae: 9.5241\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.0036 - mae: 11.4853 - val_loss: 9.4882 - val_mae: 9.9653\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.0884 - mae: 11.5751 - val_loss: 8.9993 - val_mae: 9.4918\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.7789 - mae: 11.2749 - val_loss: 9.2263 - val_mae: 9.7156\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.1944 - mae: 11.6851 - val_loss: 9.1128 - val_mae: 9.6092\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.4852 - mae: 10.9695 - val_loss: 9.3021 - val_mae: 9.7924\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.8383 - mae: 11.3266 - val_loss: 9.3728 - val_mae: 9.8605\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 10.3952 - mae: 10.8786 - val_loss: 9.3381 - val_mae: 9.8273\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 10.4501 - mae: 10.9331 - val_loss: 9.4700 - val_mae: 9.9463\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.3154 - mae: 10.7991 - val_loss: 9.0878 - val_mae: 9.5782\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 10.3036 - mae: 10.7924 - val_loss: 9.6199 - val_mae: 10.1044\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 10.3096 - mae: 10.7980 - val_loss: 9.0937 - val_mae: 9.5929\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 10.2142 - mae: 10.6972 - val_loss: 9.0825 - val_mae: 9.5825\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.4196 - mae: 10.9110 - val_loss: 9.8882 - val_mae: 10.3837\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.4187 - mae: 10.9047 - val_loss: 8.9818 - val_mae: 9.4775\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.0491 - mae: 10.5346 - val_loss: 9.9104 - val_mae: 10.3887\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 10.2261 - mae: 10.7127 - val_loss: 9.1294 - val_mae: 9.6294\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.9468 - mae: 10.4320 - val_loss: 9.3948 - val_mae: 9.8765\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.9942 - mae: 10.4783 - val_loss: 9.3573 - val_mae: 9.8468\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.9032 - mae: 10.3880 - val_loss: 9.3559 - val_mae: 9.8436\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.8875 - mae: 10.3721 - val_loss: 9.6152 - val_mae: 10.1020\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.9043 - mae: 10.3883 - val_loss: 9.2394 - val_mae: 9.7361\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.8392 - mae: 10.3277 - val_loss: 9.4437 - val_mae: 9.9117\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.8363 - mae: 10.3233 - val_loss: 9.5727 - val_mae: 10.0593\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.7880 - mae: 10.2769 - val_loss: 9.1449 - val_mae: 9.6448\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.7431 - mae: 10.2266 - val_loss: 9.9786 - val_mae: 10.4569\n",
      "Epoch 82/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.8338 - mae: 10.3161 - val_loss: 9.0331 - val_mae: 9.5330\n",
      "Epoch 83/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.7415 - mae: 10.2242 - val_loss: 9.4262 - val_mae: 9.9020\n",
      "Epoch 84/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.8816 - mae: 10.3689 - val_loss: 10.2742 - val_mae: 10.7637\n",
      "Epoch 85/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.8767 - mae: 10.3582 - val_loss: 8.9502 - val_mae: 9.4477\n",
      "Epoch 86/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.6869 - mae: 10.1706 - val_loss: 9.9651 - val_mae: 10.4497\n",
      "Epoch 87/150\n",
      "3/3 [==============================] - 0s 50ms/step - loss: 9.6851 - mae: 10.1717 - val_loss: 9.1119 - val_mae: 9.6066\n",
      "Epoch 88/150\n",
      "3/3 [==============================] - 0s 50ms/step - loss: 9.6549 - mae: 10.1381 - val_loss: 9.4456 - val_mae: 9.9336\n",
      "Epoch 89/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.6435 - mae: 10.1327 - val_loss: 9.9338 - val_mae: 10.4090\n",
      "Epoch 90/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.6112 - mae: 10.0951 - val_loss: 8.8598 - val_mae: 9.3455\n",
      "Epoch 91/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 9.7010 - mae: 10.1843 - val_loss: 9.6448 - val_mae: 10.1189\n",
      "Epoch 92/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.6848 - mae: 10.1687 - val_loss: 9.8088 - val_mae: 10.2934\n",
      "Epoch 93/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.6601 - mae: 10.1492 - val_loss: 9.8353 - val_mae: 10.3038\n",
      "Epoch 94/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.6192 - mae: 10.1062 - val_loss: 9.1673 - val_mae: 9.6613\n",
      "Epoch 95/150\n",
      "3/3 [==============================] - 0s 40ms/step - loss: 9.4322 - mae: 9.9168 - val_loss: 9.5003 - val_mae: 9.9850\n",
      "Epoch 96/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.4679 - mae: 9.9466 - val_loss: 8.9575 - val_mae: 9.4427\n",
      "Epoch 97/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.7831 - mae: 10.2667 - val_loss: 11.1587 - val_mae: 11.6538\n",
      "Epoch 98/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.9621 - mae: 10.4524 - val_loss: 9.1410 - val_mae: 9.6257\n",
      "Epoch 99/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.3976 - mae: 9.8799 - val_loss: 8.7663 - val_mae: 9.2641\n",
      "Epoch 100/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.5107 - mae: 9.9905 - val_loss: 10.6524 - val_mae: 11.1458\n",
      "Epoch 101/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.6619 - mae: 10.1515 - val_loss: 9.1855 - val_mae: 9.6695\n",
      "Epoch 102/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.3675 - mae: 9.8501 - val_loss: 8.8134 - val_mae: 9.3077\n",
      "Epoch 103/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.5808 - mae: 10.0665 - val_loss: 12.7288 - val_mae: 13.2288\n",
      "Epoch 104/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 10.4069 - mae: 10.8870 - val_loss: 9.0742 - val_mae: 9.5516\n",
      "Epoch 105/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.4448 - mae: 9.9279 - val_loss: 8.9471 - val_mae: 9.4217\n",
      "Epoch 106/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.3064 - mae: 10.7921 - val_loss: 10.7868 - val_mae: 11.2813\n",
      "Epoch 107/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.9994 - mae: 10.4846 - val_loss: 10.0947 - val_mae: 10.5721\n",
      "Epoch 108/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.4753 - mae: 9.9589 - val_loss: 8.7715 - val_mae: 9.2613\n",
      "Epoch 109/150\n",
      "3/3 [==============================] - 0s 38ms/step - loss: 9.3422 - mae: 9.8264 - val_loss: 11.5186 - val_mae: 12.0144\n",
      "Epoch 110/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 9.8115 - mae: 10.2994 - val_loss: 9.1041 - val_mae: 9.5763\n",
      "Epoch 111/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.2049 - mae: 9.6859 - val_loss: 9.0320 - val_mae: 9.5119\n",
      "Epoch 112/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.2201 - mae: 9.7072 - val_loss: 9.3610 - val_mae: 9.8487\n",
      "Epoch 113/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 9.2108 - mae: 9.6951 - val_loss: 9.0508 - val_mae: 9.5265\n",
      "Epoch 114/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.1782 - mae: 9.6533 - val_loss: 9.5731 - val_mae: 10.0689\n",
      "Epoch 115/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.6212 - mae: 10.1054 - val_loss: 10.7863 - val_mae: 11.2816\n",
      "Epoch 116/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 9.4296 - mae: 9.9155 - val_loss: 8.8505 - val_mae: 9.3447\n",
      "Epoch 117/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.2306 - mae: 9.7069 - val_loss: 8.9873 - val_mae: 9.4679\n",
      "Epoch 118/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 9.4370 - mae: 9.9053 - val_loss: 11.6125 - val_mae: 12.1107\n",
      "Epoch 119/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.7576 - mae: 10.2431 - val_loss: 8.9058 - val_mae: 9.3894\n",
      "Epoch 1/150\n",
      "3/3 [==============================] - 4s 406ms/step - loss: 78.4508 - mae: 78.9494 - val_loss: 24.7340 - val_mae: 25.2335\n",
      "Epoch 2/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 23.0540 - mae: 23.5507 - val_loss: 11.8354 - val_mae: 12.3291\n",
      "Epoch 3/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 16.3473 - mae: 16.8429 - val_loss: 18.2200 - val_mae: 18.7070\n",
      "Epoch 4/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 16.4776 - mae: 16.9704 - val_loss: 13.0498 - val_mae: 13.5430\n",
      "Epoch 5/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 17.0758 - mae: 17.5711 - val_loss: 11.4249 - val_mae: 11.9168\n",
      "Epoch 6/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 13.7001 - mae: 14.1936 - val_loss: 11.3659 - val_mae: 11.8560\n",
      "Epoch 7/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 13.6917 - mae: 14.1798 - val_loss: 11.3162 - val_mae: 11.8115\n",
      "Epoch 8/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 13.3482 - mae: 13.8393 - val_loss: 11.3508 - val_mae: 11.8466\n",
      "Epoch 9/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.9378 - mae: 13.4350 - val_loss: 11.1785 - val_mae: 11.6655\n",
      "Epoch 10/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.1534 - mae: 12.6416 - val_loss: 11.1026 - val_mae: 11.5930\n",
      "Epoch 11/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 12.1294 - mae: 12.6128 - val_loss: 11.0795 - val_mae: 11.5775\n",
      "Epoch 12/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.7709 - mae: 12.2580 - val_loss: 11.0927 - val_mae: 11.5764\n",
      "Epoch 13/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.7950 - mae: 12.2705 - val_loss: 10.9231 - val_mae: 11.4186\n",
      "Epoch 14/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.3679 - mae: 11.8555 - val_loss: 10.9189 - val_mae: 11.4052\n",
      "Epoch 15/150\n",
      "3/3 [==============================] - 0s 28ms/step - loss: 11.3470 - mae: 11.8252 - val_loss: 10.7952 - val_mae: 11.2901\n",
      "Epoch 16/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.9914 - mae: 11.4784 - val_loss: 10.9230 - val_mae: 11.4108\n",
      "Epoch 17/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.1113 - mae: 11.5854 - val_loss: 10.7341 - val_mae: 11.2294\n",
      "Epoch 18/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.9009 - mae: 11.3843 - val_loss: 11.3251 - val_mae: 11.8057\n",
      "Epoch 19/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.3738 - mae: 11.8648 - val_loss: 11.3004 - val_mae: 11.7934\n",
      "Epoch 20/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.5134 - mae: 12.0005 - val_loss: 12.4619 - val_mae: 12.9442\n",
      "Epoch 21/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 11.8000 - mae: 12.2626 - val_loss: 11.6779 - val_mae: 12.1617\n",
      "Epoch 22/150\n",
      "3/3 [==============================] - 0s 65ms/step - loss: 12.0800 - mae: 12.5743 - val_loss: 13.2820 - val_mae: 13.7793\n",
      "Epoch 23/150\n",
      "3/3 [==============================] - 0s 68ms/step - loss: 12.4198 - mae: 12.9031 - val_loss: 12.0048 - val_mae: 12.4878\n",
      "Epoch 24/150\n",
      "3/3 [==============================] - 0s 77ms/step - loss: 12.3185 - mae: 12.8097 - val_loss: 13.3518 - val_mae: 13.8441\n",
      "Epoch 25/150\n",
      "3/3 [==============================] - 0s 45ms/step - loss: 12.5634 - mae: 13.0435 - val_loss: 12.4983 - val_mae: 12.9874\n",
      "Epoch 26/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 13.1746 - mae: 13.6690 - val_loss: 12.0936 - val_mae: 12.5770\n",
      "Epoch 27/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 12.2264 - mae: 12.7195 - val_loss: 10.7941 - val_mae: 11.2717\n",
      "Epoch 28/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 12.8450 - mae: 13.3410 - val_loss: 11.0628 - val_mae: 11.5208\n",
      "Epoch 29/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 11.2785 - mae: 11.7683 - val_loss: 10.5249 - val_mae: 11.0185\n",
      "Epoch 30/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.8346 - mae: 12.3244 - val_loss: 11.0705 - val_mae: 11.5404\n",
      "Epoch 31/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.0383 - mae: 11.5255 - val_loss: 10.5381 - val_mae: 11.0257\n",
      "Epoch 32/150\n",
      "3/3 [==============================] - 0s 36ms/step - loss: 11.6047 - mae: 12.0976 - val_loss: 11.1108 - val_mae: 11.5939\n",
      "Epoch 33/150\n",
      "3/3 [==============================] - 0s 27ms/step - loss: 10.7659 - mae: 11.2506 - val_loss: 10.5693 - val_mae: 11.0625\n",
      "Epoch 34/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.1068 - mae: 11.5992 - val_loss: 11.0288 - val_mae: 11.5099\n",
      "Epoch 35/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.5133 - mae: 11.0000 - val_loss: 10.5600 - val_mae: 11.0550\n",
      "Epoch 36/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.5036 - mae: 10.9911 - val_loss: 11.2977 - val_mae: 11.7867\n",
      "Epoch 37/150\n",
      "3/3 [==============================] - 0s 37ms/step - loss: 10.4634 - mae: 10.9502 - val_loss: 10.4917 - val_mae: 10.9741\n",
      "Epoch 38/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.4689 - mae: 10.9544 - val_loss: 11.6027 - val_mae: 12.0901\n",
      "Epoch 39/150\n",
      "3/3 [==============================] - 0s 30ms/step - loss: 10.4444 - mae: 10.9161 - val_loss: 10.5971 - val_mae: 11.0874\n",
      "Epoch 40/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 10.5669 - mae: 11.0524 - val_loss: 11.3869 - val_mae: 11.8769\n",
      "Epoch 41/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.6954 - mae: 11.1802 - val_loss: 11.9781 - val_mae: 12.4682\n",
      "Epoch 42/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.5759 - mae: 12.0622 - val_loss: 12.9566 - val_mae: 13.4524\n",
      "Epoch 43/150\n",
      "3/3 [==============================] - 0s 40ms/step - loss: 11.7678 - mae: 12.2464 - val_loss: 11.0588 - val_mae: 11.5339\n",
      "Epoch 44/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 11.5752 - mae: 12.0588 - val_loss: 12.0929 - val_mae: 12.5880\n",
      "Epoch 45/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 11.3109 - mae: 11.8015 - val_loss: 10.9994 - val_mae: 11.4725\n",
      "Epoch 46/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.7808 - mae: 12.2733 - val_loss: 11.6792 - val_mae: 12.1603\n",
      "Epoch 47/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 11.5247 - mae: 12.0141 - val_loss: 10.8288 - val_mae: 11.3192\n",
      "Epoch 48/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 12.0027 - mae: 12.4940 - val_loss: 11.0750 - val_mae: 11.5633\n",
      "Epoch 49/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 10.5614 - mae: 11.0403 - val_loss: 10.4517 - val_mae: 10.9359\n",
      "Epoch 50/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 10.3530 - mae: 10.8423 - val_loss: 11.2912 - val_mae: 11.7819\n",
      "Epoch 51/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 10.2036 - mae: 10.6909 - val_loss: 10.3963 - val_mae: 10.8760\n",
      "Epoch 52/150\n",
      "3/3 [==============================] - 0s 29ms/step - loss: 10.0526 - mae: 10.5202 - val_loss: 11.6539 - val_mae: 12.1385\n",
      "Epoch 53/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 10.1461 - mae: 10.6168 - val_loss: 10.9473 - val_mae: 11.4281\n",
      "Epoch 54/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 10.1149 - mae: 10.5852 - val_loss: 12.5886 - val_mae: 13.0874\n",
      "Epoch 55/150\n",
      "3/3 [==============================] - 0s 35ms/step - loss: 11.0647 - mae: 11.5546 - val_loss: 11.8696 - val_mae: 12.3623\n",
      "Epoch 56/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 11.2770 - mae: 11.7699 - val_loss: 13.0028 - val_mae: 13.4964\n",
      "Epoch 57/150\n",
      "3/3 [==============================] - 0s 39ms/step - loss: 11.6851 - mae: 12.1759 - val_loss: 11.2968 - val_mae: 11.7827\n",
      "Epoch 58/150\n",
      "3/3 [==============================] - 0s 104ms/step - loss: 11.4720 - mae: 11.9613 - val_loss: 11.9861 - val_mae: 12.4855\n",
      "Epoch 59/150\n",
      "3/3 [==============================] - 0s 87ms/step - loss: 10.8775 - mae: 11.3670 - val_loss: 10.6103 - val_mae: 11.0998\n",
      "Epoch 60/150\n",
      "3/3 [==============================] - 0s 75ms/step - loss: 11.3839 - mae: 11.8717 - val_loss: 11.0947 - val_mae: 11.5829\n",
      "Epoch 61/150\n",
      "3/3 [==============================] - 0s 92ms/step - loss: 10.2692 - mae: 10.7499 - val_loss: 10.3767 - val_mae: 10.8666\n",
      "Epoch 62/150\n",
      "3/3 [==============================] - 0s 99ms/step - loss: 10.5329 - mae: 11.0212 - val_loss: 10.8376 - val_mae: 11.3200\n",
      "Epoch 63/150\n",
      "3/3 [==============================] - 0s 91ms/step - loss: 9.7229 - mae: 10.2081 - val_loss: 10.6294 - val_mae: 11.1206\n",
      "Epoch 64/150\n",
      "3/3 [==============================] - 0s 64ms/step - loss: 9.7202 - mae: 10.1948 - val_loss: 11.1508 - val_mae: 11.6408\n",
      "Epoch 65/150\n",
      "3/3 [==============================] - 0s 96ms/step - loss: 9.7546 - mae: 10.2271 - val_loss: 10.6447 - val_mae: 11.1270\n",
      "Epoch 66/150\n",
      "3/3 [==============================] - 0s 85ms/step - loss: 9.7434 - mae: 10.2032 - val_loss: 11.4068 - val_mae: 11.8929\n",
      "Epoch 67/150\n",
      "3/3 [==============================] - 0s 97ms/step - loss: 9.8440 - mae: 10.3159 - val_loss: 11.1597 - val_mae: 11.6528\n",
      "Epoch 68/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 9.9342 - mae: 10.4093 - val_loss: 11.0261 - val_mae: 11.5158\n",
      "Epoch 69/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.6163 - mae: 10.0973 - val_loss: 10.4751 - val_mae: 10.9529\n",
      "Epoch 70/150\n",
      "3/3 [==============================] - 0s 46ms/step - loss: 9.3290 - mae: 9.8051 - val_loss: 10.6074 - val_mae: 11.0978\n",
      "Epoch 71/150\n",
      "3/3 [==============================] - 0s 75ms/step - loss: 9.2846 - mae: 9.7513 - val_loss: 10.4429 - val_mae: 10.9349\n",
      "Epoch 72/150\n",
      "3/3 [==============================] - 0s 76ms/step - loss: 9.2511 - mae: 9.7168 - val_loss: 10.8074 - val_mae: 11.2859\n",
      "Epoch 73/150\n",
      "3/3 [==============================] - 0s 48ms/step - loss: 9.3016 - mae: 9.7734 - val_loss: 10.4843 - val_mae: 10.9770\n",
      "Epoch 74/150\n",
      "3/3 [==============================] - 0s 59ms/step - loss: 9.3349 - mae: 9.8156 - val_loss: 10.5041 - val_mae: 10.9867\n",
      "Epoch 75/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.3271 - mae: 9.8042 - val_loss: 10.8184 - val_mae: 11.2932\n",
      "Epoch 76/150\n",
      "3/3 [==============================] - 0s 34ms/step - loss: 9.3510 - mae: 9.8174 - val_loss: 10.6439 - val_mae: 11.1244\n",
      "Epoch 77/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.2443 - mae: 9.7149 - val_loss: 10.7634 - val_mae: 11.2451\n",
      "Epoch 78/150\n",
      "3/3 [==============================] - 0s 33ms/step - loss: 9.2013 - mae: 9.6730 - val_loss: 10.5427 - val_mae: 11.0377\n",
      "Epoch 79/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.0701 - mae: 9.5342 - val_loss: 10.5666 - val_mae: 11.0504\n",
      "Epoch 80/150\n",
      "3/3 [==============================] - 0s 32ms/step - loss: 9.0559 - mae: 9.5158 - val_loss: 10.5362 - val_mae: 11.0351\n",
      "Epoch 81/150\n",
      "3/3 [==============================] - 0s 31ms/step - loss: 9.0533 - mae: 9.5227 - val_loss: 11.0527 - val_mae: 11.5374\n"
     ]
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "def plot_series(time, series, format=\"-\", start=0, end=None):\n",
    "    plt.plot(time[start:end], series[start:end], format)\n",
    "    plt.xlabel(\"Time\")\n",
    "    plt.ylabel(\"Value\")\n",
    "    plt.grid(True)\n",
    "\n",
    "def compute_metrics(true_series, forecast):\n",
    "    \n",
    "    mse = tf.keras.metrics.mean_squared_error(true_series, forecast).numpy()\n",
    "    mae = tf.keras.metrics.mean_absolute_error(true_series, forecast).numpy()\n",
    "\n",
    "    return mse, mae\n",
    "\n",
    "def model_forecast(model, series, window_size):\n",
    "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
    "    ds = ds.window(window_size, shift=1, drop_remainder=True)\n",
    "    ds = ds.flat_map(lambda w: w.batch(window_size))\n",
    "    ds = ds.batch(32).prefetch(1)\n",
    "    forecast = model.predict(ds)\n",
    "    return forecast"
   ],
   "metadata": {
    "id": "NFj3EaRnW7TE",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 10,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "rnn_forecast = model_forecast(model, temp_series, window_size).squeeze()\n",
    "rnn_forecast = rnn_forecast[split_time - window_size:-1]\n",
    "\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "plot_series(time_valid, series_valid)\n",
    "plot_series(time_valid, rnn_forecast)"
   ],
   "metadata": {
    "id": "6hp8dLM6XcgI",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 405
    },
    "outputId": "1542e5b5-6269-41cf-b4cf-efa2c191f683",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 11,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "11/11 [==============================] - 1s 10ms/step\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "mse, mae = compute_metrics(series_valid, rnn_forecast)\n",
    "print(f\"mse: {mse}, mae: {mae} for forecast\")"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "T7FD1oqgGB5t",
    "outputId": "97577b4a-045f-4ec4-85ea-50eda99917ae",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": 13,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "mse: 535.8865966796875, mae: 15.045037269592285 for forecast\n"
     ]
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [],
   "metadata": {
    "id": "cWpHNuW0Khvu",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": null,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [],
   "metadata": {
    "id": "el_d3SnBKhxe",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": null,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [],
   "metadata": {
    "id": "C9-LttffKhzd",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": null,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [],
   "metadata": {
    "id": "6kK09ACMKnsl",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "execution_count": null,
   "outputs": []
  }
 ]
}