multivariate_LSTM_scaled_grid_search.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scikeras in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (0.11.0)\n",
      "Requirement already satisfied: packaging>=0.21 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikeras) (23.1)\n",
      "Requirement already satisfied: scikit-learn>=1.0.0 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikeras) (1.3.0)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem<0.32,>=0.23.1 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikeras) (0.31.0)\n",
      "Requirement already satisfied: numpy>=1.17.3 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-learn>=1.0.0->scikeras) (1.24.2)\n",
      "Requirement already satisfied: scipy>=1.5.0 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-learn>=1.0.0->scikeras) (1.11.1)\n",
      "Requirement already satisfied: joblib>=1.1.1 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-learn>=1.0.0->scikeras) (1.3.1)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\user\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-learn>=1.0.0->scikeras) (3.2.0)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import os\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "import keras\n",
    "!pip install scikeras\n",
    "import scikeras\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from scikeras.wrappers import KerasClassifier\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### Import CSV, clean and transform data-set\n",
    "\n",
    "In this section, the dataset is imported, cleaned and transformed into a fitting format. Columns with missing values are left out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Food_Price</th>\n",
       "      <th>Oil_price (USD per gallon)</th>\n",
       "      <th>Exchange_rate ($/€)</th>\n",
       "      <th>Air_Temperature (°C)</th>\n",
       "      <th>Precipitation (l/m²)</th>\n",
       "      <th>PP_food_products</th>\n",
       "      <th>PP_bakery_farinaceous_products</th>\n",
       "      <th>PP_prepared_feed_farm_animals</th>\n",
       "      <th>PP_pesticides_agrochemicals_products</th>\n",
       "      <th>PP_Grain_Starch</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1996-01-01</th>\n",
       "      <td>74.9</td>\n",
       "      <td>16.52</td>\n",
       "      <td>1.2918</td>\n",
       "      <td>-2.82</td>\n",
       "      <td>7.0</td>\n",
       "      <td>78.6</td>\n",
       "      <td>70.2</td>\n",
       "      <td>62.9</td>\n",
       "      <td>86.1</td>\n",
       "      <td>87.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996-02-01</th>\n",
       "      <td>75.5</td>\n",
       "      <td>17.76</td>\n",
       "      <td>1.2886</td>\n",
       "      <td>-2.18</td>\n",
       "      <td>48.6</td>\n",
       "      <td>78.4</td>\n",
       "      <td>70.1</td>\n",
       "      <td>63.2</td>\n",
       "      <td>85.4</td>\n",
       "      <td>87.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996-03-01</th>\n",
       "      <td>76.1</td>\n",
       "      <td>19.41</td>\n",
       "      <td>1.2813</td>\n",
       "      <td>1.00</td>\n",
       "      <td>23.3</td>\n",
       "      <td>78.5</td>\n",
       "      <td>70.9</td>\n",
       "      <td>62.8</td>\n",
       "      <td>85.4</td>\n",
       "      <td>87.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996-04-01</th>\n",
       "      <td>76.4</td>\n",
       "      <td>19.02</td>\n",
       "      <td>1.2639</td>\n",
       "      <td>8.40</td>\n",
       "      <td>23.1</td>\n",
       "      <td>78.6</td>\n",
       "      <td>70.8</td>\n",
       "      <td>63.3</td>\n",
       "      <td>85.4</td>\n",
       "      <td>87.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996-05-01</th>\n",
       "      <td>76.2</td>\n",
       "      <td>17.80</td>\n",
       "      <td>1.2467</td>\n",
       "      <td>11.15</td>\n",
       "      <td>91.4</td>\n",
       "      <td>79</td>\n",
       "      <td>70.8</td>\n",
       "      <td>65.7</td>\n",
       "      <td>84.9</td>\n",
       "      <td>88.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-08-01</th>\n",
       "      <td>114.2</td>\n",
       "      <td>72.99</td>\n",
       "      <td>1.1772</td>\n",
       "      <td>16.40</td>\n",
       "      <td>102.5</td>\n",
       "      <td>111.2</td>\n",
       "      <td>114.4</td>\n",
       "      <td>118.1</td>\n",
       "      <td>107.6</td>\n",
       "      <td>107.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-09-01</th>\n",
       "      <td>114.2</td>\n",
       "      <td>78.52</td>\n",
       "      <td>1.1770</td>\n",
       "      <td>15.16</td>\n",
       "      <td>35.3</td>\n",
       "      <td>111.3</td>\n",
       "      <td>115.2</td>\n",
       "      <td>121.1</td>\n",
       "      <td>107.9</td>\n",
       "      <td>108.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-01</th>\n",
       "      <td>114.2</td>\n",
       "      <td>84.38</td>\n",
       "      <td>1.1601</td>\n",
       "      <td>9.63</td>\n",
       "      <td>44.8</td>\n",
       "      <td>112.1</td>\n",
       "      <td>115.3</td>\n",
       "      <td>124.1</td>\n",
       "      <td>110.3</td>\n",
       "      <td>112.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-11-01</th>\n",
       "      <td>114.8</td>\n",
       "      <td>70.57</td>\n",
       "      <td>1.1414</td>\n",
       "      <td>4.90</td>\n",
       "      <td>49.7</td>\n",
       "      <td>113.5</td>\n",
       "      <td>115.6</td>\n",
       "      <td>129.6</td>\n",
       "      <td>110.7</td>\n",
       "      <td>115.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-12-01</th>\n",
       "      <td>116.0</td>\n",
       "      <td>77.78</td>\n",
       "      <td>1.1304</td>\n",
       "      <td>2.58</td>\n",
       "      <td>61.4</td>\n",
       "      <td>114.9</td>\n",
       "      <td>116.7</td>\n",
       "      <td>133.6</td>\n",
       "      <td>110.7</td>\n",
       "      <td>117.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>312 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Food_Price  Oil_price (USD per gallon)  Exchange_rate ($/€)  \\\n",
       "Date                                                                      \n",
       "1996-01-01        74.9                       16.52               1.2918   \n",
       "1996-02-01        75.5                       17.76               1.2886   \n",
       "1996-03-01        76.1                       19.41               1.2813   \n",
       "1996-04-01        76.4                       19.02               1.2639   \n",
       "1996-05-01        76.2                       17.80               1.2467   \n",
       "...                ...                         ...                  ...   \n",
       "2021-08-01       114.2                       72.99               1.1772   \n",
       "2021-09-01       114.2                       78.52               1.1770   \n",
       "2021-10-01       114.2                       84.38               1.1601   \n",
       "2021-11-01       114.8                       70.57               1.1414   \n",
       "2021-12-01       116.0                       77.78               1.1304   \n",
       "\n",
       "            Air_Temperature (°C)  Precipitation (l/m²) PP_food_products  \\\n",
       "Date                                                                      \n",
       "1996-01-01                 -2.82                   7.0             78.6   \n",
       "1996-02-01                 -2.18                  48.6             78.4   \n",
       "1996-03-01                  1.00                  23.3             78.5   \n",
       "1996-04-01                  8.40                  23.1             78.6   \n",
       "1996-05-01                 11.15                  91.4               79   \n",
       "...                          ...                   ...              ...   \n",
       "2021-08-01                 16.40                 102.5            111.2   \n",
       "2021-09-01                 15.16                  35.3            111.3   \n",
       "2021-10-01                  9.63                  44.8            112.1   \n",
       "2021-11-01                  4.90                  49.7            113.5   \n",
       "2021-12-01                  2.58                  61.4            114.9   \n",
       "\n",
       "           PP_bakery_farinaceous_products PP_prepared_feed_farm_animals  \\\n",
       "Date                                                                      \n",
       "1996-01-01                           70.2                          62.9   \n",
       "1996-02-01                           70.1                          63.2   \n",
       "1996-03-01                           70.9                          62.8   \n",
       "1996-04-01                           70.8                          63.3   \n",
       "1996-05-01                           70.8                          65.7   \n",
       "...                                   ...                           ...   \n",
       "2021-08-01                          114.4                         118.1   \n",
       "2021-09-01                          115.2                         121.1   \n",
       "2021-10-01                          115.3                         124.1   \n",
       "2021-11-01                          115.6                         129.6   \n",
       "2021-12-01                          116.7                         133.6   \n",
       "\n",
       "           PP_pesticides_agrochemicals_products PP_Grain_Starch  \n",
       "Date                                                             \n",
       "1996-01-01                                 86.1            87.3  \n",
       "1996-02-01                                 85.4            87.4  \n",
       "1996-03-01                                 85.4            87.7  \n",
       "1996-04-01                                 85.4            87.8  \n",
       "1996-05-01                                 84.9            88.8  \n",
       "...                                         ...             ...  \n",
       "2021-08-01                                107.6           107.2  \n",
       "2021-09-01                                107.9           108.2  \n",
       "2021-10-01                                110.3           112.8  \n",
       "2021-11-01                                110.7           115.2  \n",
       "2021-12-01                                110.7           117.1  \n",
       "\n",
       "[312 rows x 10 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df = pd.read_csv(\"Datasets\\data_merging_new.csv\")\n",
    "df = df.drop(columns=[\"Unnamed: 0\"])   #Delete redundant Columns\n",
    "df = df.drop(columns=[\"Unnamed: 0.1\"])   #Delete redundant Columns\n",
    "df.index = pd.to_datetime(df['Date'], format = '%Y-%m') #Dates as Index\n",
    "df = df[df.Date <= \"2021-12-01\"]  #Limit Timeframe\n",
    "df = df.drop(columns=[\"Date\"], axis = 1)\n",
    "df = df.drop(columns=[\"€/kWh\"], axis = 1)\n",
    "df = df.drop(columns=[\"PP_other_food_products\"], axis = 1)\n",
    "df\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Standardization\n",
    "\n",
    "In this section, the dataset is standardized by normalization. The mean value of each column is substracted from each individual value and divided by the columns standard deviation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Food_Price</th>\n",
       "      <th>Oil_price (USD per gallon)</th>\n",
       "      <th>Exchange_rate ($/€)</th>\n",
       "      <th>Air_Temperature (°C)</th>\n",
       "      <th>Precipitation (l/m²)</th>\n",
       "      <th>PP_food_products</th>\n",
       "      <th>PP_bakery_farinaceous_products</th>\n",
       "      <th>PP_prepared_feed_farm_animals</th>\n",
       "      <th>PP_pesticides_agrochemicals_products</th>\n",
       "      <th>PP_Grain_Starch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.255570</td>\n",
       "      <td>-1.288947</td>\n",
       "      <td>0.632412</td>\n",
       "      <td>-1.876104</td>\n",
       "      <td>-2.075057</td>\n",
       "      <td>-1.238273</td>\n",
       "      <td>-1.276179</td>\n",
       "      <td>-1.022989</td>\n",
       "      <td>-0.831987</td>\n",
       "      <td>-0.671747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.205758</td>\n",
       "      <td>-1.249765</td>\n",
       "      <td>0.611276</td>\n",
       "      <td>-1.777546</td>\n",
       "      <td>-0.600640</td>\n",
       "      <td>-1.256749</td>\n",
       "      <td>-1.283468</td>\n",
       "      <td>-1.008166</td>\n",
       "      <td>-0.921737</td>\n",
       "      <td>-0.662489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.155946</td>\n",
       "      <td>-1.197627</td>\n",
       "      <td>0.563061</td>\n",
       "      <td>-1.287835</td>\n",
       "      <td>-1.497341</td>\n",
       "      <td>-1.247511</td>\n",
       "      <td>-1.225154</td>\n",
       "      <td>-1.027930</td>\n",
       "      <td>-0.921737</td>\n",
       "      <td>-0.634715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.131040</td>\n",
       "      <td>-1.209951</td>\n",
       "      <td>0.448137</td>\n",
       "      <td>-0.148257</td>\n",
       "      <td>-1.504430</td>\n",
       "      <td>-1.238273</td>\n",
       "      <td>-1.232444</td>\n",
       "      <td>-1.003225</td>\n",
       "      <td>-0.921737</td>\n",
       "      <td>-0.625457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.147644</td>\n",
       "      <td>-1.248501</td>\n",
       "      <td>0.334534</td>\n",
       "      <td>0.275235</td>\n",
       "      <td>0.916308</td>\n",
       "      <td>-1.201319</td>\n",
       "      <td>-1.232444</td>\n",
       "      <td>-0.884640</td>\n",
       "      <td>-0.985843</td>\n",
       "      <td>-0.532876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>307</th>\n",
       "      <td>2.007114</td>\n",
       "      <td>0.495420</td>\n",
       "      <td>-0.124501</td>\n",
       "      <td>1.083720</td>\n",
       "      <td>1.309722</td>\n",
       "      <td>1.773449</td>\n",
       "      <td>1.945668</td>\n",
       "      <td>1.704475</td>\n",
       "      <td>1.924593</td>\n",
       "      <td>1.170617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>308</th>\n",
       "      <td>2.007114</td>\n",
       "      <td>0.670159</td>\n",
       "      <td>-0.125822</td>\n",
       "      <td>0.892763</td>\n",
       "      <td>-1.072029</td>\n",
       "      <td>1.782687</td>\n",
       "      <td>2.003982</td>\n",
       "      <td>1.852707</td>\n",
       "      <td>1.963056</td>\n",
       "      <td>1.263198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309</th>\n",
       "      <td>2.007114</td>\n",
       "      <td>0.855327</td>\n",
       "      <td>-0.237443</td>\n",
       "      <td>0.041160</td>\n",
       "      <td>-0.735323</td>\n",
       "      <td>1.856595</td>\n",
       "      <td>2.011272</td>\n",
       "      <td>2.000939</td>\n",
       "      <td>2.270768</td>\n",
       "      <td>1.689072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310</th>\n",
       "      <td>2.056926</td>\n",
       "      <td>0.418951</td>\n",
       "      <td>-0.360954</td>\n",
       "      <td>-0.687246</td>\n",
       "      <td>-0.561653</td>\n",
       "      <td>1.985932</td>\n",
       "      <td>2.033139</td>\n",
       "      <td>2.272697</td>\n",
       "      <td>2.322053</td>\n",
       "      <td>1.911266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>311</th>\n",
       "      <td>2.156550</td>\n",
       "      <td>0.646777</td>\n",
       "      <td>-0.433607</td>\n",
       "      <td>-1.044520</td>\n",
       "      <td>-0.146974</td>\n",
       "      <td>2.115270</td>\n",
       "      <td>2.113321</td>\n",
       "      <td>2.470339</td>\n",
       "      <td>2.322053</td>\n",
       "      <td>2.087170</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>312 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Food_Price  Oil_price (USD per gallon)  Exchange_rate ($/€)  \\\n",
       "0     -1.255570                   -1.288947             0.632412   \n",
       "1     -1.205758                   -1.249765             0.611276   \n",
       "2     -1.155946                   -1.197627             0.563061   \n",
       "3     -1.131040                   -1.209951             0.448137   \n",
       "4     -1.147644                   -1.248501             0.334534   \n",
       "..          ...                         ...                  ...   \n",
       "307    2.007114                    0.495420            -0.124501   \n",
       "308    2.007114                    0.670159            -0.125822   \n",
       "309    2.007114                    0.855327            -0.237443   \n",
       "310    2.056926                    0.418951            -0.360954   \n",
       "311    2.156550                    0.646777            -0.433607   \n",
       "\n",
       "     Air_Temperature (°C)  Precipitation (l/m²)  PP_food_products  \\\n",
       "0               -1.876104             -2.075057         -1.238273   \n",
       "1               -1.777546             -0.600640         -1.256749   \n",
       "2               -1.287835             -1.497341         -1.247511   \n",
       "3               -0.148257             -1.504430         -1.238273   \n",
       "4                0.275235              0.916308         -1.201319   \n",
       "..                    ...                   ...               ...   \n",
       "307              1.083720              1.309722          1.773449   \n",
       "308              0.892763             -1.072029          1.782687   \n",
       "309              0.041160             -0.735323          1.856595   \n",
       "310             -0.687246             -0.561653          1.985932   \n",
       "311             -1.044520             -0.146974          2.115270   \n",
       "\n",
       "     PP_bakery_farinaceous_products  PP_prepared_feed_farm_animals  \\\n",
       "0                         -1.276179                      -1.022989   \n",
       "1                         -1.283468                      -1.008166   \n",
       "2                         -1.225154                      -1.027930   \n",
       "3                         -1.232444                      -1.003225   \n",
       "4                         -1.232444                      -0.884640   \n",
       "..                              ...                            ...   \n",
       "307                        1.945668                       1.704475   \n",
       "308                        2.003982                       1.852707   \n",
       "309                        2.011272                       2.000939   \n",
       "310                        2.033139                       2.272697   \n",
       "311                        2.113321                       2.470339   \n",
       "\n",
       "     PP_pesticides_agrochemicals_products  PP_Grain_Starch  \n",
       "0                               -0.831987        -0.671747  \n",
       "1                               -0.921737        -0.662489  \n",
       "2                               -0.921737        -0.634715  \n",
       "3                               -0.921737        -0.625457  \n",
       "4                               -0.985843        -0.532876  \n",
       "..                                    ...              ...  \n",
       "307                              1.924593         1.170617  \n",
       "308                              1.963056         1.263198  \n",
       "309                              2.270768         1.689072  \n",
       "310                              2.322053         1.911266  \n",
       "311                              2.322053         2.087170  \n",
       "\n",
       "[312 rows x 10 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ss = StandardScaler()\n",
    "df = pd.DataFrame(ss.fit_transform(df),columns = df.columns)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creation of Sliding-Window Arrays Function\n",
    "\n",
    "This section creates a function which transforms the original dataset into sliding windows. A matrix in the shape of (a,b,c) for the feature observations and a corresponding array for the y-labels (food-price) is created. By defining the window-size parameter, the observation window for the model training is specified. It defines how many previous periods of X-feature observations the model should take into account for predicting the food-price in the current period. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df_to_X_y(df, window_size=5): #set dataframe and window-size for defining range of previous periods \n",
    "    df_as_np = df.to_numpy()\n",
    "    X = [] # Matrix of feature observations\n",
    "    y = [] # array of labels\n",
    "\n",
    "    for i in range(len(df_as_np)-window_size): #iteration over each period dataset\n",
    "        row = [r for r in df_as_np[i:i+window_size]] #create array of x-feature observation periods in the size of \"window size\"\n",
    "        X.append(row)\n",
    "        label = df_as_np[i+window_size][0]\n",
    "        y.append(label)\n",
    "    return np.array(X), np.array(y)          #return X and y matrices."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Checking the format of the resulting data-set\n",
    "X-matrix shape is (a,b,c) whereas a is the total number of periods, b is the window size, and c is the total number of features. the y-array of labels has the shape of (a)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((303, 9, 10), (303,))"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "window = 9\n",
    "X, y = df_to_X_y(df, window)\n",
    "X.shape, y.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Creation of a Train-, Test- and Validation-Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((250, 9, 10), (250,), (25, 9, 10), (25,), (28, 9, 10), (28,))"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, y_train = X[:250], y[:250]\n",
    "X_val, y_val = X[250:275], y[250:275]\n",
    "X_test, y_test = X[275:], y[275:]\n",
    "\n",
    "\n",
    "\n",
    "X_train = np.asarray(X_train).astype('float32')\n",
    "y_train = np.asarray(y_train).astype('float32')\n",
    "X_val = np.asarray(X_val).astype('float32')\n",
    "y_val = np.asarray(y_val).astype('float32')\n",
    "X_test = np.asarray(X_test).astype('float32')\n",
    "y_test = np.asarray(y_test).astype('float32')\n",
    "\n",
    "X_train.shape, y_train.shape, X_val.shape, y_val.shape, X_test.shape, y_test.shape  #Check shape of sets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model-Definition & Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configuring the LSTM-Model with Tensorflow\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed = 42\n",
    "\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import * \n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "from tensorflow.keras.losses import MeanSquaredError\n",
    "from tensorflow.keras.metrics import RootMeanSquaredError\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.optimizers import RMSprop, Adagrad\n",
    "\n",
    "#definiere Model als Function für GridSearch\n",
    "\n",
    "def define_model(drop=0.1,optimizer =\"Adam\"):\n",
    "    model_n = Sequential()\n",
    "    model_n.add(LSTM(64, activation='relu', input_shape=(window, 10), return_sequences=True))\n",
    "    model_n.add(LSTM(32, activation='relu', return_sequences=False))\n",
    "    model_n.add(Dropout(drop))\n",
    "    model_n.add(Dense(1))\n",
    "\n",
    "    model_n.compile(optimizer=optimizer, loss='mape', metrics=[\"mape\"])\n",
    "    \n",
    "    return model_n\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Executing Grid-Search\n",
    "\n",
    "In this section, a grid-search is performed. It iterates over a predefined selection of dropout-rates and optimizers in order to find the best-performing combination. As a metric it uses the mean-average-percentage-error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "f:\\Git\\Food-price-forecast - Kopie\\venv\\Lib\\site-packages\\scikeras\\wrappers.py:915: UserWarning: ``build_fn`` will be renamed to ``model`` in a future release, at which point use of ``build_fn`` will raise an Error instead.\n",
      "  X, y = self._initialize(X, y)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/300\n",
      "8/8 [==============================] - 2s 6ms/step - loss: 77.0046 - mape: 77.0046\n",
      "Epoch 2/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 47.6894 - mape: 47.6894\n",
      "Epoch 3/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 33.9574 - mape: 33.9574\n",
      "Epoch 4/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 25.0459 - mape: 25.0459\n",
      "Epoch 5/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 16.6181 - mape: 16.6181\n",
      "Epoch 6/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 15.2682 - mape: 15.2682\n",
      "Epoch 7/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 13.3591 - mape: 13.3591\n",
      "Epoch 8/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 14.8995 - mape: 14.8995\n",
      "Epoch 9/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 14.4340 - mape: 14.4340\n",
      "Epoch 10/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 17.0031 - mape: 17.0031\n",
      "Epoch 11/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 13.2584 - mape: 13.2584\n",
      "Epoch 12/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 14.2639 - mape: 14.2639\n",
      "Epoch 13/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 14.3475 - mape: 14.3475\n",
      "Epoch 14/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 13.0848 - mape: 13.0848\n",
      "Epoch 15/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 11.1057 - mape: 11.1057\n",
      "Epoch 16/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 10.0262 - mape: 10.0262\n",
      "Epoch 17/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 11.5216 - mape: 11.5216\n",
      "Epoch 18/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 10.9625 - mape: 10.9625\n",
      "Epoch 19/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.9669 - mape: 9.9669\n",
      "Epoch 20/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 10.2694 - mape: 10.2694\n",
      "Epoch 21/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 9.3677 - mape: 9.3677\n",
      "Epoch 22/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 11.9793 - mape: 11.9793\n",
      "Epoch 23/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 12.3033 - mape: 12.3033\n",
      "Epoch 24/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 12.6105 - mape: 12.6105\n",
      "Epoch 25/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 10.6819 - mape: 10.6819\n",
      "Epoch 26/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 10.1933 - mape: 10.1933\n",
      "Epoch 27/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.4369 - mape: 9.4369\n",
      "Epoch 28/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.0646 - mape: 9.0646\n",
      "Epoch 29/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.1871 - mape: 9.1871\n",
      "Epoch 30/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 10.0541 - mape: 10.0541\n",
      "Epoch 31/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.7587 - mape: 8.7587\n",
      "Epoch 32/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.0673 - mape: 9.0673\n",
      "Epoch 33/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.3174 - mape: 9.3174\n",
      "Epoch 34/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.2210 - mape: 9.2210\n",
      "Epoch 35/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.8668 - mape: 8.8668\n",
      "Epoch 36/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.9601 - mape: 8.9601\n",
      "Epoch 37/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.6284 - mape: 8.6284\n",
      "Epoch 38/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.3515 - mape: 8.3515\n",
      "Epoch 39/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.0325 - mape: 8.0325\n",
      "Epoch 40/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.6070 - mape: 9.6070\n",
      "Epoch 41/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.8840 - mape: 8.8840\n",
      "Epoch 42/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.1252 - mape: 9.1252\n",
      "Epoch 43/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.4940 - mape: 9.4940\n",
      "Epoch 44/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.7113 - mape: 7.7113\n",
      "Epoch 45/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.3127 - mape: 9.3127\n",
      "Epoch 46/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 9.0782 - mape: 9.0782\n",
      "Epoch 47/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 9.8688 - mape: 9.8688\n",
      "Epoch 48/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.4676 - mape: 8.4676\n",
      "Epoch 49/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.7793 - mape: 7.7793\n",
      "Epoch 50/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.7397 - mape: 8.7397\n",
      "Epoch 51/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.0134 - mape: 8.0134\n",
      "Epoch 52/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.8782 - mape: 7.8782\n",
      "Epoch 53/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.6370 - mape: 7.6370\n",
      "Epoch 54/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.0384 - mape: 8.0384\n",
      "Epoch 55/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.5584 - mape: 8.5584\n",
      "Epoch 56/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.0285 - mape: 8.0285\n",
      "Epoch 57/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.5752 - mape: 8.5752\n",
      "Epoch 58/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.8601 - mape: 7.8601\n",
      "Epoch 59/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.7874 - mape: 8.7874\n",
      "Epoch 60/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.8686 - mape: 7.8686\n",
      "Epoch 61/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.7254 - mape: 7.7254\n",
      "Epoch 62/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.7753 - mape: 8.7753\n",
      "Epoch 63/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.0717 - mape: 8.0717\n",
      "Epoch 64/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 9.3121 - mape: 9.3121\n",
      "Epoch 65/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.6511 - mape: 7.6511\n",
      "Epoch 66/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.1031 - mape: 8.1031\n",
      "Epoch 67/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.7283 - mape: 8.7283\n",
      "Epoch 68/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.9920 - mape: 7.9920\n",
      "Epoch 69/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.7905 - mape: 7.7905\n",
      "Epoch 70/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.5462 - mape: 7.5462\n",
      "Epoch 71/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.7837 - mape: 7.7837\n",
      "Epoch 72/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 7.9142 - mape: 7.9142\n",
      "Epoch 73/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.2453 - mape: 7.2453\n",
      "Epoch 74/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.0732 - mape: 8.0732\n",
      "Epoch 75/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.0577 - mape: 7.0577\n",
      "Epoch 76/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.8673 - mape: 7.8673\n",
      "Epoch 77/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.2532 - mape: 7.2532\n",
      "Epoch 78/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.9914 - mape: 6.9914\n",
      "Epoch 79/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.4468 - mape: 7.4468\n",
      "Epoch 80/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.5735 - mape: 6.5735\n",
      "Epoch 81/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.4609 - mape: 7.4609\n",
      "Epoch 82/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4990 - mape: 6.4990\n",
      "Epoch 83/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.8251 - mape: 6.8251\n",
      "Epoch 84/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9149 - mape: 5.9149\n",
      "Epoch 85/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9358 - mape: 5.9358\n",
      "Epoch 86/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.6162 - mape: 6.6162\n",
      "Epoch 87/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.8557 - mape: 6.8557\n",
      "Epoch 88/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.1117 - mape: 6.1117\n",
      "Epoch 89/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.6983 - mape: 6.6983\n",
      "Epoch 90/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.4358 - mape: 7.4358\n",
      "Epoch 91/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.3737 - mape: 7.3737\n",
      "Epoch 92/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.4851 - mape: 7.4851\n",
      "Epoch 93/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.4437 - mape: 8.4437\n",
      "Epoch 94/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 8.0114 - mape: 8.0114\n",
      "Epoch 95/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.7290 - mape: 6.7290\n",
      "Epoch 96/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 6.5192 - mape: 6.5192\n",
      "Epoch 97/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4488 - mape: 6.4488\n",
      "Epoch 98/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 7.4580 - mape: 7.4580\n",
      "Epoch 99/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.7833 - mape: 6.7833\n",
      "Epoch 100/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 9.2382 - mape: 9.2382\n",
      "Epoch 101/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 8.0278 - mape: 8.0278\n",
      "Epoch 102/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.5217 - mape: 6.5217\n",
      "Epoch 103/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.5878 - mape: 6.5878\n",
      "Epoch 104/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.1319 - mape: 7.1319\n",
      "Epoch 105/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.1561 - mape: 7.1561\n",
      "Epoch 106/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.2052 - mape: 7.2052\n",
      "Epoch 107/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4957 - mape: 6.4957\n",
      "Epoch 108/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.9896 - mape: 6.9896\n",
      "Epoch 109/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.9181 - mape: 7.9181\n",
      "Epoch 110/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.9702 - mape: 7.9702\n",
      "Epoch 111/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4543 - mape: 6.4543\n",
      "Epoch 112/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.2720 - mape: 6.2720\n",
      "Epoch 113/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7669 - mape: 5.7669\n",
      "Epoch 114/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6708 - mape: 5.6708\n",
      "Epoch 115/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.2802 - mape: 6.2802\n",
      "Epoch 116/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.9959 - mape: 6.9959\n",
      "Epoch 117/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.9700 - mape: 6.9700\n",
      "Epoch 118/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.7924 - mape: 6.7924\n",
      "Epoch 119/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.2745 - mape: 7.2745\n",
      "Epoch 120/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 6.8426 - mape: 6.8426\n",
      "Epoch 121/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9631 - mape: 5.9631\n",
      "Epoch 122/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.3503 - mape: 6.3503\n",
      "Epoch 123/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7221 - mape: 5.7221\n",
      "Epoch 124/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7408 - mape: 5.7408\n",
      "Epoch 125/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.1155 - mape: 7.1155\n",
      "Epoch 126/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.7228 - mape: 6.7228\n",
      "Epoch 127/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.1671 - mape: 7.1671\n",
      "Epoch 128/300\n",
      "8/8 [==============================] - 0s 26ms/step - loss: 6.0348 - mape: 6.0348\n",
      "Epoch 129/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.0290 - mape: 7.0290\n",
      "Epoch 130/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.6586 - mape: 5.6586\n",
      "Epoch 131/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6705 - mape: 5.6705\n",
      "Epoch 132/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.2244 - mape: 6.2244\n",
      "Epoch 133/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.8172 - mape: 5.8172\n",
      "Epoch 134/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.1180 - mape: 6.1180\n",
      "Epoch 135/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3682 - mape: 5.3682\n",
      "Epoch 136/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.1714 - mape: 6.1714\n",
      "Epoch 137/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.2645 - mape: 5.2645\n",
      "Epoch 138/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.3959 - mape: 6.3959\n",
      "Epoch 139/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.3826 - mape: 6.3826\n",
      "Epoch 140/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6590 - mape: 5.6590\n",
      "Epoch 141/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6256 - mape: 5.6256\n",
      "Epoch 142/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 6.8907 - mape: 6.8907\n",
      "Epoch 143/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7829 - mape: 5.7829\n",
      "Epoch 144/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.3688 - mape: 6.3688\n",
      "Epoch 145/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 7.3954 - mape: 7.3954\n",
      "Epoch 146/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4405 - mape: 6.4405\n",
      "Epoch 147/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4381 - mape: 6.4381\n",
      "Epoch 148/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.6272 - mape: 6.6272\n",
      "Epoch 149/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.2874 - mape: 6.2874\n",
      "Epoch 150/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.3326 - mape: 6.3326\n",
      "Epoch 151/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.5270 - mape: 6.5270\n",
      "Epoch 152/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.9380 - mape: 5.9380\n",
      "Epoch 153/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.4750 - mape: 5.4750\n",
      "Epoch 154/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9912 - mape: 5.9912\n",
      "Epoch 155/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.6467 - mape: 6.6467\n",
      "Epoch 156/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.0435 - mape: 6.0435\n",
      "Epoch 157/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.4010 - mape: 6.4010\n",
      "Epoch 158/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.0364 - mape: 6.0364\n",
      "Epoch 159/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9650 - mape: 5.9650\n",
      "Epoch 160/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8829 - mape: 5.8829\n",
      "Epoch 161/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.7199 - mape: 6.7199\n",
      "Epoch 162/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7816 - mape: 5.7816\n",
      "Epoch 163/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.7015 - mape: 5.7015\n",
      "Epoch 164/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.8958 - mape: 6.8958\n",
      "Epoch 165/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.4726 - mape: 6.4726\n",
      "Epoch 166/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8711 - mape: 5.8711\n",
      "Epoch 167/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.4698 - mape: 6.4698\n",
      "Epoch 168/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.0210 - mape: 6.0210\n",
      "Epoch 169/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8358 - mape: 5.8358\n",
      "Epoch 170/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.3918 - mape: 6.3918\n",
      "Epoch 171/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.6322 - mape: 5.6322\n",
      "Epoch 172/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4659 - mape: 5.4659\n",
      "Epoch 173/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3661 - mape: 5.3661\n",
      "Epoch 174/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.0004 - mape: 6.0004\n",
      "Epoch 175/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.8898 - mape: 6.8898\n",
      "Epoch 176/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6513 - mape: 5.6513\n",
      "Epoch 177/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8995 - mape: 5.8995\n",
      "Epoch 178/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6462 - mape: 5.6462\n",
      "Epoch 179/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9431 - mape: 4.9431\n",
      "Epoch 180/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1078 - mape: 5.1078\n",
      "Epoch 181/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.8497 - mape: 4.8497\n",
      "Epoch 182/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.7244 - mape: 5.7244\n",
      "Epoch 183/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8372 - mape: 5.8372\n",
      "Epoch 184/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7047 - mape: 5.7047\n",
      "Epoch 185/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.0090 - mape: 5.0090\n",
      "Epoch 186/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.4201 - mape: 6.4201\n",
      "Epoch 187/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3655 - mape: 5.3655\n",
      "Epoch 188/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8760 - mape: 5.8760\n",
      "Epoch 189/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.0642 - mape: 6.0642\n",
      "Epoch 190/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 5.4431 - mape: 5.4431\n",
      "Epoch 191/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.0186 - mape: 5.0186\n",
      "Epoch 192/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.0861 - mape: 5.0861\n",
      "Epoch 193/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.7169 - mape: 4.7169\n",
      "Epoch 194/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.5098 - mape: 5.5098\n",
      "Epoch 195/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.5354 - mape: 4.5354\n",
      "Epoch 196/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2670 - mape: 5.2670\n",
      "Epoch 197/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.2462 - mape: 6.2462\n",
      "Epoch 198/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.2031 - mape: 5.2031\n",
      "Epoch 199/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1596 - mape: 5.1596\n",
      "Epoch 200/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.0666 - mape: 5.0666\n",
      "Epoch 201/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2382 - mape: 5.2382\n",
      "Epoch 202/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8881 - mape: 5.8881\n",
      "Epoch 203/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3442 - mape: 5.3442\n",
      "Epoch 204/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.1866 - mape: 5.1866\n",
      "Epoch 205/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6152 - mape: 5.6152\n",
      "Epoch 206/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9087 - mape: 4.9087\n",
      "Epoch 207/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.0279 - mape: 5.0279\n",
      "Epoch 208/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9944 - mape: 5.9944\n",
      "Epoch 209/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2291 - mape: 5.2291\n",
      "Epoch 210/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.3802 - mape: 6.3802\n",
      "Epoch 211/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.1372 - mape: 6.1372\n",
      "Epoch 212/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 5.8337 - mape: 5.8337\n",
      "Epoch 213/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.5763 - mape: 5.5763\n",
      "Epoch 214/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.7187 - mape: 4.7187\n",
      "Epoch 215/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.1678 - mape: 6.1678\n",
      "Epoch 216/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2835 - mape: 5.2835\n",
      "Epoch 217/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.8732 - mape: 4.8732\n",
      "Epoch 218/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2989 - mape: 5.2989\n",
      "Epoch 219/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1888 - mape: 5.1888\n",
      "Epoch 220/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1050 - mape: 5.1050\n",
      "Epoch 221/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3119 - mape: 5.3119\n",
      "Epoch 222/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2255 - mape: 5.2255\n",
      "Epoch 223/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.0181 - mape: 5.0181\n",
      "Epoch 224/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.2912 - mape: 5.2912\n",
      "Epoch 225/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.4980 - mape: 5.4980\n",
      "Epoch 226/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.9236 - mape: 5.9236\n",
      "Epoch 227/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.9933 - mape: 6.9933\n",
      "Epoch 228/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.7886 - mape: 6.7886\n",
      "Epoch 229/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6314 - mape: 5.6314\n",
      "Epoch 230/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.5200 - mape: 6.5200\n",
      "Epoch 231/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 6.5485 - mape: 6.5485\n",
      "Epoch 232/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4931 - mape: 5.4931\n",
      "Epoch 233/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 5.0539 - mape: 5.0539\n",
      "Epoch 234/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.1013 - mape: 5.1013\n",
      "Epoch 235/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.7051 - mape: 5.7051\n",
      "Epoch 236/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.9501 - mape: 5.9501\n",
      "Epoch 237/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.6536 - mape: 4.6536\n",
      "Epoch 238/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7222 - mape: 5.7222\n",
      "Epoch 239/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4361 - mape: 5.4361\n",
      "Epoch 240/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4634 - mape: 5.4634\n",
      "Epoch 241/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3077 - mape: 5.3077\n",
      "Epoch 242/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 6.0343 - mape: 6.0343\n",
      "Epoch 243/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.2841 - mape: 5.2841\n",
      "Epoch 244/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9248 - mape: 4.9248\n",
      "Epoch 245/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.5034 - mape: 4.5034\n",
      "Epoch 246/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.6985 - mape: 4.6985\n",
      "Epoch 247/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4668 - mape: 5.4668\n",
      "Epoch 248/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.5848 - mape: 5.5848\n",
      "Epoch 249/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.6469 - mape: 4.6469\n",
      "Epoch 250/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1298 - mape: 5.1298\n",
      "Epoch 251/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.5790 - mape: 5.5790\n",
      "Epoch 252/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.7527 - mape: 5.7527\n",
      "Epoch 253/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.0996 - mape: 5.0996\n",
      "Epoch 254/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.4531 - mape: 5.4531\n",
      "Epoch 255/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.0286 - mape: 5.0286\n",
      "Epoch 256/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.9449 - mape: 4.9449\n",
      "Epoch 257/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.8159 - mape: 5.8159\n",
      "Epoch 258/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.1320 - mape: 5.1320\n",
      "Epoch 259/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.9267 - mape: 4.9267\n",
      "Epoch 260/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.5603 - mape: 4.5603\n",
      "Epoch 261/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.0988 - mape: 4.0988\n",
      "Epoch 262/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.2875 - mape: 4.2875\n",
      "Epoch 263/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.8479 - mape: 4.8479\n",
      "Epoch 264/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.1574 - mape: 5.1574\n",
      "Epoch 265/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.6494 - mape: 4.6494\n",
      "Epoch 266/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.2396 - mape: 4.2396\n",
      "Epoch 267/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.5415 - mape: 4.5415\n",
      "Epoch 268/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.0856 - mape: 4.0856\n",
      "Epoch 269/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 3.9488 - mape: 3.9488\n",
      "Epoch 270/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.8381 - mape: 4.8381\n",
      "Epoch 271/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.3601 - mape: 4.3601\n",
      "Epoch 272/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.2159 - mape: 4.2159\n",
      "Epoch 273/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.7903 - mape: 4.7903\n",
      "Epoch 274/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.3317 - mape: 5.3317\n",
      "Epoch 275/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2087 - mape: 5.2087\n",
      "Epoch 276/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 4.6345 - mape: 4.6345\n",
      "Epoch 277/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.6254 - mape: 4.6254\n",
      "Epoch 278/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.8900 - mape: 4.8900\n",
      "Epoch 279/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2243 - mape: 5.2243\n",
      "Epoch 280/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.2568 - mape: 5.2568\n",
      "Epoch 281/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.6747 - mape: 5.6747\n",
      "Epoch 282/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.0852 - mape: 4.0852\n",
      "Epoch 283/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.1235 - mape: 4.1235\n",
      "Epoch 284/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 3.9466 - mape: 3.9466\n",
      "Epoch 285/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9021 - mape: 4.9021\n",
      "Epoch 286/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.3188 - mape: 4.3188\n",
      "Epoch 287/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.1396 - mape: 4.1396\n",
      "Epoch 288/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.6269 - mape: 4.6269\n",
      "Epoch 289/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9213 - mape: 4.9213\n",
      "Epoch 290/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.5058 - mape: 4.5058\n",
      "Epoch 291/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.3377 - mape: 4.3377\n",
      "Epoch 292/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.8555 - mape: 4.8555\n",
      "Epoch 293/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.6277 - mape: 4.6277\n",
      "Epoch 294/300\n",
      "8/8 [==============================] - 0s 7ms/step - loss: 5.5055 - mape: 5.5055\n",
      "Epoch 295/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.5930 - mape: 5.5930\n",
      "Epoch 296/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 4.9352 - mape: 4.9352\n",
      "Epoch 297/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 4.7668 - mape: 4.7668\n",
      "Epoch 298/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.6664 - mape: 5.6664\n",
      "Epoch 299/300\n",
      "8/8 [==============================] - 0s 5ms/step - loss: 5.0130 - mape: 5.0130\n",
      "Epoch 300/300\n",
      "8/8 [==============================] - 0s 6ms/step - loss: 5.6555 - mape: 5.6555\n"
     ]
    }
   ],
   "source": [
    "from scikeras.wrappers import KerasRegressor\n",
    "batch_size = 32\n",
    "epochs = 300\n",
    "\n",
    "model = KerasRegressor(\n",
    "                       build_fn=define_model,\n",
    "                       epochs=epochs,\n",
    "                       batch_size=batch_size,                     \n",
    "                       verbose=1,\n",
    "                       drop = 0.1,\n",
    "                       )\n",
    "\n",
    "optimizer = ['SGD', 'RMSprop', 'Adagrad']\n",
    "drop = [0.1,0.05,0.01,0.005]\n",
    "\n",
    "param_grid = dict(drop = drop,optimizer=optimizer)\n",
    "grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs = 8, cv= 5)\n",
    "\n",
    "grid_result = grid.fit(X_train, y_train)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best: -0.313347 using {'drop': 0.01, 'optimizer': 'RMSprop'}\n",
      "Mean = -0.762322 (std=1.454045) with: {'drop': 0.1, 'optimizer': 'SGD'}\n",
      "Mean = -0.560674 (std=1.899837) with: {'drop': 0.1, 'optimizer': 'RMSprop'}\n",
      "Mean = -0.872457 (std=1.544733) with: {'drop': 0.1, 'optimizer': 'Adagrad'}\n",
      "Mean = -2.024741 (std=2.908786) with: {'drop': 0.05, 'optimizer': 'SGD'}\n",
      "Mean = -2.355383 (std=3.164144) with: {'drop': 0.05, 'optimizer': 'RMSprop'}\n",
      "Mean = -3.537838 (std=6.804406) with: {'drop': 0.05, 'optimizer': 'Adagrad'}\n",
      "Mean = -0.809223 (std=1.510325) with: {'drop': 0.01, 'optimizer': 'SGD'}\n",
      "Mean = -0.313347 (std=1.109537) with: {'drop': 0.01, 'optimizer': 'RMSprop'}\n",
      "Mean = -3.018976 (std=5.140518) with: {'drop': 0.01, 'optimizer': 'Adagrad'}\n",
      "Mean = -2.208167 (std=4.208206) with: {'drop': 0.005, 'optimizer': 'SGD'}\n",
      "Mean = -0.931255 (std=1.877850) with: {'drop': 0.005, 'optimizer': 'RMSprop'}\n",
      "Mean = -1.171376 (std=1.668623) with: {'drop': 0.005, 'optimizer': 'Adagrad'}\n"
     ]
    }
   ],
   "source": [
    "print(\"Best: %f using %s\" %(grid_result.best_score_,grid_result.best_params_))\n",
    "\n",
    "means = grid_result.cv_results_['mean_test_score']\n",
    "stds = grid_result.cv_results_['std_test_score']\n",
    "params = grid_result.cv_results_['params']\n",
    "for mean,stdev,param in zip(means,stds,params):\n",
    "    print(\"Mean = %f (std=%f) with: %r\" % (mean,stdev,param))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training and Model Validation\n",
    "In this section, the best performing model defined from grid search is trained on the given test and validation data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "best_model = define_model(drop=grid_result.best_params_[\"drop\"], optimizer=grid_result.best_params_[\"optimizer\"])\n",
    "cp_n = ModelCheckpoint(\"model_n/\", save_best_only=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Training mit Training und Validation Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: model_n\\assets\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'loss': [71.84213256835938,\n",
       "  37.62737274169922,\n",
       "  27.480791091918945,\n",
       "  20.614213943481445,\n",
       "  19.031972885131836,\n",
       "  20.2956600189209,\n",
       "  16.62161636352539,\n",
       "  16.946218490600586,\n",
       "  17.51696014404297,\n",
       "  16.36197280883789,\n",
       "  16.90349769592285,\n",
       "  15.996856689453125,\n",
       "  16.35455894470215,\n",
       "  14.285415649414062,\n",
       "  16.39510154724121,\n",
       "  15.63688850402832,\n",
       "  15.582076072692871,\n",
       "  13.860311508178711,\n",
       "  14.012740135192871,\n",
       "  14.37724781036377,\n",
       "  14.230175018310547,\n",
       "  15.031641960144043,\n",
       "  16.449050903320312,\n",
       "  12.358002662658691,\n",
       "  12.821013450622559,\n",
       "  14.527270317077637,\n",
       "  13.535833358764648,\n",
       "  11.549763679504395,\n",
       "  13.513591766357422,\n",
       "  13.41964340209961,\n",
       "  14.112432479858398,\n",
       "  14.497570037841797,\n",
       "  10.61849594116211,\n",
       "  12.634737014770508,\n",
       "  13.229074478149414,\n",
       "  13.241083145141602,\n",
       "  10.566801071166992,\n",
       "  10.397174835205078,\n",
       "  13.260279655456543,\n",
       "  12.2630033493042,\n",
       "  10.251762390136719,\n",
       "  12.750268936157227,\n",
       "  10.060876846313477,\n",
       "  14.267651557922363,\n",
       "  10.086051940917969,\n",
       "  11.836166381835938,\n",
       "  12.041860580444336,\n",
       "  12.323890686035156,\n",
       "  11.151262283325195,\n",
       "  11.119672775268555,\n",
       "  13.246173858642578,\n",
       "  10.617903709411621,\n",
       "  11.240455627441406,\n",
       "  8.882596015930176,\n",
       "  11.234992027282715,\n",
       "  10.602014541625977,\n",
       "  10.926383972167969,\n",
       "  10.476204872131348,\n",
       "  10.806028366088867,\n",
       "  10.177753448486328,\n",
       "  13.002172470092773,\n",
       "  10.588665008544922,\n",
       "  12.688042640686035,\n",
       "  11.20759391784668,\n",
       "  9.844188690185547,\n",
       "  10.86207103729248,\n",
       "  11.831186294555664,\n",
       "  9.609359741210938,\n",
       "  10.213603973388672,\n",
       "  9.768255233764648,\n",
       "  10.876269340515137,\n",
       "  10.869553565979004,\n",
       "  10.621725082397461,\n",
       "  9.632428169250488,\n",
       "  10.511184692382812,\n",
       "  11.05257797241211,\n",
       "  11.11996078491211,\n",
       "  8.991172790527344,\n",
       "  8.44760513305664,\n",
       "  13.235373497009277,\n",
       "  10.071029663085938,\n",
       "  11.077529907226562,\n",
       "  9.571946144104004,\n",
       "  8.7649564743042,\n",
       "  10.454276084899902,\n",
       "  9.558585166931152,\n",
       "  7.94511604309082,\n",
       "  9.71408462524414,\n",
       "  8.922264099121094,\n",
       "  10.059677124023438,\n",
       "  8.626221656799316,\n",
       "  10.548242568969727,\n",
       "  9.619033813476562,\n",
       "  10.4122953414917,\n",
       "  8.717698097229004,\n",
       "  10.772566795349121,\n",
       "  8.32815170288086,\n",
       "  9.957245826721191,\n",
       "  9.191422462463379,\n",
       "  10.977322578430176,\n",
       "  10.03602123260498,\n",
       "  10.23031997680664,\n",
       "  10.693882942199707,\n",
       "  10.338234901428223,\n",
       "  9.540824890136719,\n",
       "  10.108179092407227,\n",
       "  8.517462730407715,\n",
       "  9.841012001037598,\n",
       "  10.384299278259277,\n",
       "  9.324297904968262,\n",
       "  9.585082054138184,\n",
       "  8.712425231933594,\n",
       "  8.538002014160156,\n",
       "  11.363584518432617,\n",
       "  8.859743118286133,\n",
       "  9.593547821044922,\n",
       "  8.527316093444824,\n",
       "  9.5687894821167,\n",
       "  8.649155616760254,\n",
       "  9.493870735168457,\n",
       "  9.105796813964844,\n",
       "  9.59396743774414,\n",
       "  8.515303611755371,\n",
       "  10.036194801330566,\n",
       "  8.30874252319336,\n",
       "  10.318338394165039,\n",
       "  10.237028121948242,\n",
       "  8.542516708374023,\n",
       "  8.163163185119629,\n",
       "  9.906356811523438,\n",
       "  7.963354587554932,\n",
       "  8.078166961669922,\n",
       "  9.808929443359375,\n",
       "  8.237049102783203,\n",
       "  9.183206558227539,\n",
       "  8.852538108825684,\n",
       "  7.9218573570251465,\n",
       "  10.293723106384277,\n",
       "  7.848324298858643,\n",
       "  8.914780616760254,\n",
       "  8.215850830078125,\n",
       "  8.369915008544922,\n",
       "  9.162160873413086,\n",
       "  9.41540813446045,\n",
       "  8.14696979522705,\n",
       "  7.197159767150879,\n",
       "  8.889883041381836,\n",
       "  8.256574630737305,\n",
       "  9.257499694824219,\n",
       "  7.919360160827637,\n",
       "  8.400856018066406,\n",
       "  9.022744178771973,\n",
       "  8.765847206115723,\n",
       "  8.087747573852539,\n",
       "  9.123649597167969,\n",
       "  8.369596481323242,\n",
       "  7.876274108886719,\n",
       "  8.46153450012207,\n",
       "  7.995826244354248,\n",
       "  9.828357696533203,\n",
       "  7.8582892417907715,\n",
       "  6.760960578918457,\n",
       "  8.333741188049316,\n",
       "  8.364931106567383,\n",
       "  7.938633918762207,\n",
       "  8.206713676452637,\n",
       "  7.79386568069458,\n",
       "  8.742019653320312,\n",
       "  9.783883094787598,\n",
       "  7.793449401855469,\n",
       "  7.9505462646484375,\n",
       "  7.449461460113525,\n",
       "  7.663033485412598,\n",
       "  8.617952346801758,\n",
       "  7.1702656745910645,\n",
       "  7.95973539352417,\n",
       "  8.196053504943848,\n",
       "  8.871017456054688,\n",
       "  7.82059383392334,\n",
       "  6.761845588684082,\n",
       "  8.955889701843262,\n",
       "  7.8398213386535645,\n",
       "  8.41222095489502,\n",
       "  9.189316749572754,\n",
       "  7.2618255615234375,\n",
       "  8.446221351623535,\n",
       "  7.7628936767578125,\n",
       "  9.142935752868652,\n",
       "  8.6626558303833,\n",
       "  7.94429874420166,\n",
       "  7.754384994506836,\n",
       "  7.0225019454956055,\n",
       "  6.480770587921143,\n",
       "  8.39262580871582,\n",
       "  7.724310398101807,\n",
       "  6.680017471313477,\n",
       "  8.305519104003906,\n",
       "  8.31091022491455,\n",
       "  7.135566234588623,\n",
       "  7.232278823852539,\n",
       "  8.166488647460938,\n",
       "  7.446841716766357,\n",
       "  7.265542984008789,\n",
       "  7.389220237731934,\n",
       "  8.076854705810547,\n",
       "  7.534738540649414,\n",
       "  7.749157905578613,\n",
       "  7.6468658447265625,\n",
       "  6.207286834716797,\n",
       "  7.8043036460876465,\n",
       "  7.895034313201904,\n",
       "  7.0341339111328125,\n",
       "  7.0112833976745605,\n",
       "  7.868298530578613,\n",
       "  7.380251884460449,\n",
       "  7.212913513183594,\n",
       "  8.01546859741211,\n",
       "  6.267679691314697,\n",
       "  7.480772972106934,\n",
       "  7.665228843688965,\n",
       "  6.968308448791504,\n",
       "  7.0972208976745605,\n",
       "  7.396210670471191,\n",
       "  7.529797554016113,\n",
       "  7.461294651031494,\n",
       "  6.667644023895264,\n",
       "  7.204059600830078,\n",
       "  6.530356407165527,\n",
       "  8.598024368286133,\n",
       "  6.779585838317871,\n",
       "  6.917820453643799,\n",
       "  6.287054538726807,\n",
       "  7.801990032196045,\n",
       "  7.000175952911377,\n",
       "  6.1878743171691895,\n",
       "  7.454387664794922,\n",
       "  6.878872394561768,\n",
       "  8.234322547912598,\n",
       "  6.785667419433594,\n",
       "  6.758298873901367,\n",
       "  7.134435653686523,\n",
       "  6.694054126739502,\n",
       "  7.0839409828186035,\n",
       "  7.485688209533691,\n",
       "  6.799443244934082,\n",
       "  6.768490791320801,\n",
       "  7.63771390914917,\n",
       "  6.467884063720703,\n",
       "  7.481230735778809,\n",
       "  7.533239364624023,\n",
       "  6.788309574127197,\n",
       "  6.319194316864014,\n",
       "  6.923039436340332,\n",
       "  6.816992282867432,\n",
       "  6.492179870605469,\n",
       "  7.097285747528076,\n",
       "  6.25166654586792,\n",
       "  7.68004846572876,\n",
       "  7.878579616546631,\n",
       "  6.588149547576904,\n",
       "  6.810642242431641,\n",
       "  6.498405933380127,\n",
       "  7.677929878234863,\n",
       "  7.323472023010254,\n",
       "  6.713583469390869,\n",
       "  6.040391445159912,\n",
       "  6.227593898773193,\n",
       "  7.176214694976807,\n",
       "  8.127742767333984,\n",
       "  6.825228691101074,\n",
       "  6.296786785125732,\n",
       "  6.927177429199219,\n",
       "  6.418667793273926,\n",
       "  6.494465351104736,\n",
       "  7.195237159729004,\n",
       "  6.569384574890137,\n",
       "  6.62149715423584,\n",
       "  5.802986145019531,\n",
       "  7.3670196533203125,\n",
       "  6.827986240386963,\n",
       "  5.856072902679443,\n",
       "  7.335781097412109,\n",
       "  6.543778419494629,\n",
       "  5.7210869789123535,\n",
       "  8.059049606323242,\n",
       "  5.829472541809082,\n",
       "  7.2427449226379395,\n",
       "  6.561563491821289,\n",
       "  7.020979881286621,\n",
       "  6.858595848083496,\n",
       "  6.410332679748535,\n",
       "  5.050848007202148,\n",
       "  7.195659160614014,\n",
       "  5.997049808502197,\n",
       "  6.034939289093018,\n",
       "  6.514571189880371,\n",
       "  7.372781276702881,\n",
       "  6.795056343078613,\n",
       "  6.040008068084717,\n",
       "  6.336002349853516,\n",
       "  6.999300003051758,\n",
       "  6.005327224731445,\n",
       "  7.45535945892334,\n",
       "  6.534383296966553,\n",
       "  6.67570161819458,\n",
       "  6.666004657745361,\n",
       "  6.759767532348633,\n",
       "  6.204916000366211,\n",
       "  5.875921726226807,\n",
       "  6.594703674316406,\n",
       "  6.478421688079834,\n",
       "  6.591340065002441,\n",
       "  7.127359867095947,\n",
       "  5.73553466796875,\n",
       "  6.547192096710205,\n",
       "  7.667715549468994,\n",
       "  5.854055881500244,\n",
       "  6.763865947723389,\n",
       "  5.952390670776367,\n",
       "  5.945060729980469,\n",
       "  6.97554874420166,\n",
       "  6.085504055023193,\n",
       "  6.124736785888672,\n",
       "  6.205444812774658,\n",
       "  5.931761741638184,\n",
       "  5.916987419128418,\n",
       "  6.576712131500244,\n",
       "  6.963075637817383,\n",
       "  6.0466766357421875,\n",
       "  6.23590087890625,\n",
       "  5.6171345710754395,\n",
       "  6.638739585876465,\n",
       "  6.5844268798828125,\n",
       "  5.481698513031006,\n",
       "  6.935417175292969,\n",
       "  6.80951452255249,\n",
       "  5.718283653259277,\n",
       "  5.958721160888672,\n",
       "  6.6796464920043945,\n",
       "  6.2743024826049805,\n",
       "  5.4649481773376465,\n",
       "  5.280467510223389,\n",
       "  6.37878942489624,\n",
       "  5.662500858306885,\n",
       "  5.696841716766357,\n",
       "  6.108622074127197,\n",
       "  6.863558292388916,\n",
       "  6.800041675567627,\n",
       "  5.673215389251709,\n",
       "  5.570806503295898,\n",
       "  5.676109790802002,\n",
       "  6.667434215545654,\n",
       "  6.012537479400635,\n",
       "  5.52169942855835,\n",
       "  5.843184471130371,\n",
       "  5.718277931213379,\n",
       "  5.374483585357666,\n",
       "  6.640198230743408,\n",
       "  6.4122161865234375,\n",
       "  6.267284870147705,\n",
       "  5.81055212020874,\n",
       "  6.136805534362793,\n",
       "  6.599950790405273,\n",
       "  6.705836772918701,\n",
       "  5.479508399963379,\n",
       "  5.282500743865967,\n",
       "  6.625036716461182,\n",
       "  6.17461633682251,\n",
       "  6.615903377532959,\n",
       "  5.8258771896362305,\n",
       "  6.130025863647461,\n",
       "  5.554210662841797,\n",
       "  5.804098129272461,\n",
       "  5.115693092346191,\n",
       "  6.7227301597595215,\n",
       "  6.303426742553711,\n",
       "  5.689883708953857,\n",
       "  5.23836612701416,\n",
       "  5.874710559844971,\n",
       "  6.6328206062316895,\n",
       "  6.057425498962402,\n",
       "  7.073878288269043,\n",
       "  5.867114543914795,\n",
       "  6.2701416015625,\n",
       "  5.146039009094238,\n",
       "  5.096616268157959,\n",
       "  5.609146595001221,\n",
       "  4.758760452270508,\n",
       "  5.721353530883789,\n",
       "  5.389133930206299,\n",
       "  7.056066989898682,\n",
       "  4.760824203491211,\n",
       "  5.926266670227051,\n",
       "  6.792305946350098,\n",
       "  5.669538497924805,\n",
       "  5.353050708770752,\n",
       "  4.74342155456543,\n",
       "  5.6696672439575195,\n",
       "  5.726839065551758,\n",
       "  5.7107625007629395,\n",
       "  6.176100730895996,\n",
       "  5.673054218292236,\n",
       "  5.48049783706665,\n",
       "  5.188520431518555,\n",
       "  5.972352027893066,\n",
       "  6.031451225280762,\n",
       "  5.629551887512207,\n",
       "  7.156277656555176,\n",
       "  5.959214210510254,\n",
       "  5.407836437225342,\n",
       "  5.615020275115967,\n",
       "  5.113141059875488,\n",
       "  6.150355339050293,\n",
       "  5.77457857131958,\n",
       "  5.783969879150391,\n",
       "  5.515460014343262,\n",
       "  5.805061340332031,\n",
       "  5.080686569213867,\n",
       "  5.919998645782471,\n",
       "  5.719204902648926,\n",
       "  5.330305099487305,\n",
       "  6.052525997161865,\n",
       "  5.76485538482666,\n",
       "  5.848709583282471,\n",
       "  5.871634006500244,\n",
       "  6.188107967376709,\n",
       "  5.015067100524902,\n",
       "  5.688781261444092,\n",
       "  5.515068054199219,\n",
       "  5.61531400680542,\n",
       "  4.842020511627197,\n",
       "  5.778710842132568,\n",
       "  6.045665740966797,\n",
       "  5.7311601638793945,\n",
       "  5.854386806488037,\n",
       "  5.084052562713623,\n",
       "  5.2331366539001465,\n",
       "  5.563003063201904,\n",
       "  5.774744510650635,\n",
       "  5.848795413970947,\n",
       "  5.928769111633301,\n",
       "  5.096157550811768,\n",
       "  5.238570690155029,\n",
       "  5.887909412384033,\n",
       "  5.409764289855957,\n",
       "  4.40825891494751,\n",
       "  5.092676639556885,\n",
       "  5.556719779968262,\n",
       "  5.8568501472473145,\n",
       "  5.2536492347717285,\n",
       "  5.03596305847168,\n",
       "  6.178236484527588,\n",
       "  5.30592155456543,\n",
       "  4.868119716644287,\n",
       "  6.175187110900879,\n",
       "  4.663137912750244,\n",
       "  6.042108058929443,\n",
       "  4.847924709320068,\n",
       "  5.070185661315918,\n",
       "  5.196875095367432,\n",
       "  5.100446701049805,\n",
       "  5.394138813018799,\n",
       "  5.286220073699951,\n",
       "  6.424188137054443,\n",
       "  5.247403621673584,\n",
       "  5.182949066162109,\n",
       "  5.253805160522461,\n",
       "  5.048996925354004,\n",
       "  4.923771858215332,\n",
       "  4.96286153793335,\n",
       "  5.412273406982422,\n",
       "  5.554286956787109,\n",
       "  5.207097053527832,\n",
       "  6.07520055770874,\n",
       "  4.976685523986816,\n",
       "  5.081509113311768,\n",
       "  5.326171875,\n",
       "  5.4420928955078125,\n",
       "  4.5683159828186035,\n",
       "  5.917113780975342,\n",
       "  5.079334735870361,\n",
       "  5.563129425048828,\n",
       "  4.7679243087768555,\n",
       "  5.528578758239746,\n",
       "  3.8719534873962402,\n",
       "  4.750521659851074,\n",
       "  5.701635360717773,\n",
       "  5.742092609405518,\n",
       "  4.9177045822143555,\n",
       "  5.760986804962158,\n",
       "  5.261519908905029,\n",
       "  4.418517112731934,\n",
       "  5.210386276245117,\n",
       "  5.5773515701293945,\n",
       "  4.867336750030518,\n",
       "  4.988964080810547,\n",
       "  5.17173433303833,\n",
       "  5.866097450256348,\n",
       "  5.628702163696289,\n",
       "  4.99838924407959,\n",
       "  4.199275493621826,\n",
       "  5.159941673278809,\n",
       "  4.487574100494385,\n",
       "  5.1383819580078125,\n",
       "  5.056176662445068,\n",
       "  6.0571088790893555,\n",
       "  4.769680023193359,\n",
       "  5.787027835845947,\n",
       "  5.4376935958862305,\n",
       "  4.737827777862549,\n",
       "  4.605546951293945,\n",
       "  5.179082870483398,\n",
       "  4.475465297698975,\n",
       "  5.482834815979004,\n",
       "  5.2783522605896,\n",
       "  4.736172676086426,\n",
       "  4.037018775939941,\n",
       "  5.056282043457031,\n",
       "  5.561960220336914,\n",
       "  5.084578990936279,\n",
       "  5.3997321128845215,\n",
       "  5.606127738952637,\n",
       "  4.5007195472717285,\n",
       "  5.044325351715088,\n",
       "  5.243644714355469,\n",
       "  4.7694268226623535,\n",
       "  5.257894039154053,\n",
       "  4.9729180335998535,\n",
       "  4.75773811340332,\n",
       "  5.225069046020508,\n",
       "  5.1748738288879395,\n",
       "  4.633727073669434,\n",
       "  4.551016807556152,\n",
       "  4.309856414794922,\n",
       "  4.5417351722717285,\n",
       "  4.853643417358398,\n",
       "  5.10538387298584,\n",
       "  5.139662265777588,\n",
       "  5.258285999298096,\n",
       "  4.501973628997803,\n",
       "  5.008573055267334,\n",
       "  5.052850723266602,\n",
       "  4.951686382293701,\n",
       "  4.66091775894165,\n",
       "  4.803896903991699,\n",
       "  5.3039140701293945,\n",
       "  5.3527703285217285,\n",
       "  4.116580963134766,\n",
       "  5.58057975769043,\n",
       "  4.748333930969238,\n",
       "  5.256656169891357,\n",
       "  4.602118015289307,\n",
       "  4.942258834838867,\n",
       "  4.811962604522705,\n",
       "  4.09275484085083,\n",
       "  5.550209045410156,\n",
       "  5.157925128936768,\n",
       "  5.5393967628479,\n",
       "  4.451102256774902,\n",
       "  4.8215765953063965,\n",
       "  5.118683815002441,\n",
       "  4.489153861999512,\n",
       "  4.698952674865723,\n",
       "  3.739699125289917,\n",
       "  4.49470853805542,\n",
       "  5.5609636306762695,\n",
       "  4.419373512268066,\n",
       "  4.440491676330566,\n",
       "  4.657710075378418,\n",
       "  4.60772180557251,\n",
       "  4.986242294311523,\n",
       "  4.736499786376953,\n",
       "  4.463547706604004,\n",
       "  4.913445472717285,\n",
       "  5.1469645500183105,\n",
       "  4.443927764892578,\n",
       "  5.379999160766602,\n",
       "  4.616513252258301,\n",
       "  4.71842622756958,\n",
       "  4.884434700012207,\n",
       "  4.640573501586914,\n",
       "  5.0231428146362305,\n",
       "  4.71310567855835,\n",
       "  4.8584794998168945,\n",
       "  5.1766486167907715,\n",
       "  4.108058452606201,\n",
       "  5.680817127227783,\n",
       "  4.3322930335998535,\n",
       "  4.635502815246582,\n",
       "  4.548344135284424,\n",
       "  5.018511772155762,\n",
       "  4.727330207824707,\n",
       "  4.503907203674316,\n",
       "  4.908830165863037,\n",
       "  4.834928035736084,\n",
       "  4.2016143798828125,\n",
       "  4.617180824279785,\n",
       "  4.272310733795166,\n",
       "  4.630500316619873,\n",
       "  5.086462020874023],\n",
       " 'mape': [71.84213256835938,\n",
       "  37.62737274169922,\n",
       "  27.480791091918945,\n",
       "  20.614213943481445,\n",
       "  19.031972885131836,\n",
       "  20.2956600189209,\n",
       "  16.62161636352539,\n",
       "  16.946218490600586,\n",
       "  17.51696014404297,\n",
       "  16.36197280883789,\n",
       "  16.90349769592285,\n",
       "  15.996856689453125,\n",
       "  16.35455894470215,\n",
       "  14.285415649414062,\n",
       "  16.39510154724121,\n",
       "  15.63688850402832,\n",
       "  15.582076072692871,\n",
       "  13.860311508178711,\n",
       "  14.012740135192871,\n",
       "  14.37724781036377,\n",
       "  14.230175018310547,\n",
       "  15.031641960144043,\n",
       "  16.449050903320312,\n",
       "  12.358002662658691,\n",
       "  12.821013450622559,\n",
       "  14.527270317077637,\n",
       "  13.535834312438965,\n",
       "  11.549763679504395,\n",
       "  13.513591766357422,\n",
       "  13.419642448425293,\n",
       "  14.112432479858398,\n",
       "  14.497570037841797,\n",
       "  10.61849594116211,\n",
       "  12.634737014770508,\n",
       "  13.229074478149414,\n",
       "  13.241083145141602,\n",
       "  10.566801071166992,\n",
       "  10.397174835205078,\n",
       "  13.260279655456543,\n",
       "  12.2630033493042,\n",
       "  10.251762390136719,\n",
       "  12.750268936157227,\n",
       "  10.060876846313477,\n",
       "  14.267651557922363,\n",
       "  10.086051940917969,\n",
       "  11.836166381835938,\n",
       "  12.041860580444336,\n",
       "  12.323890686035156,\n",
       "  11.151262283325195,\n",
       "  11.119672775268555,\n",
       "  13.246173858642578,\n",
       "  10.617903709411621,\n",
       "  11.240455627441406,\n",
       "  8.882596015930176,\n",
       "  11.234992027282715,\n",
       "  10.602014541625977,\n",
       "  10.926383972167969,\n",
       "  10.476204872131348,\n",
       "  10.806028366088867,\n",
       "  10.177753448486328,\n",
       "  13.002172470092773,\n",
       "  10.588664054870605,\n",
       "  12.688042640686035,\n",
       "  11.20759391784668,\n",
       "  9.844188690185547,\n",
       "  10.86207103729248,\n",
       "  11.831186294555664,\n",
       "  9.609359741210938,\n",
       "  10.213603973388672,\n",
       "  9.768255233764648,\n",
       "  10.876269340515137,\n",
       "  10.869553565979004,\n",
       "  10.621725082397461,\n",
       "  9.632428169250488,\n",
       "  10.511184692382812,\n",
       "  11.05257797241211,\n",
       "  11.11996078491211,\n",
       "  8.991172790527344,\n",
       "  8.44760513305664,\n",
       "  13.235373497009277,\n",
       "  10.071029663085938,\n",
       "  11.077529907226562,\n",
       "  9.571946144104004,\n",
       "  8.7649564743042,\n",
       "  10.454276084899902,\n",
       "  9.558585166931152,\n",
       "  7.94511604309082,\n",
       "  9.71408462524414,\n",
       "  8.922264099121094,\n",
       "  10.059677124023438,\n",
       "  8.626221656799316,\n",
       "  10.548242568969727,\n",
       "  9.619033813476562,\n",
       "  10.4122953414917,\n",
       "  8.717698097229004,\n",
       "  10.772566795349121,\n",
       "  8.32815170288086,\n",
       "  9.957245826721191,\n",
       "  9.191422462463379,\n",
       "  10.977322578430176,\n",
       "  10.03602123260498,\n",
       "  10.23031997680664,\n",
       "  10.693882942199707,\n",
       "  10.338234901428223,\n",
       "  9.540824890136719,\n",
       "  10.108179092407227,\n",
       "  8.517462730407715,\n",
       "  9.841012001037598,\n",
       "  10.384299278259277,\n",
       "  9.324297904968262,\n",
       "  9.585082054138184,\n",
       "  8.712425231933594,\n",
       "  8.538002014160156,\n",
       "  11.363584518432617,\n",
       "  8.859743118286133,\n",
       "  9.593547821044922,\n",
       "  8.527316093444824,\n",
       "  9.5687894821167,\n",
       "  8.649155616760254,\n",
       "  9.493870735168457,\n",
       "  9.105796813964844,\n",
       "  9.59396743774414,\n",
       "  8.515303611755371,\n",
       "  10.036194801330566,\n",
       "  8.30874252319336,\n",
       "  10.318338394165039,\n",
       "  10.237028121948242,\n",
       "  8.542516708374023,\n",
       "  8.163163185119629,\n",
       "  9.906356811523438,\n",
       "  7.963354587554932,\n",
       "  8.078166961669922,\n",
       "  9.808929443359375,\n",
       "  8.237049102783203,\n",
       "  9.183206558227539,\n",
       "  8.852538108825684,\n",
       "  7.9218573570251465,\n",
       "  10.293723106384277,\n",
       "  7.848324298858643,\n",
       "  8.914780616760254,\n",
       "  8.215850830078125,\n",
       "  8.369915008544922,\n",
       "  9.162160873413086,\n",
       "  9.41540813446045,\n",
       "  8.14696979522705,\n",
       "  7.197159767150879,\n",
       "  8.889883041381836,\n",
       "  8.256574630737305,\n",
       "  9.257499694824219,\n",
       "  7.919360160827637,\n",
       "  8.400856018066406,\n",
       "  9.022744178771973,\n",
       "  8.765847206115723,\n",
       "  8.087747573852539,\n",
       "  9.123649597167969,\n",
       "  8.369596481323242,\n",
       "  7.876274108886719,\n",
       "  8.46153450012207,\n",
       "  7.995826244354248,\n",
       "  9.828357696533203,\n",
       "  7.8582892417907715,\n",
       "  6.760960578918457,\n",
       "  8.333741188049316,\n",
       "  8.364931106567383,\n",
       "  7.938633918762207,\n",
       "  8.206713676452637,\n",
       "  7.79386568069458,\n",
       "  8.742019653320312,\n",
       "  9.783883094787598,\n",
       "  7.793449401855469,\n",
       "  7.9505462646484375,\n",
       "  7.449461460113525,\n",
       "  7.663033485412598,\n",
       "  8.617952346801758,\n",
       "  7.1702656745910645,\n",
       "  7.95973539352417,\n",
       "  8.196053504943848,\n",
       "  8.871017456054688,\n",
       "  7.82059383392334,\n",
       "  6.761845588684082,\n",
       "  8.955889701843262,\n",
       "  7.8398213386535645,\n",
       "  8.41222095489502,\n",
       "  9.189316749572754,\n",
       "  7.2618255615234375,\n",
       "  8.446221351623535,\n",
       "  7.7628936767578125,\n",
       "  9.142935752868652,\n",
       "  8.6626558303833,\n",
       "  7.94429874420166,\n",
       "  7.754384994506836,\n",
       "  7.0225019454956055,\n",
       "  6.480770587921143,\n",
       "  8.392626762390137,\n",
       "  7.724310398101807,\n",
       "  6.680017471313477,\n",
       "  8.305519104003906,\n",
       "  8.31091022491455,\n",
       "  7.135565757751465,\n",
       "  7.232278823852539,\n",
       "  8.166488647460938,\n",
       "  7.446841716766357,\n",
       "  7.265542984008789,\n",
       "  7.389220237731934,\n",
       "  8.076854705810547,\n",
       "  7.534738540649414,\n",
       "  7.749157905578613,\n",
       "  7.6468658447265625,\n",
       "  6.207286834716797,\n",
       "  7.8043036460876465,\n",
       "  7.895034313201904,\n",
       "  7.0341339111328125,\n",
       "  7.0112833976745605,\n",
       "  7.868298530578613,\n",
       "  7.380251884460449,\n",
       "  7.212913513183594,\n",
       "  8.01546859741211,\n",
       "  6.267679691314697,\n",
       "  7.480772972106934,\n",
       "  7.665228843688965,\n",
       "  6.968308448791504,\n",
       "  7.0972208976745605,\n",
       "  7.396210670471191,\n",
       "  7.529797554016113,\n",
       "  7.461294651031494,\n",
       "  6.667644023895264,\n",
       "  7.204059600830078,\n",
       "  6.530356407165527,\n",
       "  8.598024368286133,\n",
       "  6.779585838317871,\n",
       "  6.917820453643799,\n",
       "  6.287054538726807,\n",
       "  7.801990032196045,\n",
       "  7.000175952911377,\n",
       "  6.1878743171691895,\n",
       "  7.454387664794922,\n",
       "  6.878872394561768,\n",
       "  8.234322547912598,\n",
       "  6.785667419433594,\n",
       "  6.758298873901367,\n",
       "  7.134435653686523,\n",
       "  6.694054126739502,\n",
       "  7.0839409828186035,\n",
       "  7.485688209533691,\n",
       "  6.799443244934082,\n",
       "  6.768490791320801,\n",
       "  7.63771390914917,\n",
       "  6.467884063720703,\n",
       "  7.481230735778809,\n",
       "  7.533239364624023,\n",
       "  6.788309574127197,\n",
       "  6.319194316864014,\n",
       "  6.923039436340332,\n",
       "  6.816992282867432,\n",
       "  6.492179870605469,\n",
       "  7.097285747528076,\n",
       "  6.25166654586792,\n",
       "  7.68004846572876,\n",
       "  7.878579616546631,\n",
       "  6.588149547576904,\n",
       "  6.810642242431641,\n",
       "  6.498405933380127,\n",
       "  7.677929878234863,\n",
       "  7.323472023010254,\n",
       "  6.713583469390869,\n",
       "  6.040391445159912,\n",
       "  6.227593898773193,\n",
       "  7.176214694976807,\n",
       "  8.127742767333984,\n",
       "  6.825228691101074,\n",
       "  6.296786785125732,\n",
       "  6.927177429199219,\n",
       "  6.418667793273926,\n",
       "  6.494465351104736,\n",
       "  7.195237159729004,\n",
       "  6.569384574890137,\n",
       "  6.62149715423584,\n",
       "  5.802986145019531,\n",
       "  7.3670196533203125,\n",
       "  6.827986240386963,\n",
       "  5.856072902679443,\n",
       "  7.335781097412109,\n",
       "  6.543778419494629,\n",
       "  5.7210869789123535,\n",
       "  8.059049606323242,\n",
       "  5.829472541809082,\n",
       "  7.2427449226379395,\n",
       "  6.561563491821289,\n",
       "  7.020979881286621,\n",
       "  6.858595848083496,\n",
       "  6.410332679748535,\n",
       "  5.050848007202148,\n",
       "  7.195659160614014,\n",
       "  5.997049808502197,\n",
       "  6.034939289093018,\n",
       "  6.514571189880371,\n",
       "  7.372781276702881,\n",
       "  6.795056343078613,\n",
       "  6.040008068084717,\n",
       "  6.336002349853516,\n",
       "  6.999300003051758,\n",
       "  6.005327224731445,\n",
       "  7.45535945892334,\n",
       "  6.534383296966553,\n",
       "  6.67570161819458,\n",
       "  6.666004657745361,\n",
       "  6.759767532348633,\n",
       "  6.204916000366211,\n",
       "  5.875921726226807,\n",
       "  6.594703674316406,\n",
       "  6.478421688079834,\n",
       "  6.591340065002441,\n",
       "  7.127359867095947,\n",
       "  5.73553466796875,\n",
       "  6.547192096710205,\n",
       "  7.667715549468994,\n",
       "  5.854055881500244,\n",
       "  6.763865947723389,\n",
       "  5.952390670776367,\n",
       "  5.945060729980469,\n",
       "  6.97554874420166,\n",
       "  6.085504055023193,\n",
       "  6.124736785888672,\n",
       "  6.205444812774658,\n",
       "  5.931761741638184,\n",
       "  5.916987419128418,\n",
       "  6.576712131500244,\n",
       "  6.963075637817383,\n",
       "  6.0466766357421875,\n",
       "  6.23590087890625,\n",
       "  5.6171345710754395,\n",
       "  6.638739585876465,\n",
       "  6.5844268798828125,\n",
       "  5.481698513031006,\n",
       "  6.935417175292969,\n",
       "  6.80951452255249,\n",
       "  5.718283653259277,\n",
       "  5.958721160888672,\n",
       "  6.6796464920043945,\n",
       "  6.2743024826049805,\n",
       "  5.4649481773376465,\n",
       "  5.280467510223389,\n",
       "  6.37878942489624,\n",
       "  5.662500858306885,\n",
       "  5.696841716766357,\n",
       "  6.108622074127197,\n",
       "  6.863558292388916,\n",
       "  6.800041675567627,\n",
       "  5.673215389251709,\n",
       "  5.570806503295898,\n",
       "  5.676109790802002,\n",
       "  6.6674346923828125,\n",
       "  6.012537479400635,\n",
       "  5.52169942855835,\n",
       "  5.843184471130371,\n",
       "  5.718277931213379,\n",
       "  5.374483585357666,\n",
       "  6.640198230743408,\n",
       "  6.4122161865234375,\n",
       "  6.267284870147705,\n",
       "  5.81055212020874,\n",
       "  6.136805534362793,\n",
       "  6.599950790405273,\n",
       "  6.705836772918701,\n",
       "  5.479507923126221,\n",
       "  5.282500743865967,\n",
       "  6.625036716461182,\n",
       "  6.17461633682251,\n",
       "  6.615903377532959,\n",
       "  5.8258771896362305,\n",
       "  6.130025863647461,\n",
       "  5.554210662841797,\n",
       "  5.804098129272461,\n",
       "  5.115693092346191,\n",
       "  6.7227301597595215,\n",
       "  6.303426742553711,\n",
       "  5.689883708953857,\n",
       "  5.23836612701416,\n",
       "  5.874710559844971,\n",
       "  6.6328206062316895,\n",
       "  6.057425498962402,\n",
       "  7.073878288269043,\n",
       "  5.867114543914795,\n",
       "  6.2701416015625,\n",
       "  5.146039009094238,\n",
       "  5.096616268157959,\n",
       "  5.609146595001221,\n",
       "  4.758760452270508,\n",
       "  5.721353530883789,\n",
       "  5.389133930206299,\n",
       "  7.056066989898682,\n",
       "  4.760824203491211,\n",
       "  5.926266670227051,\n",
       "  6.792305946350098,\n",
       "  5.669538497924805,\n",
       "  5.353050708770752,\n",
       "  4.74342155456543,\n",
       "  5.6696672439575195,\n",
       "  5.726839065551758,\n",
       "  5.7107625007629395,\n",
       "  6.176100730895996,\n",
       "  5.673054218292236,\n",
       "  5.48049783706665,\n",
       "  5.188520431518555,\n",
       "  5.972352027893066,\n",
       "  6.031451225280762,\n",
       "  5.629551887512207,\n",
       "  7.156277656555176,\n",
       "  5.959214210510254,\n",
       "  5.407836437225342,\n",
       "  5.615020275115967,\n",
       "  5.113141059875488,\n",
       "  6.150355339050293,\n",
       "  5.77457857131958,\n",
       "  5.783969879150391,\n",
       "  5.515460014343262,\n",
       "  5.805061340332031,\n",
       "  5.080686569213867,\n",
       "  5.919998645782471,\n",
       "  5.719204902648926,\n",
       "  5.330305099487305,\n",
       "  6.052525997161865,\n",
       "  5.76485538482666,\n",
       "  5.848709583282471,\n",
       "  5.871634006500244,\n",
       "  6.188107967376709,\n",
       "  5.015067100524902,\n",
       "  5.688781261444092,\n",
       "  5.515068054199219,\n",
       "  5.61531400680542,\n",
       "  4.842020511627197,\n",
       "  5.778710842132568,\n",
       "  6.045665740966797,\n",
       "  5.7311601638793945,\n",
       "  5.854386806488037,\n",
       "  5.084052562713623,\n",
       "  5.2331366539001465,\n",
       "  5.563003063201904,\n",
       "  5.774744510650635,\n",
       "  5.848795413970947,\n",
       "  5.928769111633301,\n",
       "  5.096157550811768,\n",
       "  5.238570690155029,\n",
       "  5.887909412384033,\n",
       "  5.409764289855957,\n",
       "  4.40825891494751,\n",
       "  5.092676639556885,\n",
       "  5.556719779968262,\n",
       "  5.8568501472473145,\n",
       "  5.2536492347717285,\n",
       "  5.03596305847168,\n",
       "  6.178236484527588,\n",
       "  5.30592155456543,\n",
       "  4.868119716644287,\n",
       "  6.175187110900879,\n",
       "  4.663137912750244,\n",
       "  6.042108058929443,\n",
       "  4.847924709320068,\n",
       "  5.070185661315918,\n",
       "  5.196875095367432,\n",
       "  5.100446701049805,\n",
       "  5.394138813018799,\n",
       "  5.286220073699951,\n",
       "  6.424188137054443,\n",
       "  5.247403621673584,\n",
       "  5.182949066162109,\n",
       "  5.253805160522461,\n",
       "  5.048996925354004,\n",
       "  4.923771858215332,\n",
       "  4.96286153793335,\n",
       "  5.412273406982422,\n",
       "  5.554286956787109,\n",
       "  5.207097053527832,\n",
       "  6.07520055770874,\n",
       "  4.976685523986816,\n",
       "  5.081509113311768,\n",
       "  5.326171875,\n",
       "  5.4420928955078125,\n",
       "  4.5683159828186035,\n",
       "  5.917113780975342,\n",
       "  5.079334735870361,\n",
       "  5.563129425048828,\n",
       "  4.7679243087768555,\n",
       "  5.528578758239746,\n",
       "  3.8719534873962402,\n",
       "  4.750521659851074,\n",
       "  5.701635360717773,\n",
       "  5.742092609405518,\n",
       "  4.9177045822143555,\n",
       "  5.760986804962158,\n",
       "  5.261519908905029,\n",
       "  4.418517112731934,\n",
       "  5.210386276245117,\n",
       "  5.5773515701293945,\n",
       "  4.867336750030518,\n",
       "  4.988964080810547,\n",
       "  5.17173433303833,\n",
       "  5.866097450256348,\n",
       "  5.628702163696289,\n",
       "  4.99838924407959,\n",
       "  4.199275493621826,\n",
       "  5.159941673278809,\n",
       "  4.487574100494385,\n",
       "  5.1383819580078125,\n",
       "  5.056176662445068,\n",
       "  6.0571088790893555,\n",
       "  4.769680023193359,\n",
       "  5.787027835845947,\n",
       "  5.4376935958862305,\n",
       "  4.737827777862549,\n",
       "  4.605546951293945,\n",
       "  5.179082870483398,\n",
       "  4.475465297698975,\n",
       "  5.482834815979004,\n",
       "  5.2783522605896,\n",
       "  4.736172676086426,\n",
       "  4.037018775939941,\n",
       "  5.056282043457031,\n",
       "  5.561960220336914,\n",
       "  5.084578990936279,\n",
       "  5.3997321128845215,\n",
       "  5.606127738952637,\n",
       "  4.5007195472717285,\n",
       "  5.044325351715088,\n",
       "  5.243644714355469,\n",
       "  4.7694268226623535,\n",
       "  5.257894039154053,\n",
       "  4.9729180335998535,\n",
       "  4.75773811340332,\n",
       "  5.225069046020508,\n",
       "  5.1748738288879395,\n",
       "  4.633727073669434,\n",
       "  4.551016807556152,\n",
       "  4.309856414794922,\n",
       "  4.5417351722717285,\n",
       "  4.853643417358398,\n",
       "  5.10538387298584,\n",
       "  5.139662265777588,\n",
       "  5.258285999298096,\n",
       "  4.501973628997803,\n",
       "  5.008573055267334,\n",
       "  5.052850723266602,\n",
       "  4.951686382293701,\n",
       "  4.66091775894165,\n",
       "  4.803896903991699,\n",
       "  5.3039140701293945,\n",
       "  5.3527703285217285,\n",
       "  4.116580963134766,\n",
       "  5.58057975769043,\n",
       "  4.748333930969238,\n",
       "  5.256656169891357,\n",
       "  4.602118015289307,\n",
       "  4.942258834838867,\n",
       "  4.811962604522705,\n",
       "  4.09275484085083,\n",
       "  5.550209045410156,\n",
       "  5.157925128936768,\n",
       "  5.5393967628479,\n",
       "  4.451102256774902,\n",
       "  4.8215765953063965,\n",
       "  5.118683815002441,\n",
       "  4.489153861999512,\n",
       "  4.6989521980285645,\n",
       "  3.739699125289917,\n",
       "  4.49470853805542,\n",
       "  5.5609636306762695,\n",
       "  4.419373512268066,\n",
       "  4.440491676330566,\n",
       "  4.657710075378418,\n",
       "  4.60772180557251,\n",
       "  4.986242294311523,\n",
       "  4.736499786376953,\n",
       "  4.463547706604004,\n",
       "  4.913445472717285,\n",
       "  5.146964073181152,\n",
       "  4.443927764892578,\n",
       "  5.379999160766602,\n",
       "  4.616513252258301,\n",
       "  4.71842622756958,\n",
       "  4.884434700012207,\n",
       "  4.640573501586914,\n",
       "  5.0231428146362305,\n",
       "  4.71310567855835,\n",
       "  4.8584794998168945,\n",
       "  5.1766486167907715,\n",
       "  4.108058452606201,\n",
       "  5.680817127227783,\n",
       "  4.3322930335998535,\n",
       "  4.635502815246582,\n",
       "  4.548344135284424,\n",
       "  5.018511772155762,\n",
       "  4.727330207824707,\n",
       "  4.503907203674316,\n",
       "  4.908830165863037,\n",
       "  4.834928035736084,\n",
       "  4.2016143798828125,\n",
       "  4.617180824279785,\n",
       "  4.272310733795166,\n",
       "  4.630500316619873,\n",
       "  5.086462020874023],\n",
       " 'val_loss': [53.29063415527344,\n",
       "  28.729333877563477,\n",
       "  73.74231719970703,\n",
       "  14.930423736572266,\n",
       "  16.6611270904541,\n",
       "  25.584260940551758,\n",
       "  12.212021827697754,\n",
       "  34.78976821899414,\n",
       "  31.96874237060547,\n",
       "  11.07560920715332,\n",
       "  16.333162307739258,\n",
       "  33.531436920166016,\n",
       "  25.698453903198242,\n",
       "  8.87343978881836,\n",
       "  10.137455940246582,\n",
       "  13.857712745666504,\n",
       "  20.93545913696289,\n",
       "  33.559364318847656,\n",
       "  8.161493301391602,\n",
       "  7.541574478149414,\n",
       "  22.73699188232422,\n",
       "  11.979292869567871,\n",
       "  7.840504169464111,\n",
       "  15.88094711303711,\n",
       "  7.7266526222229,\n",
       "  8.780428886413574,\n",
       "  15.123218536376953,\n",
       "  8.278851509094238,\n",
       "  8.606639862060547,\n",
       "  14.460674285888672,\n",
       "  7.823577404022217,\n",
       "  8.514206886291504,\n",
       "  14.772841453552246,\n",
       "  6.962386608123779,\n",
       "  7.973356246948242,\n",
       "  7.212589740753174,\n",
       "  12.900840759277344,\n",
       "  20.092063903808594,\n",
       "  9.538280487060547,\n",
       "  14.824227333068848,\n",
       "  23.922029495239258,\n",
       "  11.771953582763672,\n",
       "  9.66606330871582,\n",
       "  11.81421184539795,\n",
       "  12.427289009094238,\n",
       "  28.765281677246094,\n",
       "  6.613513946533203,\n",
       "  6.795417308807373,\n",
       "  18.687822341918945,\n",
       "  7.428153038024902,\n",
       "  23.697717666625977,\n",
       "  6.289120674133301,\n",
       "  14.353482246398926,\n",
       "  7.210375785827637,\n",
       "  15.155774116516113,\n",
       "  8.745835304260254,\n",
       "  5.94697380065918,\n",
       "  6.29397439956665,\n",
       "  9.098139762878418,\n",
       "  8.417657852172852,\n",
       "  6.4495439529418945,\n",
       "  7.813793182373047,\n",
       "  8.834159851074219,\n",
       "  7.9834208488464355,\n",
       "  6.30343770980835,\n",
       "  11.534502029418945,\n",
       "  10.96074104309082,\n",
       "  6.0465264320373535,\n",
       "  6.96540641784668,\n",
       "  13.190866470336914,\n",
       "  14.850268363952637,\n",
       "  19.874168395996094,\n",
       "  5.530066013336182,\n",
       "  15.879683494567871,\n",
       "  11.640302658081055,\n",
       "  6.199772357940674,\n",
       "  9.355660438537598,\n",
       "  10.601536750793457,\n",
       "  8.643421173095703,\n",
       "  6.800670623779297,\n",
       "  11.259071350097656,\n",
       "  7.796281814575195,\n",
       "  4.798348426818848,\n",
       "  5.0271453857421875,\n",
       "  8.813859939575195,\n",
       "  11.850349426269531,\n",
       "  9.471029281616211,\n",
       "  14.701189041137695,\n",
       "  4.585611343383789,\n",
       "  5.0442795753479,\n",
       "  5.729183197021484,\n",
       "  6.1546759605407715,\n",
       "  17.441173553466797,\n",
       "  5.468178749084473,\n",
       "  5.965064525604248,\n",
       "  4.9601826667785645,\n",
       "  5.007981777191162,\n",
       "  9.91482162475586,\n",
       "  8.544275283813477,\n",
       "  17.501609802246094,\n",
       "  6.803743839263916,\n",
       "  6.927257061004639,\n",
       "  8.964740753173828,\n",
       "  4.140946388244629,\n",
       "  9.722648620605469,\n",
       "  14.429662704467773,\n",
       "  7.100452423095703,\n",
       "  10.917973518371582,\n",
       "  5.123836517333984,\n",
       "  5.362590312957764,\n",
       "  6.788051605224609,\n",
       "  7.16236686706543,\n",
       "  5.247028350830078,\n",
       "  4.397543430328369,\n",
       "  5.6444478034973145,\n",
       "  4.926358222961426,\n",
       "  4.14362907409668,\n",
       "  9.281618118286133,\n",
       "  5.188228130340576,\n",
       "  5.306182384490967,\n",
       "  4.898017883300781,\n",
       "  5.406959056854248,\n",
       "  8.37790298461914,\n",
       "  3.8549795150756836,\n",
       "  4.671802520751953,\n",
       "  3.649524211883545,\n",
       "  4.229393005371094,\n",
       "  5.9475417137146,\n",
       "  3.597731351852417,\n",
       "  3.68355655670166,\n",
       "  3.4296627044677734,\n",
       "  4.880404949188232,\n",
       "  3.1678147315979004,\n",
       "  2.838067054748535,\n",
       "  2.784727096557617,\n",
       "  3.216620445251465,\n",
       "  3.221193790435791,\n",
       "  5.236837863922119,\n",
       "  3.832801103591919,\n",
       "  5.398064136505127,\n",
       "  3.576542615890503,\n",
       "  12.899548530578613,\n",
       "  9.602356910705566,\n",
       "  3.1053943634033203,\n",
       "  6.087162971496582,\n",
       "  3.991111993789673,\n",
       "  3.607283353805542,\n",
       "  3.3493659496307373,\n",
       "  10.352433204650879,\n",
       "  6.882247447967529,\n",
       "  8.537496566772461,\n",
       "  6.414183139801025,\n",
       "  5.137759208679199,\n",
       "  5.6702704429626465,\n",
       "  4.158064365386963,\n",
       "  4.653733253479004,\n",
       "  4.621649742126465,\n",
       "  6.373188495635986,\n",
       "  5.953242301940918,\n",
       "  3.9594037532806396,\n",
       "  3.200650691986084,\n",
       "  2.9764962196350098,\n",
       "  2.624833583831787,\n",
       "  7.4724531173706055,\n",
       "  3.0964386463165283,\n",
       "  2.3794925212860107,\n",
       "  3.074812412261963,\n",
       "  3.39876127243042,\n",
       "  2.941572904586792,\n",
       "  5.148664951324463,\n",
       "  4.075094699859619,\n",
       "  5.937179088592529,\n",
       "  8.419109344482422,\n",
       "  4.318003177642822,\n",
       "  11.45469856262207,\n",
       "  2.7030107975006104,\n",
       "  5.3324480056762695,\n",
       "  3.5945024490356445,\n",
       "  3.30399227142334,\n",
       "  3.1367015838623047,\n",
       "  3.4417667388916016,\n",
       "  5.274641036987305,\n",
       "  5.034524440765381,\n",
       "  3.8655459880828857,\n",
       "  7.031256198883057,\n",
       "  3.9045379161834717,\n",
       "  8.798295021057129,\n",
       "  6.619394302368164,\n",
       "  3.3526434898376465,\n",
       "  6.523923873901367,\n",
       "  3.3073458671569824,\n",
       "  3.5587034225463867,\n",
       "  3.7147281169891357,\n",
       "  6.1608076095581055,\n",
       "  7.720147132873535,\n",
       "  7.203759670257568,\n",
       "  6.351304531097412,\n",
       "  5.034461975097656,\n",
       "  5.342652797698975,\n",
       "  6.084295272827148,\n",
       "  4.030337810516357,\n",
       "  4.173795700073242,\n",
       "  8.549942970275879,\n",
       "  3.685474157333374,\n",
       "  4.746246814727783,\n",
       "  9.059246063232422,\n",
       "  5.284670829772949,\n",
       "  10.440149307250977,\n",
       "  4.350831508636475,\n",
       "  5.212509632110596,\n",
       "  4.011274337768555,\n",
       "  3.8945584297180176,\n",
       "  8.712343215942383,\n",
       "  7.713636875152588,\n",
       "  9.582688331604004,\n",
       "  7.797133922576904,\n",
       "  11.21169662475586,\n",
       "  5.285256862640381,\n",
       "  6.559321403503418,\n",
       "  5.7567620277404785,\n",
       "  5.6533732414245605,\n",
       "  8.461394309997559,\n",
       "  10.705538749694824,\n",
       "  9.802725791931152,\n",
       "  5.764666557312012,\n",
       "  8.093265533447266,\n",
       "  7.911177158355713,\n",
       "  5.878359317779541,\n",
       "  9.609759330749512,\n",
       "  5.187238693237305,\n",
       "  6.326935291290283,\n",
       "  10.5763521194458,\n",
       "  5.655768871307373,\n",
       "  6.325952529907227,\n",
       "  12.502849578857422,\n",
       "  9.050594329833984,\n",
       "  10.220783233642578,\n",
       "  10.016122817993164,\n",
       "  9.185370445251465,\n",
       "  6.388772010803223,\n",
       "  6.77215576171875,\n",
       "  10.117025375366211,\n",
       "  7.4779534339904785,\n",
       "  7.260883808135986,\n",
       "  7.2789812088012695,\n",
       "  8.129229545593262,\n",
       "  4.511934280395508,\n",
       "  5.189913749694824,\n",
       "  12.535602569580078,\n",
       "  11.968770980834961,\n",
       "  8.3356351852417,\n",
       "  5.870573043823242,\n",
       "  8.447949409484863,\n",
       "  9.115769386291504,\n",
       "  11.902510643005371,\n",
       "  16.0690975189209,\n",
       "  12.185351371765137,\n",
       "  12.978960037231445,\n",
       "  9.969937324523926,\n",
       "  6.353248119354248,\n",
       "  5.671882152557373,\n",
       "  6.3890275955200195,\n",
       "  6.010112762451172,\n",
       "  6.607509136199951,\n",
       "  6.0255818367004395,\n",
       "  9.8319730758667,\n",
       "  9.684510231018066,\n",
       "  11.791390419006348,\n",
       "  7.6150007247924805,\n",
       "  6.673311710357666,\n",
       "  9.186190605163574,\n",
       "  5.8535943031311035,\n",
       "  7.330247402191162,\n",
       "  7.429415225982666,\n",
       "  10.787637710571289,\n",
       "  9.648706436157227,\n",
       "  8.963693618774414,\n",
       "  7.033394813537598,\n",
       "  6.8082709312438965,\n",
       "  7.592709541320801,\n",
       "  13.894222259521484,\n",
       "  12.939031600952148,\n",
       "  10.938892364501953,\n",
       "  8.221512794494629,\n",
       "  12.129683494567871,\n",
       "  12.814980506896973,\n",
       "  8.818940162658691,\n",
       "  10.205672264099121,\n",
       "  13.634321212768555,\n",
       "  6.574211597442627,\n",
       "  8.68920612335205,\n",
       "  5.887176036834717,\n",
       "  8.78370189666748,\n",
       "  11.800922393798828,\n",
       "  11.288544654846191,\n",
       "  13.345771789550781,\n",
       "  15.044912338256836,\n",
       "  8.145452499389648,\n",
       "  14.633234977722168,\n",
       "  18.18604850769043,\n",
       "  13.224967956542969,\n",
       "  12.3259916305542,\n",
       "  8.251152038574219,\n",
       "  11.457542419433594,\n",
       "  9.517267227172852,\n",
       "  7.465667247772217,\n",
       "  7.717828750610352,\n",
       "  17.81892967224121,\n",
       "  12.521821022033691,\n",
       "  13.225926399230957,\n",
       "  11.716719627380371,\n",
       "  9.948308944702148,\n",
       "  8.787873268127441,\n",
       "  16.90575408935547,\n",
       "  15.238066673278809,\n",
       "  11.649158477783203,\n",
       "  9.423333168029785,\n",
       "  16.20494842529297,\n",
       "  13.606237411499023,\n",
       "  11.720756530761719,\n",
       "  7.944421768188477,\n",
       "  12.124920845031738,\n",
       "  13.01611614227295,\n",
       "  10.710526466369629,\n",
       "  10.10244369506836,\n",
       "  8.65882682800293,\n",
       "  14.500387191772461,\n",
       "  11.473392486572266,\n",
       "  10.522632598876953,\n",
       "  13.595331192016602,\n",
       "  8.780606269836426,\n",
       "  7.199328422546387,\n",
       "  18.157869338989258,\n",
       "  9.760110855102539,\n",
       "  10.221796035766602,\n",
       "  10.834941864013672,\n",
       "  9.916707992553711,\n",
       "  12.55569076538086,\n",
       "  12.053567886352539,\n",
       "  14.637443542480469,\n",
       "  10.708627700805664,\n",
       "  11.821877479553223,\n",
       "  7.82381534576416,\n",
       "  9.02008056640625,\n",
       "  12.428248405456543,\n",
       "  10.427868843078613,\n",
       "  9.180992126464844,\n",
       "  15.490019798278809,\n",
       "  8.924798011779785,\n",
       "  16.733074188232422,\n",
       "  13.518442153930664,\n",
       "  14.095406532287598,\n",
       "  12.997623443603516,\n",
       "  11.81186294555664,\n",
       "  12.0254545211792,\n",
       "  12.366233825683594,\n",
       "  16.758434295654297,\n",
       "  12.852492332458496,\n",
       "  15.346632957458496,\n",
       "  10.718253135681152,\n",
       "  11.630565643310547,\n",
       "  12.395066261291504,\n",
       "  9.969108581542969,\n",
       "  10.683416366577148,\n",
       "  14.373713493347168,\n",
       "  11.166089057922363,\n",
       "  15.048149108886719,\n",
       "  10.914427757263184,\n",
       "  11.008566856384277,\n",
       "  16.524808883666992,\n",
       "  9.52989387512207,\n",
       "  13.659111022949219,\n",
       "  11.364513397216797,\n",
       "  10.731497764587402,\n",
       "  13.177482604980469,\n",
       "  11.425951957702637,\n",
       "  11.892483711242676,\n",
       "  15.177541732788086,\n",
       "  15.538658142089844,\n",
       "  14.503735542297363,\n",
       "  13.097733497619629,\n",
       "  17.230712890625,\n",
       "  14.247814178466797,\n",
       "  12.540838241577148,\n",
       "  8.408023834228516,\n",
       "  12.776668548583984,\n",
       "  10.970606803894043,\n",
       "  18.544679641723633,\n",
       "  10.123261451721191,\n",
       "  14.510218620300293,\n",
       "  12.040958404541016,\n",
       "  12.768497467041016,\n",
       "  11.565390586853027,\n",
       "  8.987548828125,\n",
       "  17.891056060791016,\n",
       "  15.437291145324707,\n",
       "  9.891288757324219,\n",
       "  9.960949897766113,\n",
       "  15.0935697555542,\n",
       "  10.571636199951172,\n",
       "  12.931386947631836,\n",
       "  15.435927391052246,\n",
       "  17.060937881469727,\n",
       "  18.26757049560547,\n",
       "  16.868867874145508,\n",
       "  11.366254806518555,\n",
       "  15.403498649597168,\n",
       "  18.506511688232422,\n",
       "  12.378643035888672,\n",
       "  14.333457946777344,\n",
       "  10.917391777038574,\n",
       "  13.605046272277832,\n",
       "  12.244959831237793,\n",
       "  18.946327209472656,\n",
       "  17.228862762451172,\n",
       "  12.048527717590332,\n",
       "  15.411198616027832,\n",
       "  11.985538482666016,\n",
       "  15.993197441101074,\n",
       "  11.526385307312012,\n",
       "  13.709606170654297,\n",
       "  15.633078575134277,\n",
       "  18.82403564453125,\n",
       "  15.651449203491211,\n",
       "  14.36054801940918,\n",
       "  11.938127517700195,\n",
       "  16.640409469604492,\n",
       "  11.436578750610352,\n",
       "  12.57502555847168,\n",
       "  11.681038856506348,\n",
       "  16.931385040283203,\n",
       "  11.755975723266602,\n",
       "  13.179415702819824,\n",
       "  11.189959526062012,\n",
       "  13.909457206726074,\n",
       "  15.777192115783691,\n",
       "  17.26078224182129,\n",
       "  11.841290473937988,\n",
       "  13.378271102905273,\n",
       "  10.057754516601562,\n",
       "  11.361461639404297,\n",
       "  14.711077690124512,\n",
       "  10.431571006774902,\n",
       "  15.458518028259277,\n",
       "  11.492165565490723,\n",
       "  16.817487716674805,\n",
       "  12.8748140335083,\n",
       "  15.834193229675293,\n",
       "  14.84949016571045,\n",
       "  16.444805145263672,\n",
       "  16.02796173095703,\n",
       "  14.985795974731445,\n",
       "  16.310108184814453,\n",
       "  13.030791282653809,\n",
       "  9.496703147888184,\n",
       "  14.44662094116211,\n",
       "  12.254014015197754,\n",
       "  13.012238502502441,\n",
       "  14.838573455810547,\n",
       "  14.050694465637207,\n",
       "  15.923832893371582,\n",
       "  10.404012680053711,\n",
       "  19.014724731445312,\n",
       "  16.47197723388672,\n",
       "  11.006677627563477,\n",
       "  13.775691986083984,\n",
       "  13.782129287719727,\n",
       "  12.64394760131836,\n",
       "  12.900748252868652,\n",
       "  15.51565933227539,\n",
       "  10.939562797546387,\n",
       "  16.30808448791504,\n",
       "  15.790104866027832,\n",
       "  16.298105239868164,\n",
       "  15.192822456359863,\n",
       "  16.047006607055664,\n",
       "  12.716517448425293,\n",
       "  14.66356086730957,\n",
       "  14.455521583557129,\n",
       "  16.732624053955078,\n",
       "  12.49001693725586,\n",
       "  13.783480644226074,\n",
       "  16.720989227294922,\n",
       "  11.364859580993652,\n",
       "  10.645459175109863,\n",
       "  12.813026428222656,\n",
       "  20.205732345581055,\n",
       "  18.7498836517334,\n",
       "  16.181859970092773,\n",
       "  15.501730918884277,\n",
       "  15.355228424072266,\n",
       "  17.246244430541992,\n",
       "  15.458952903747559,\n",
       "  11.356783866882324,\n",
       "  13.792675018310547,\n",
       "  10.36743450164795,\n",
       "  15.136687278747559,\n",
       "  14.922197341918945,\n",
       "  15.698395729064941,\n",
       "  18.410669326782227,\n",
       "  18.06251335144043,\n",
       "  16.00379753112793,\n",
       "  12.421723365783691,\n",
       "  19.066822052001953,\n",
       "  11.961528778076172,\n",
       "  16.99488067626953,\n",
       "  15.374433517456055,\n",
       "  20.03278160095215,\n",
       "  19.13322639465332,\n",
       "  13.232514381408691,\n",
       "  14.010061264038086,\n",
       "  14.3250150680542,\n",
       "  15.670807838439941,\n",
       "  15.730683326721191,\n",
       "  17.300939559936523,\n",
       "  12.158812522888184,\n",
       "  19.27826499938965,\n",
       "  15.772296905517578,\n",
       "  15.603034973144531,\n",
       "  19.026382446289062,\n",
       "  16.408191680908203,\n",
       "  11.674939155578613,\n",
       "  13.93232536315918,\n",
       "  17.452781677246094,\n",
       "  15.961139678955078,\n",
       "  19.018733978271484,\n",
       "  20.238128662109375,\n",
       "  14.482071876525879,\n",
       "  13.913093566894531,\n",
       "  13.902453422546387,\n",
       "  12.190146446228027,\n",
       "  20.5694522857666,\n",
       "  19.502967834472656,\n",
       "  16.5478515625,\n",
       "  15.072381973266602,\n",
       "  14.918322563171387,\n",
       "  17.8337345123291,\n",
       "  22.805898666381836,\n",
       "  18.34781837463379,\n",
       "  12.3644380569458,\n",
       "  15.200214385986328,\n",
       "  15.548871040344238,\n",
       "  12.733726501464844,\n",
       "  14.018355369567871,\n",
       "  17.40847396850586,\n",
       "  18.26883888244629,\n",
       "  19.084081649780273,\n",
       "  16.935874938964844,\n",
       "  15.323038101196289,\n",
       "  12.737046241760254,\n",
       "  19.655603408813477,\n",
       "  16.21343231201172,\n",
       "  16.363069534301758,\n",
       "  17.237621307373047,\n",
       "  14.241643905639648,\n",
       "  17.590972900390625,\n",
       "  12.066545486450195,\n",
       "  16.5583438873291,\n",
       "  13.632301330566406,\n",
       "  16.79709815979004,\n",
       "  16.148303985595703,\n",
       "  17.05228614807129,\n",
       "  16.614315032958984,\n",
       "  13.849008560180664,\n",
       "  13.663151741027832,\n",
       "  16.788711547851562,\n",
       "  18.025480270385742,\n",
       "  14.727730751037598,\n",
       "  17.904726028442383,\n",
       "  14.93217658996582,\n",
       "  17.843778610229492,\n",
       "  13.406999588012695,\n",
       "  16.770221710205078,\n",
       "  16.22726821899414,\n",
       "  17.874908447265625,\n",
       "  14.863337516784668,\n",
       "  14.212210655212402,\n",
       "  12.613410949707031,\n",
       "  18.830768585205078,\n",
       "  16.59371566772461,\n",
       "  16.505680084228516,\n",
       "  14.951499938964844,\n",
       "  15.670379638671875,\n",
       "  15.508214950561523,\n",
       "  15.576377868652344,\n",
       "  18.457630157470703,\n",
       "  12.460521697998047,\n",
       "  21.027820587158203,\n",
       "  13.751361846923828,\n",
       "  17.88991928100586,\n",
       "  11.904324531555176,\n",
       "  15.674676895141602,\n",
       "  17.298019409179688,\n",
       "  11.769865989685059,\n",
       "  12.800402641296387,\n",
       "  11.407265663146973,\n",
       "  13.03292465209961,\n",
       "  14.232281684875488,\n",
       "  15.572345733642578,\n",
       "  12.51910400390625],\n",
       " 'val_mape': [53.29063415527344,\n",
       "  28.729333877563477,\n",
       "  73.74231719970703,\n",
       "  14.930423736572266,\n",
       "  16.6611270904541,\n",
       "  25.584260940551758,\n",
       "  12.212021827697754,\n",
       "  34.78976821899414,\n",
       "  31.96874237060547,\n",
       "  11.07560920715332,\n",
       "  16.333162307739258,\n",
       "  33.531436920166016,\n",
       "  25.698453903198242,\n",
       "  8.87343978881836,\n",
       "  10.137455940246582,\n",
       "  13.857712745666504,\n",
       "  20.93545913696289,\n",
       "  33.559364318847656,\n",
       "  8.161493301391602,\n",
       "  7.541574478149414,\n",
       "  22.73699188232422,\n",
       "  11.979292869567871,\n",
       "  7.840504169464111,\n",
       "  15.88094711303711,\n",
       "  7.7266526222229,\n",
       "  8.780428886413574,\n",
       "  15.123218536376953,\n",
       "  8.278851509094238,\n",
       "  8.606639862060547,\n",
       "  14.460674285888672,\n",
       "  7.823577404022217,\n",
       "  8.514206886291504,\n",
       "  14.772841453552246,\n",
       "  6.962386608123779,\n",
       "  7.973356246948242,\n",
       "  7.212589740753174,\n",
       "  12.900840759277344,\n",
       "  20.092063903808594,\n",
       "  9.538280487060547,\n",
       "  14.824227333068848,\n",
       "  23.922029495239258,\n",
       "  11.771953582763672,\n",
       "  9.66606330871582,\n",
       "  11.81421184539795,\n",
       "  12.427289009094238,\n",
       "  28.765281677246094,\n",
       "  6.613513946533203,\n",
       "  6.795417308807373,\n",
       "  18.687822341918945,\n",
       "  7.428153038024902,\n",
       "  23.697717666625977,\n",
       "  6.289120674133301,\n",
       "  14.353482246398926,\n",
       "  7.210375785827637,\n",
       "  15.155774116516113,\n",
       "  8.745835304260254,\n",
       "  5.94697380065918,\n",
       "  6.29397439956665,\n",
       "  9.098139762878418,\n",
       "  8.417657852172852,\n",
       "  6.4495439529418945,\n",
       "  7.813793182373047,\n",
       "  8.834159851074219,\n",
       "  7.9834208488464355,\n",
       "  6.30343770980835,\n",
       "  11.534502029418945,\n",
       "  10.96074104309082,\n",
       "  6.0465264320373535,\n",
       "  6.96540641784668,\n",
       "  13.190866470336914,\n",
       "  14.850268363952637,\n",
       "  19.874168395996094,\n",
       "  5.530066013336182,\n",
       "  15.879683494567871,\n",
       "  11.640302658081055,\n",
       "  6.199772357940674,\n",
       "  9.355660438537598,\n",
       "  10.601536750793457,\n",
       "  8.643421173095703,\n",
       "  6.800670623779297,\n",
       "  11.259071350097656,\n",
       "  7.796281814575195,\n",
       "  4.798348426818848,\n",
       "  5.0271453857421875,\n",
       "  8.813859939575195,\n",
       "  11.850349426269531,\n",
       "  9.471029281616211,\n",
       "  14.701189041137695,\n",
       "  4.585611343383789,\n",
       "  5.0442795753479,\n",
       "  5.729183197021484,\n",
       "  6.1546759605407715,\n",
       "  17.441173553466797,\n",
       "  5.468178749084473,\n",
       "  5.965064525604248,\n",
       "  4.9601826667785645,\n",
       "  5.007981777191162,\n",
       "  9.91482162475586,\n",
       "  8.544275283813477,\n",
       "  17.501609802246094,\n",
       "  6.803743839263916,\n",
       "  6.927257061004639,\n",
       "  8.964740753173828,\n",
       "  4.140946388244629,\n",
       "  9.722648620605469,\n",
       "  14.429662704467773,\n",
       "  7.100452423095703,\n",
       "  10.917973518371582,\n",
       "  5.123836517333984,\n",
       "  5.362590312957764,\n",
       "  6.788051605224609,\n",
       "  7.16236686706543,\n",
       "  5.247028350830078,\n",
       "  4.397543430328369,\n",
       "  5.6444478034973145,\n",
       "  4.926358222961426,\n",
       "  4.14362907409668,\n",
       "  9.281618118286133,\n",
       "  5.188228130340576,\n",
       "  5.306182384490967,\n",
       "  4.898017883300781,\n",
       "  5.406959056854248,\n",
       "  8.37790298461914,\n",
       "  3.8549795150756836,\n",
       "  4.671802520751953,\n",
       "  3.649524211883545,\n",
       "  4.229393005371094,\n",
       "  5.9475417137146,\n",
       "  3.597731351852417,\n",
       "  3.68355655670166,\n",
       "  3.4296627044677734,\n",
       "  4.880404949188232,\n",
       "  3.1678147315979004,\n",
       "  2.838067054748535,\n",
       "  2.784727096557617,\n",
       "  3.216620445251465,\n",
       "  3.221193790435791,\n",
       "  5.236837863922119,\n",
       "  3.832801103591919,\n",
       "  5.398064136505127,\n",
       "  3.576542615890503,\n",
       "  12.899548530578613,\n",
       "  9.602356910705566,\n",
       "  3.1053943634033203,\n",
       "  6.087162971496582,\n",
       "  3.991111993789673,\n",
       "  3.607283353805542,\n",
       "  3.3493659496307373,\n",
       "  10.352433204650879,\n",
       "  6.882247447967529,\n",
       "  8.537496566772461,\n",
       "  6.414183139801025,\n",
       "  5.137759208679199,\n",
       "  5.6702704429626465,\n",
       "  4.158064365386963,\n",
       "  4.653733253479004,\n",
       "  4.621649742126465,\n",
       "  6.373188495635986,\n",
       "  5.953242301940918,\n",
       "  3.9594037532806396,\n",
       "  3.200650691986084,\n",
       "  2.9764962196350098,\n",
       "  2.624833583831787,\n",
       "  7.4724531173706055,\n",
       "  3.0964386463165283,\n",
       "  2.3794925212860107,\n",
       "  3.074812412261963,\n",
       "  3.39876127243042,\n",
       "  2.941572904586792,\n",
       "  5.148664951324463,\n",
       "  4.075094699859619,\n",
       "  5.937179088592529,\n",
       "  8.419109344482422,\n",
       "  4.318003177642822,\n",
       "  11.45469856262207,\n",
       "  2.7030107975006104,\n",
       "  5.3324480056762695,\n",
       "  3.5945024490356445,\n",
       "  3.30399227142334,\n",
       "  3.1367015838623047,\n",
       "  3.4417667388916016,\n",
       "  5.274641036987305,\n",
       "  5.034524440765381,\n",
       "  3.8655459880828857,\n",
       "  7.031256198883057,\n",
       "  3.9045379161834717,\n",
       "  8.798295021057129,\n",
       "  6.619394302368164,\n",
       "  3.3526434898376465,\n",
       "  6.523923873901367,\n",
       "  3.3073458671569824,\n",
       "  3.5587034225463867,\n",
       "  3.7147281169891357,\n",
       "  6.1608076095581055,\n",
       "  7.720147132873535,\n",
       "  7.203759670257568,\n",
       "  6.351304531097412,\n",
       "  5.034461975097656,\n",
       "  5.342652797698975,\n",
       "  6.084295272827148,\n",
       "  4.030337810516357,\n",
       "  4.173795700073242,\n",
       "  8.549942970275879,\n",
       "  3.685474157333374,\n",
       "  4.746246814727783,\n",
       "  9.059246063232422,\n",
       "  5.284670829772949,\n",
       "  10.440149307250977,\n",
       "  4.350831508636475,\n",
       "  5.212509632110596,\n",
       "  4.011274337768555,\n",
       "  3.8945584297180176,\n",
       "  8.712343215942383,\n",
       "  7.713636875152588,\n",
       "  9.582688331604004,\n",
       "  7.797133922576904,\n",
       "  11.21169662475586,\n",
       "  5.285256862640381,\n",
       "  6.559321403503418,\n",
       "  5.7567620277404785,\n",
       "  5.6533732414245605,\n",
       "  8.461394309997559,\n",
       "  10.705538749694824,\n",
       "  9.802725791931152,\n",
       "  5.764666557312012,\n",
       "  8.093265533447266,\n",
       "  7.911177158355713,\n",
       "  5.878359317779541,\n",
       "  9.609759330749512,\n",
       "  5.187238693237305,\n",
       "  6.326935291290283,\n",
       "  10.5763521194458,\n",
       "  5.655768871307373,\n",
       "  6.325952529907227,\n",
       "  12.502849578857422,\n",
       "  9.050594329833984,\n",
       "  10.220783233642578,\n",
       "  10.016122817993164,\n",
       "  9.185370445251465,\n",
       "  6.388772010803223,\n",
       "  6.77215576171875,\n",
       "  10.117025375366211,\n",
       "  7.4779534339904785,\n",
       "  7.260883808135986,\n",
       "  7.2789812088012695,\n",
       "  8.129229545593262,\n",
       "  4.511934280395508,\n",
       "  5.189913749694824,\n",
       "  12.535602569580078,\n",
       "  11.968770980834961,\n",
       "  8.3356351852417,\n",
       "  5.870573043823242,\n",
       "  8.447949409484863,\n",
       "  9.115769386291504,\n",
       "  11.902510643005371,\n",
       "  16.0690975189209,\n",
       "  12.185351371765137,\n",
       "  12.978960037231445,\n",
       "  9.969937324523926,\n",
       "  6.353248119354248,\n",
       "  5.671882152557373,\n",
       "  6.3890275955200195,\n",
       "  6.010112762451172,\n",
       "  6.607509136199951,\n",
       "  6.0255818367004395,\n",
       "  9.8319730758667,\n",
       "  9.684510231018066,\n",
       "  11.791390419006348,\n",
       "  7.6150007247924805,\n",
       "  6.673311710357666,\n",
       "  9.186190605163574,\n",
       "  5.8535943031311035,\n",
       "  7.330247402191162,\n",
       "  7.429415225982666,\n",
       "  10.787637710571289,\n",
       "  9.648706436157227,\n",
       "  8.963693618774414,\n",
       "  7.033394813537598,\n",
       "  6.8082709312438965,\n",
       "  7.592709541320801,\n",
       "  13.894222259521484,\n",
       "  12.939031600952148,\n",
       "  10.938892364501953,\n",
       "  8.221512794494629,\n",
       "  12.129683494567871,\n",
       "  12.814980506896973,\n",
       "  8.818940162658691,\n",
       "  10.205672264099121,\n",
       "  13.634321212768555,\n",
       "  6.574211597442627,\n",
       "  8.68920612335205,\n",
       "  5.887176036834717,\n",
       "  8.78370189666748,\n",
       "  11.800922393798828,\n",
       "  11.288544654846191,\n",
       "  13.345771789550781,\n",
       "  15.044912338256836,\n",
       "  8.145452499389648,\n",
       "  14.633234977722168,\n",
       "  18.18604850769043,\n",
       "  13.224967956542969,\n",
       "  12.3259916305542,\n",
       "  8.251152038574219,\n",
       "  11.457542419433594,\n",
       "  9.517267227172852,\n",
       "  7.465667247772217,\n",
       "  7.717828750610352,\n",
       "  17.81892967224121,\n",
       "  12.521821022033691,\n",
       "  13.225926399230957,\n",
       "  11.716719627380371,\n",
       "  9.948308944702148,\n",
       "  8.787873268127441,\n",
       "  16.90575408935547,\n",
       "  15.238066673278809,\n",
       "  11.649158477783203,\n",
       "  9.423333168029785,\n",
       "  16.20494842529297,\n",
       "  13.606237411499023,\n",
       "  11.720756530761719,\n",
       "  7.944421768188477,\n",
       "  12.124920845031738,\n",
       "  13.01611614227295,\n",
       "  10.710526466369629,\n",
       "  10.10244369506836,\n",
       "  8.65882682800293,\n",
       "  14.500387191772461,\n",
       "  11.473392486572266,\n",
       "  10.522632598876953,\n",
       "  13.595331192016602,\n",
       "  8.780606269836426,\n",
       "  7.199328422546387,\n",
       "  18.157869338989258,\n",
       "  9.760110855102539,\n",
       "  10.221796035766602,\n",
       "  10.834941864013672,\n",
       "  9.916707992553711,\n",
       "  12.55569076538086,\n",
       "  12.053567886352539,\n",
       "  14.637443542480469,\n",
       "  10.708627700805664,\n",
       "  11.821877479553223,\n",
       "  7.82381534576416,\n",
       "  9.02008056640625,\n",
       "  12.428248405456543,\n",
       "  10.427868843078613,\n",
       "  9.180992126464844,\n",
       "  15.490019798278809,\n",
       "  8.924798011779785,\n",
       "  16.733074188232422,\n",
       "  13.518442153930664,\n",
       "  14.095406532287598,\n",
       "  12.997623443603516,\n",
       "  11.81186294555664,\n",
       "  12.0254545211792,\n",
       "  12.366233825683594,\n",
       "  16.758434295654297,\n",
       "  12.852492332458496,\n",
       "  15.346632957458496,\n",
       "  10.718253135681152,\n",
       "  11.630565643310547,\n",
       "  12.395066261291504,\n",
       "  9.969108581542969,\n",
       "  10.683416366577148,\n",
       "  14.373713493347168,\n",
       "  11.166089057922363,\n",
       "  15.048149108886719,\n",
       "  10.914427757263184,\n",
       "  11.008566856384277,\n",
       "  16.524808883666992,\n",
       "  9.52989387512207,\n",
       "  13.659111022949219,\n",
       "  11.364513397216797,\n",
       "  10.731497764587402,\n",
       "  13.177482604980469,\n",
       "  11.425951957702637,\n",
       "  11.892483711242676,\n",
       "  15.177541732788086,\n",
       "  15.538658142089844,\n",
       "  14.503735542297363,\n",
       "  13.097733497619629,\n",
       "  17.230712890625,\n",
       "  14.247814178466797,\n",
       "  12.540838241577148,\n",
       "  8.408023834228516,\n",
       "  12.776668548583984,\n",
       "  10.970606803894043,\n",
       "  18.544679641723633,\n",
       "  10.123261451721191,\n",
       "  14.510218620300293,\n",
       "  12.040958404541016,\n",
       "  12.768497467041016,\n",
       "  11.565390586853027,\n",
       "  8.987548828125,\n",
       "  17.891056060791016,\n",
       "  15.437291145324707,\n",
       "  9.891288757324219,\n",
       "  9.960949897766113,\n",
       "  15.0935697555542,\n",
       "  10.571636199951172,\n",
       "  12.931386947631836,\n",
       "  15.435927391052246,\n",
       "  17.060937881469727,\n",
       "  18.26757049560547,\n",
       "  16.868867874145508,\n",
       "  11.366254806518555,\n",
       "  15.403498649597168,\n",
       "  18.506511688232422,\n",
       "  12.378643035888672,\n",
       "  14.333457946777344,\n",
       "  10.917391777038574,\n",
       "  13.605046272277832,\n",
       "  12.244959831237793,\n",
       "  18.946327209472656,\n",
       "  17.228862762451172,\n",
       "  12.048527717590332,\n",
       "  15.411198616027832,\n",
       "  11.985538482666016,\n",
       "  15.993197441101074,\n",
       "  11.526385307312012,\n",
       "  13.709606170654297,\n",
       "  15.633078575134277,\n",
       "  18.82403564453125,\n",
       "  15.651449203491211,\n",
       "  14.36054801940918,\n",
       "  11.938127517700195,\n",
       "  16.640409469604492,\n",
       "  11.436578750610352,\n",
       "  12.57502555847168,\n",
       "  11.681038856506348,\n",
       "  16.931385040283203,\n",
       "  11.755975723266602,\n",
       "  13.179415702819824,\n",
       "  11.189959526062012,\n",
       "  13.909457206726074,\n",
       "  15.777192115783691,\n",
       "  17.26078224182129,\n",
       "  11.841290473937988,\n",
       "  13.378271102905273,\n",
       "  10.057754516601562,\n",
       "  11.361461639404297,\n",
       "  14.711077690124512,\n",
       "  10.431571006774902,\n",
       "  15.458518028259277,\n",
       "  11.492165565490723,\n",
       "  16.817487716674805,\n",
       "  12.8748140335083,\n",
       "  15.834193229675293,\n",
       "  14.84949016571045,\n",
       "  16.444805145263672,\n",
       "  16.02796173095703,\n",
       "  14.985795974731445,\n",
       "  16.310108184814453,\n",
       "  13.030791282653809,\n",
       "  9.496703147888184,\n",
       "  14.44662094116211,\n",
       "  12.254014015197754,\n",
       "  13.012238502502441,\n",
       "  14.838573455810547,\n",
       "  14.050694465637207,\n",
       "  15.923832893371582,\n",
       "  10.404012680053711,\n",
       "  19.014724731445312,\n",
       "  16.47197723388672,\n",
       "  11.006677627563477,\n",
       "  13.775691986083984,\n",
       "  13.782129287719727,\n",
       "  12.64394760131836,\n",
       "  12.900748252868652,\n",
       "  15.51565933227539,\n",
       "  10.939562797546387,\n",
       "  16.30808448791504,\n",
       "  15.790104866027832,\n",
       "  16.298105239868164,\n",
       "  15.192822456359863,\n",
       "  16.047006607055664,\n",
       "  12.716517448425293,\n",
       "  14.66356086730957,\n",
       "  14.455521583557129,\n",
       "  16.732624053955078,\n",
       "  12.49001693725586,\n",
       "  13.783480644226074,\n",
       "  16.720989227294922,\n",
       "  11.364859580993652,\n",
       "  10.645459175109863,\n",
       "  12.813026428222656,\n",
       "  20.205732345581055,\n",
       "  18.7498836517334,\n",
       "  16.181859970092773,\n",
       "  15.501730918884277,\n",
       "  15.355228424072266,\n",
       "  17.246244430541992,\n",
       "  15.458952903747559,\n",
       "  11.356783866882324,\n",
       "  13.792675018310547,\n",
       "  10.36743450164795,\n",
       "  15.136687278747559,\n",
       "  14.922197341918945,\n",
       "  15.698395729064941,\n",
       "  18.410669326782227,\n",
       "  18.06251335144043,\n",
       "  16.00379753112793,\n",
       "  12.421723365783691,\n",
       "  19.066822052001953,\n",
       "  11.961528778076172,\n",
       "  16.99488067626953,\n",
       "  15.374433517456055,\n",
       "  20.03278160095215,\n",
       "  19.13322639465332,\n",
       "  13.232514381408691,\n",
       "  14.010061264038086,\n",
       "  14.3250150680542,\n",
       "  15.670807838439941,\n",
       "  15.730683326721191,\n",
       "  17.300939559936523,\n",
       "  12.158812522888184,\n",
       "  19.27826499938965,\n",
       "  15.772296905517578,\n",
       "  15.603034973144531,\n",
       "  19.026382446289062,\n",
       "  16.408191680908203,\n",
       "  11.674939155578613,\n",
       "  13.93232536315918,\n",
       "  17.452781677246094,\n",
       "  15.961139678955078,\n",
       "  19.018733978271484,\n",
       "  20.238128662109375,\n",
       "  14.482071876525879,\n",
       "  13.913093566894531,\n",
       "  13.902453422546387,\n",
       "  12.190146446228027,\n",
       "  20.5694522857666,\n",
       "  19.502967834472656,\n",
       "  16.5478515625,\n",
       "  15.072381973266602,\n",
       "  14.918322563171387,\n",
       "  17.8337345123291,\n",
       "  22.805898666381836,\n",
       "  18.34781837463379,\n",
       "  12.3644380569458,\n",
       "  15.200214385986328,\n",
       "  15.548871040344238,\n",
       "  12.733726501464844,\n",
       "  14.018355369567871,\n",
       "  17.40847396850586,\n",
       "  18.26883888244629,\n",
       "  19.084081649780273,\n",
       "  16.935874938964844,\n",
       "  15.323038101196289,\n",
       "  12.737046241760254,\n",
       "  19.655603408813477,\n",
       "  16.21343231201172,\n",
       "  16.363069534301758,\n",
       "  17.237621307373047,\n",
       "  14.241643905639648,\n",
       "  17.590972900390625,\n",
       "  12.066545486450195,\n",
       "  16.5583438873291,\n",
       "  13.632301330566406,\n",
       "  16.79709815979004,\n",
       "  16.148303985595703,\n",
       "  17.05228614807129,\n",
       "  16.614315032958984,\n",
       "  13.849008560180664,\n",
       "  13.663151741027832,\n",
       "  16.788711547851562,\n",
       "  18.025480270385742,\n",
       "  14.727730751037598,\n",
       "  17.904726028442383,\n",
       "  14.93217658996582,\n",
       "  17.843778610229492,\n",
       "  13.406999588012695,\n",
       "  16.770221710205078,\n",
       "  16.22726821899414,\n",
       "  17.874908447265625,\n",
       "  14.863337516784668,\n",
       "  14.212210655212402,\n",
       "  12.613410949707031,\n",
       "  18.830768585205078,\n",
       "  16.59371566772461,\n",
       "  16.505680084228516,\n",
       "  14.951499938964844,\n",
       "  15.670379638671875,\n",
       "  15.508214950561523,\n",
       "  15.576377868652344,\n",
       "  18.457630157470703,\n",
       "  12.460521697998047,\n",
       "  21.027820587158203,\n",
       "  13.751361846923828,\n",
       "  17.88991928100586,\n",
       "  11.904324531555176,\n",
       "  15.674676895141602,\n",
       "  17.298019409179688,\n",
       "  11.769865989685059,\n",
       "  12.800402641296387,\n",
       "  11.407265663146973,\n",
       "  13.03292465209961,\n",
       "  14.232281684875488,\n",
       "  15.572345733642578,\n",
       "  12.51910400390625]}"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model.fit(X_train, y_train, epochs=600, validation_data = (X_val,y_val), batch_size=32, verbose=0, callbacks=[cp_n]).history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction & Performance\n",
    "\n",
    "In this section, the trained model is used for predicting values of the Training-,Validation-, and Test-Set. They are depicted seperately. In the final part, the three graphs are combined and presented with their respective metrics of mean-average-percentage-error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import load_model\n",
    "model_n = load_model('model_n/')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training Data-Set isolated View"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/8 [==>...........................] - ETA: 0s"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8/8 [==============================] - 0s 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=float32, numpy=19.858633>"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_predictions = model_n.predict(X_train).flatten() \n",
    "\n",
    "\n",
    "train_results = pd.DataFrame(data={'Train Predictions':train_predictions,'Actuals': y_train}) \n",
    "train_results\n",
    "\n",
    "tf.keras.metrics.mean_absolute_percentage_error(train_results['Train Predictions'],train_results['Actuals'])  #Mean Absolute percentage Error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x262a0265dd0>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(15,6))\n",
    "plt.plot(train_results['Train Predictions'],label='Prediction', color =\"orange\")\n",
    "plt.plot(train_results['Actuals'],label='Actual Price',alpha=0.6, linewidth = 1.0, color =\"b\")\n",
    "plt.legend()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Validation Data-Set isolated View\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 20ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=float32, numpy=2.3230212>"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val_predictions = model_n.predict(X_val).flatten() #Preis-Vorhersagen des Trainingsets\n",
    "\n",
    "#Dictionary mit Preis-Vorhersagen und tatsächlichen Preisen des Trainingsets\n",
    "val_results = pd.DataFrame(data={'Validation Predictions':val_predictions,'Actuals': y_val}) \n",
    "val_results\n",
    "\n",
    "tf.keras.metrics.mean_absolute_percentage_error(val_results['Validation Predictions'],val_results['Actuals'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x262a03bb290>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,6))\n",
    "plt.plot(val_results['Validation Predictions'],label='Prediction', color=\"orange\")\n",
    "plt.plot(val_results['Actuals'],label='Actual Price',alpha=0.6, linewidth = 1.0, color =\"b\")\n",
    "plt.legend()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test Data-Set isolated View"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 20ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(), dtype=float32, numpy=5.3371744>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_predictions = model_n.predict(X_test).flatten() #Preis-Vorhersagen des Trainingsets\n",
    "\n",
    "#Dictionary mit Preis-Vorhersagen und tatsächlichen Preisen des Trainingsets\n",
    "test_results = pd.DataFrame(data={'Test Predictions':test_predictions,'Actuals': y_test}) \n",
    "test_results\n",
    "\n",
    "tf.keras.metrics.mean_absolute_percentage_error(test_results['Test Predictions'],test_results['Actuals'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x262a1803290>"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,6))\n",
    "plt.plot(test_results['Test Predictions'],label='Prediction',color = \"orange\")\n",
    "plt.plot(test_results['Actuals'],label='Actual Price',alpha=0.6, linewidth = 1.0, color =\"b\")\n",
    "plt.legend()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Combined Final View with Training- , Validation, and Test-MAPE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training-MAPE 10.503478050231934\n",
      "Validation-MAPE 2.323021173477173\n",
      "Test-MAPE 5.337174415588379\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,6))\n",
    "\n",
    "plt.plot(train_results['Train Predictions'],label='Prediction', color = \"orange\")\n",
    "plt.plot(train_results['Actuals'],label='Actual Price', color = \"b\",alpha=0.6, linewidth = 1.0)\n",
    "plt.plot(np.arange(len(val_results))+len(train_results),val_results['Validation Predictions'],color = 'orange')\n",
    "plt.plot(np.arange(len(val_results))+len(train_results),val_results['Actuals'], color = \"b\",alpha=0.6, linewidth = 1.0)\n",
    "plt.plot(np.arange(len(test_results))+len(train_results)+len(val_results),test_results['Test Predictions'], color = \"orange\")\n",
    "plt.plot(np.arange(len(test_results))+len(train_results)+len(val_results),test_results['Actuals'],color = \"b\",alpha=0.6, linewidth = 1.0)\n",
    "plt.axvline(x = 250, color = 'b', label = 'Validation')\n",
    "plt.axvline(x = 275, color = 'r', label = 'Test')\n",
    "plt.legend()\n",
    "plt.legend()\n",
    "\n",
    "print(f\"Training-MAPE {tf.keras.metrics.mean_absolute_percentage_error(train_results['Train Predictions'],train_results['Actuals'])}\")\n",
    "print(f\"Validation-MAPE {tf.keras.metrics.mean_absolute_percentage_error(val_results['Validation Predictions'],val_results['Actuals'])}\")\n",
    "print(f\"Test-MAPE {tf.keras.metrics.mean_absolute_percentage_error(test_results['Test Predictions'],test_results['Actuals'])}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.11"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}