08.4.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from google.colab import drive\n",
    "#drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data = pd.read_csv('/content/drive/My Drive/solar-forecasting/data/meteo_galicia.csv', sep=';', parse_dates=['fecha'], decimal=',',\n",
    "#                 infer_datetime_format=True, low_memory=False, index_col='fecha').sort_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cargar datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('./data/meteo_galicia.csv', sep=';', parse_dates=['fecha'], decimal=',',\n",
    "                 infer_datetime_format=True, low_memory=False, index_col='fecha').sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>estacionid</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>tamb</th>\n",
       "      <th>hum</th>\n",
       "      <th>pres</th>\n",
       "      <th>prec</th>\n",
       "      <th>rad</th>\n",
       "      <th>vvien</th>\n",
       "      <th>tipodered</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fecha</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>2019-01-01 00:10:00</th>\n",
       "      <td>20284</td>\n",
       "      <td>42.4044</td>\n",
       "      <td>-8.6730</td>\n",
       "      <td>5.4</td>\n",
       "      <td>0</td>\n",
       "      <td>1022.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20233</td>\n",
       "      <td>42.3554</td>\n",
       "      <td>-7.0086</td>\n",
       "      <td>7.3</td>\n",
       "      <td>0</td>\n",
       "      <td>892.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20218</td>\n",
       "      <td>43.1626</td>\n",
       "      <td>-7.2826</td>\n",
       "      <td>9.2</td>\n",
       "      <td>0</td>\n",
       "      <td>955.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20217</td>\n",
       "      <td>42.7071</td>\n",
       "      <td>-7.0471</td>\n",
       "      <td>8.5</td>\n",
       "      <td>0</td>\n",
       "      <td>882.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.3</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20221</td>\n",
       "      <td>43.5407</td>\n",
       "      <td>-7.0830</td>\n",
       "      <td>5.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1027.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>OFICIAL</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>2019-06-01 19:40:00</th>\n",
       "      <td>20249</td>\n",
       "      <td>42.2987</td>\n",
       "      <td>-7.5907</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>909.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>402.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20259</td>\n",
       "      <td>42.1630</td>\n",
       "      <td>-7.0896</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>922.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>374.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20255</td>\n",
       "      <td>42.4631</td>\n",
       "      <td>-6.9295</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>378.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20187</td>\n",
       "      <td>42.8760</td>\n",
       "      <td>-8.5594</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>988.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>419.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20295</td>\n",
       "      <td>42.0786</td>\n",
       "      <td>-8.6794</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>963.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>OFICIAL</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2510008 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     estacionid      lat     lon  tamb  hum    pres  prec  \\\n",
       "fecha                                                                       \n",
       "2019-01-01 00:10:00       20284  42.4044 -8.6730   5.4    0  1022.0   0.0   \n",
       "2019-01-01 00:10:00       20233  42.3554 -7.0086   7.3    0   892.1   0.0   \n",
       "2019-01-01 00:10:00       20218  43.1626 -7.2826   9.2    0   955.0   0.0   \n",
       "2019-01-01 00:10:00       20217  42.7071 -7.0471   8.5    0   882.5   0.0   \n",
       "2019-01-01 00:10:00       20221  43.5407 -7.0830   5.3    0  1027.0   0.0   \n",
       "...                         ...      ...     ...   ...  ...     ...   ...   \n",
       "2019-06-01 19:40:00       20249  42.2987 -7.5907  25.0    0   909.0   0.0   \n",
       "2019-06-01 19:40:00       20259  42.1630 -7.0896  27.0    0   922.0   0.0   \n",
       "2019-06-01 19:40:00       20255  42.4631 -6.9295  30.0    0     NaN   0.0   \n",
       "2019-06-01 19:40:00       20187  42.8760 -8.5594  28.0    0   988.0   0.0   \n",
       "2019-06-01 19:40:00       20295  42.0786 -8.6794  25.0    0   963.0   0.0   \n",
       "\n",
       "                       rad  vvien tipodered  \n",
       "fecha                                        \n",
       "2019-01-01 00:10:00    0.0    0.7   OFICIAL  \n",
       "2019-01-01 00:10:00    0.0    3.5   OFICIAL  \n",
       "2019-01-01 00:10:00    0.0    0.0   OFICIAL  \n",
       "2019-01-01 00:10:00    1.0    6.3   OFICIAL  \n",
       "2019-01-01 00:10:00    0.0    2.8   OFICIAL  \n",
       "...                    ...    ...       ...  \n",
       "2019-06-01 19:40:00  402.0    2.0   OFICIAL  \n",
       "2019-06-01 19:40:00  374.0    3.0   OFICIAL  \n",
       "2019-06-01 19:40:00  378.0    0.0   OFICIAL  \n",
       "2019-06-01 19:40:00  419.0    3.0   OFICIAL  \n",
       "2019-06-01 19:40:00   26.0    0.0   OFICIAL  \n",
       "\n",
       "[2510008 rows x 10 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Eliminar columnas innecesarias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.drop(columns=['tipodered', 'hum', 'prec', 'vvien', 'pres'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2510008"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "estacionid        0\n",
       "lat               0\n",
       "lon               0\n",
       "tamb           2475\n",
       "rad           10847\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2496903"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "estacionid    0\n",
       "lat           0\n",
       "lon           0\n",
       "tamb          0\n",
       "rad           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Escalar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Scaler:\n",
    "    def fit(self, data):\n",
    "        self._min = data.min()\n",
    "        self._max = data.max()\n",
    "    \n",
    "    def normalize(self, data):\n",
    "        return (data - self._min) / (self._max - self._min)\n",
    "    \n",
    "    def denormalize(self, data):\n",
    "        return data*(self._max-self._min)+self._min"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "tamb_scaler = Scaler()\n",
    "tamb_scaler.fit(data['tamb'])\n",
    "data['tamb'] = tamb_scaler.normalize(data['tamb'])\n",
    "\n",
    "rad_scaler = Scaler()\n",
    "rad_scaler.fit(data['rad'])\n",
    "data['rad'] = rad_scaler.normalize(data['rad'])"
   ]
  },
  {
   "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>estacionid</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>tamb</th>\n",
       "      <th>rad</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fecha</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>2019-01-01 00:10:00</th>\n",
       "      <td>20284</td>\n",
       "      <td>42.4044</td>\n",
       "      <td>-8.6730</td>\n",
       "      <td>0.470423</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20233</td>\n",
       "      <td>42.3554</td>\n",
       "      <td>-7.0086</td>\n",
       "      <td>0.497183</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20218</td>\n",
       "      <td>43.1626</td>\n",
       "      <td>-7.2826</td>\n",
       "      <td>0.523944</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20217</td>\n",
       "      <td>42.7071</td>\n",
       "      <td>-7.0471</td>\n",
       "      <td>0.514085</td>\n",
       "      <td>0.000791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 00:10:00</th>\n",
       "      <td>20221</td>\n",
       "      <td>43.5407</td>\n",
       "      <td>-7.0830</td>\n",
       "      <td>0.469014</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20249</td>\n",
       "      <td>42.2987</td>\n",
       "      <td>-7.5907</td>\n",
       "      <td>0.746479</td>\n",
       "      <td>0.317787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20259</td>\n",
       "      <td>42.1630</td>\n",
       "      <td>-7.0896</td>\n",
       "      <td>0.774648</td>\n",
       "      <td>0.295652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20255</td>\n",
       "      <td>42.4631</td>\n",
       "      <td>-6.9295</td>\n",
       "      <td>0.816901</td>\n",
       "      <td>0.298814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20187</td>\n",
       "      <td>42.8760</td>\n",
       "      <td>-8.5594</td>\n",
       "      <td>0.788732</td>\n",
       "      <td>0.331225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:40:00</th>\n",
       "      <td>20295</td>\n",
       "      <td>42.0786</td>\n",
       "      <td>-8.6794</td>\n",
       "      <td>0.746479</td>\n",
       "      <td>0.020553</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2496903 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     estacionid      lat     lon      tamb       rad\n",
       "fecha                                                               \n",
       "2019-01-01 00:10:00       20284  42.4044 -8.6730  0.470423  0.000000\n",
       "2019-01-01 00:10:00       20233  42.3554 -7.0086  0.497183  0.000000\n",
       "2019-01-01 00:10:00       20218  43.1626 -7.2826  0.523944  0.000000\n",
       "2019-01-01 00:10:00       20217  42.7071 -7.0471  0.514085  0.000791\n",
       "2019-01-01 00:10:00       20221  43.5407 -7.0830  0.469014  0.000000\n",
       "...                         ...      ...     ...       ...       ...\n",
       "2019-06-01 19:40:00       20249  42.2987 -7.5907  0.746479  0.317787\n",
       "2019-06-01 19:40:00       20259  42.1630 -7.0896  0.774648  0.295652\n",
       "2019-06-01 19:40:00       20255  42.4631 -6.9295  0.816901  0.298814\n",
       "2019-06-01 19:40:00       20187  42.8760 -8.5594  0.788732  0.331225\n",
       "2019-06-01 19:40:00       20295  42.0786 -8.6794  0.746479  0.020553\n",
       "\n",
       "[2496903 rows x 5 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Obtener id de las estaciones"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "142"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "estaciones = [s for s in set(data['estacionid'])]\n",
    "\n",
    "len(estaciones)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Obtener datos de las estaciones\n",
    "\n",
    "**Las estaciones que no tienen registro de radiación son eliminadas**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2.62 s, sys: 0 ns, total: 2.62 s\n",
      "Wall time: 2.63 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "data_by_station = []\n",
    "\n",
    "for s in estaciones:\n",
    "    filter_station = data['estacionid'] == s\n",
    "    data_station = data[filter_station].drop(columns=['estacionid'])\n",
    "    d = data_station.describe()\n",
    "    if d.rad['max'] != 0:\n",
    "        data_by_station.append(data_station)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data_by_station)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ajustar intervalos de tiempo\n",
    "\n",
    "**Ajuste de 1 hora**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 287 ms, sys: 6.22 ms, total: 294 ms\n",
      "Wall time: 293 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "data_by_station = [s.resample('1H').mean().ffill().bfill() for s in data_by_station]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>tamb</th>\n",
       "      <th>rad</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fecha</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-01-01 01:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.545070</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 02:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.540141</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 03:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.532394</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 04:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.551056</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 05:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.558803</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 15:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.714789</td>\n",
       "      <td>0.672727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 16:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.711268</td>\n",
       "      <td>0.632806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 17:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.742958</td>\n",
       "      <td>0.548419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 18:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.764085</td>\n",
       "      <td>0.471542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.732394</td>\n",
       "      <td>0.296970</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3643 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         lat     lon      tamb       rad\n",
       "fecha                                                   \n",
       "2019-01-01 01:00:00  43.3634 -8.4419  0.545070  0.000000\n",
       "2019-01-01 02:00:00  43.3634 -8.4419  0.540141  0.000000\n",
       "2019-01-01 03:00:00  43.3634 -8.4419  0.532394  0.000000\n",
       "2019-01-01 04:00:00  43.3634 -8.4419  0.551056  0.000000\n",
       "2019-01-01 05:00:00  43.3634 -8.4419  0.558803  0.000000\n",
       "...                      ...     ...       ...       ...\n",
       "2019-06-01 15:00:00  43.3634 -8.4419  0.714789  0.672727\n",
       "2019-06-01 16:00:00  43.3634 -8.4419  0.711268  0.632806\n",
       "2019-06-01 17:00:00  43.3634 -8.4419  0.742958  0.548419\n",
       "2019-06-01 18:00:00  43.3634 -8.4419  0.764085  0.471542\n",
       "2019-06-01 19:00:00  43.3634 -8.4419  0.732394  0.296970\n",
       "\n",
       "[3643 rows x 4 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_by_station[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "for d in data_by_station:\n",
    "    d['day']=[x.day for x in d.index]\n",
    "    d['month']=[x.month for x in d.index]\n",
    "    d['hour']=[x.hour for x in d.index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>tamb</th>\n",
       "      <th>rad</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fecha</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>2019-01-01 01:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.545070</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 02:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.540141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 03:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.532394</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 04:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.551056</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01 05:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.558803</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 15:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.714789</td>\n",
       "      <td>0.672727</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 16:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.711268</td>\n",
       "      <td>0.632806</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 17:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.742958</td>\n",
       "      <td>0.548419</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 18:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.764085</td>\n",
       "      <td>0.471542</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06-01 19:00:00</th>\n",
       "      <td>43.3634</td>\n",
       "      <td>-8.4419</td>\n",
       "      <td>0.732394</td>\n",
       "      <td>0.296970</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3643 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         lat     lon      tamb       rad  day  month  hour\n",
       "fecha                                                                     \n",
       "2019-01-01 01:00:00  43.3634 -8.4419  0.545070  0.000000    1      1     1\n",
       "2019-01-01 02:00:00  43.3634 -8.4419  0.540141  0.000000    1      1     2\n",
       "2019-01-01 03:00:00  43.3634 -8.4419  0.532394  0.000000    1      1     3\n",
       "2019-01-01 04:00:00  43.3634 -8.4419  0.551056  0.000000    1      1     4\n",
       "2019-01-01 05:00:00  43.3634 -8.4419  0.558803  0.000000    1      1     5\n",
       "...                      ...     ...       ...       ...  ...    ...   ...\n",
       "2019-06-01 15:00:00  43.3634 -8.4419  0.714789  0.672727    1      6    15\n",
       "2019-06-01 16:00:00  43.3634 -8.4419  0.711268  0.632806    1      6    16\n",
       "2019-06-01 17:00:00  43.3634 -8.4419  0.742958  0.548419    1      6    17\n",
       "2019-06-01 18:00:00  43.3634 -8.4419  0.764085  0.471542    1      6    18\n",
       "2019-06-01 19:00:00  43.3634 -8.4419  0.732394  0.296970    1      6    19\n",
       "\n",
       "[3643 rows x 7 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_by_station[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.        , 0.        , 0.        , ..., 0.54841897, 0.4715415 ,\n",
       "       0.2969697 ])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### rad\n",
    "data_by_station[0].values[:, -4]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualización"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set(rc={'figure.figsize': (16, 6)})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preparación de los datos para el entrenamiento de la red"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "# Se fija la semilla para generar valores aleatorios\n",
    "np.random.seed(1)\n",
    "tf.random.set_seed(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Función para preparar los datos para una RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_rnn_data(dataset, history_size=30, target_size=7):\n",
    "    start_index = history_size\n",
    "    end_index = len(dataset) - target_size\n",
    "    \n",
    "    data = []\n",
    "    labels = []\n",
    "    \n",
    "    for i in range(start_index, end_index):\n",
    "        indices = range(i - history_size, i, 1)\n",
    "        data.append(dataset[indices])\n",
    "        labels.append(dataset[i:i+target_size, -4])\n",
    "        \n",
    "    data = np.array(data)\n",
    "    labels = np.array(labels)\n",
    "    \n",
    "    return data, labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "past_history = 23 # Pasar información de 23 horas anteriores\n",
    "forecast_horizon = 1 # Obtener nivel de radiacion para la siguiente hora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 7.35 s, sys: 126 ms, total: 7.48 s\n",
      "Wall time: 7.44 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "data = [get_rnn_data(d.values, past_history, forecast_horizon) for d in data_by_station]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_input = data[0][0]\n",
    "data_target = data[0][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3619, 23, 7)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3619, 1)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  0.  ],\n",
       "       [  0.  ],\n",
       "       [  0.  ],\n",
       "       ...,\n",
       "       [800.5 ],\n",
       "       [693.75],\n",
       "       [596.5 ]])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rad_scaler.denormalize(data_target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        ],\n",
       "       [0.        ],\n",
       "       [0.        ],\n",
       "       ...,\n",
       "       [0.63280632],\n",
       "       [0.54841897],\n",
       "       [0.4715415 ]])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.9 s, sys: 2.93 s, total: 9.83 s\n",
      "Wall time: 9.84 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "for i in range(1, len(data)):\n",
    "    data_input = np.concatenate((data_input, data[i][0]))\n",
    "    data_target = np.concatenate((data_target, data[i][1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(401335, 23, 7)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(401335, 1)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_target.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GRU"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import GRU, Dense, Conv2D, Reshape\n",
    "from tensorflow.keras.losses import mae\n",
    "from tensorflow.keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "earlyStopping = EarlyStopping(monitor='val_loss', mode='min', patience=3, restore_best_weights=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 132 ms, sys: 69.7 ms, total: 202 ms\n",
      "Wall time: 202 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "X_train, X_test, y_train, y_test = train_test_split(data_input, data_target, train_size=0.8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rad_mae(x, y):\n",
    "    X = rad_scaler.denormalize(x)\n",
    "    Y = rad_scaler.denormalize(y)\n",
    "    \n",
    "    return mae(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape (Reshape)            (None, 23, 7, 1)          0         \n",
      "_________________________________________________________________\n",
      "conv2d (Conv2D)              (None, 23, 7, 64)         256       \n",
      "_________________________________________________________________\n",
      "reshape_1 (Reshape)          (None, 23, 448)           0         \n",
      "_________________________________________________________________\n",
      "gru (GRU)                    (None, 64)                98688     \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 99,009\n",
      "Trainable params: 99,009\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential([\n",
    "    Reshape(\n",
    "        input_shape=(X_train.shape[1], X_train.shape[2]), \n",
    "        target_shape=(X_train.shape[1], X_train.shape[2], 1)\n",
    "    ),\n",
    "    Conv2D(64, kernel_size=(3, 1), padding='same'),\n",
    "    Reshape(target_shape=(X_train.shape[1], 64*X_train.shape[2])),\n",
    "    GRU(64, input_shape=(X_train.shape[1], X_train.shape[2])),\n",
    "    Dense(forecast_horizon, activation='sigmoid')\n",
    "])\n",
    "\n",
    "model.compile(loss='mae', optimizer='adam', metrics=[rad_mae])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1004/1004 [==============================] - 169s 168ms/step - loss: 0.0547 - rad_mae: 69.1689 - val_loss: 0.0230 - val_rad_mae: 29.1246\n",
      "Epoch 2/10\n",
      "1004/1004 [==============================] - 169s 169ms/step - loss: 0.0230 - rad_mae: 29.0914 - val_loss: 0.0213 - val_rad_mae: 26.9837\n",
      "Epoch 3/10\n",
      "1004/1004 [==============================] - 168s 167ms/step - loss: 0.0212 - rad_mae: 26.8427 - val_loss: 0.0206 - val_rad_mae: 26.1210\n",
      "Epoch 4/10\n",
      "1004/1004 [==============================] - 173s 172ms/step - loss: 0.0200 - rad_mae: 25.2375 - val_loss: 0.0189 - val_rad_mae: 23.9675\n",
      "Epoch 5/10\n",
      "1004/1004 [==============================] - 171s 171ms/step - loss: 0.0195 - rad_mae: 24.6538 - val_loss: 0.0192 - val_rad_mae: 24.3439\n",
      "Epoch 6/10\n",
      "1004/1004 [==============================] - 169s 168ms/step - loss: 0.0189 - rad_mae: 23.9552 - val_loss: 0.0186 - val_rad_mae: 23.4939\n",
      "Epoch 7/10\n",
      "1004/1004 [==============================] - 169s 168ms/step - loss: 0.0187 - rad_mae: 23.6354 - val_loss: 0.0188 - val_rad_mae: 23.7580\n",
      "Epoch 8/10\n",
      "1004/1004 [==============================] - 168s 167ms/step - loss: 0.0186 - rad_mae: 23.5626 - val_loss: 0.0203 - val_rad_mae: 25.6382\n",
      "Epoch 9/10\n",
      "1004/1004 [==============================] - 170s 170ms/step - loss: 0.0186 - rad_mae: 23.5589 - val_loss: 0.0188 - val_rad_mae: 23.7620\n",
      "CPU times: user 1h 3min 11s, sys: 25min 37s, total: 1h 28min 49s\n",
      "Wall time: 25min 30s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "history = model.fit(X_train, y_train, epochs=10, batch_size=256, validation_split=0.2, callbacks=[earlyStopping])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'], label='loss')\n",
    "plt.plot(history.history['val_loss'], label='val_loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2509/2509 [==============================] - 35s 14ms/step - loss: 0.0181 - rad_mae: 22.8915\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.01809602975845337, 22.89145278930664]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(80267, 1)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predicted = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "YP = rad_scaler.denormalize(y_predicted)\n",
    "YT = rad_scaler.denormalize(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(YT)\n",
    "plt.plot(YP, color='salmon')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "error = YT - YP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(error)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}