ETAPrediction_FlightsFromORYtoTLS_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Description:** <br>\n",
    "Here we try to predict the ETA at a future waypoint or final destination, for flights from ORY to TLS airports. The LSTM network has been trained on the flight data from ORY to TLS using the timestamps in epoch format. In this version we set the number of time steps (lag) to be shifted (`n_in` >= **10**) in the `series_to_supervised` function and predict the latitude and longitude\n",
    "\n",
    "* __Author__: Dilip Rajkumar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:90% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:90% !important; }</style>\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# Set Random Seed for reproducible results\n",
    "np.random.seed(42)\n",
    "\n",
    "#Set some numpy print options for displaying numpy arrays to fit maximum width of cell\n",
    "np.set_printoptions(precision=3, edgeitems=30, linewidth=1000,formatter=dict(float=lambda x: \"%.8g\" % x))\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler, LabelEncoder\n",
    "from sklearn.metrics import mean_squared_error,mean_absolute_error\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, LSTM, Dropout, Activation\n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.colors import Normalize, LinearSegmentedColormap, PowerNorm\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>postime</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>icao</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1539414114000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1539414140000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.218559</td>\n",
       "      <td>1.60048</td>\n",
       "      <td>12425</td>\n",
       "      <td>215</td>\n",
       "      <td>349</td>\n",
       "      <td>-1664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1539414147000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.211040</td>\n",
       "      <td>1.59303</td>\n",
       "      <td>12250</td>\n",
       "      <td>215</td>\n",
       "      <td>346</td>\n",
       "      <td>-1664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1539414149000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.207119</td>\n",
       "      <td>1.58909</td>\n",
       "      <td>12175</td>\n",
       "      <td>215</td>\n",
       "      <td>346</td>\n",
       "      <td>-1600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>1539414154000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>309</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          postime aircraftreg     icao        lat      lon    alt  course  \\\n",
       "1   1539414114000      F-GKXE  3746532  48.729340  2.35702      0     128   \n",
       "5   1539414140000      F-GTAP  3755023  44.218559  1.60048  12425     215   \n",
       "21  1539414147000      F-GTAP  3755023  44.211040  1.59303  12250     215   \n",
       "25  1539414149000      F-GTAP  3755023  44.207119  1.58909  12175     215   \n",
       "39  1539414154000      F-GKXE  3746532  48.729351  2.35701      0     309   \n",
       "\n",
       "    speed  verticalspeed  \n",
       "1       0              0  \n",
       "5     349          -1664  \n",
       "21    346          -1664  \n",
       "25    346          -1600  \n",
       "39      0              0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('data/ADSB_SampleData.csv',parse_dates = ['postime'])\n",
    "cols_of_interest = ['postime','aircraftreg','icao','lat','lon','alt','course','speed','verticalspeed']\n",
    "flightpaths = df.groupby(['fromairport']).get_group('ORY')[cols_of_interest]\n",
    "flightpaths.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 11904 entries, 1 to 97971\n",
      "Data columns (total 9 columns):\n",
      "postime          11904 non-null int64\n",
      "aircraftreg      11904 non-null object\n",
      "icao             11904 non-null int64\n",
      "lat              11904 non-null float64\n",
      "lon              11904 non-null float64\n",
      "alt              11904 non-null int64\n",
      "course           11904 non-null int64\n",
      "speed            11904 non-null int64\n",
      "verticalspeed    11904 non-null int64\n",
      "dtypes: float64(2), int64(6), object(1)\n",
      "memory usage: 930.0+ KB\n"
     ]
    }
   ],
   "source": [
    "flightpaths[['postime']] = flightpaths[['postime']].apply(pd.to_numeric)\n",
    "flightpaths.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>postime</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>icao</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>1539414158000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>309</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1539414164000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1539414173000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.177410</td>\n",
       "      <td>1.55975</td>\n",
       "      <td>11375</td>\n",
       "      <td>215</td>\n",
       "      <td>342</td>\n",
       "      <td>-2496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>1539414187000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.159729</td>\n",
       "      <td>1.54224</td>\n",
       "      <td>11025</td>\n",
       "      <td>215</td>\n",
       "      <td>336</td>\n",
       "      <td>-1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>1539414185000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265</th>\n",
       "      <td>1539414230000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>373</th>\n",
       "      <td>1539414270000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>309</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>420</th>\n",
       "      <td>1539414284000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>435</th>\n",
       "      <td>1539414292000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.044289</td>\n",
       "      <td>1.41678</td>\n",
       "      <td>8775</td>\n",
       "      <td>227</td>\n",
       "      <td>283</td>\n",
       "      <td>-2112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>457</th>\n",
       "      <td>1539414294000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466</th>\n",
       "      <td>1539414304000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.033482</td>\n",
       "      <td>1.40050</td>\n",
       "      <td>8350</td>\n",
       "      <td>227</td>\n",
       "      <td>283</td>\n",
       "      <td>-2304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>482</th>\n",
       "      <td>1539414305000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>247</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>534</th>\n",
       "      <td>1539414320000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>545</th>\n",
       "      <td>1539414328000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.012562</td>\n",
       "      <td>1.36888</td>\n",
       "      <td>7600</td>\n",
       "      <td>227</td>\n",
       "      <td>283</td>\n",
       "      <td>-1472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>550</th>\n",
       "      <td>1539414329000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>685</th>\n",
       "      <td>1539414380000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>732</th>\n",
       "      <td>1539414399000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>783</th>\n",
       "      <td>1539414414000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>853</th>\n",
       "      <td>1539414440000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>877</th>\n",
       "      <td>1539414449000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1016</th>\n",
       "      <td>1539414494000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1042</th>\n",
       "      <td>1539414504000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35704</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1146</th>\n",
       "      <td>1539414535000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1167</th>\n",
       "      <td>1539414544000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35704</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1198</th>\n",
       "      <td>1539414557000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>43.810131</td>\n",
       "      <td>1.18193</td>\n",
       "      <td>3350</td>\n",
       "      <td>183</td>\n",
       "      <td>174</td>\n",
       "      <td>-1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1216</th>\n",
       "      <td>1539414557000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>43.810131</td>\n",
       "      <td>1.18193</td>\n",
       "      <td>3350</td>\n",
       "      <td>183</td>\n",
       "      <td>174</td>\n",
       "      <td>-1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1228</th>\n",
       "      <td>1539414557000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>43.810131</td>\n",
       "      <td>1.18193</td>\n",
       "      <td>3350</td>\n",
       "      <td>183</td>\n",
       "      <td>174</td>\n",
       "      <td>-1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1233</th>\n",
       "      <td>1539414557000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>43.810131</td>\n",
       "      <td>1.18193</td>\n",
       "      <td>3350</td>\n",
       "      <td>183</td>\n",
       "      <td>174</td>\n",
       "      <td>-1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1237</th>\n",
       "      <td>1539414564000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1381</th>\n",
       "      <td>1539414615000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95952</th>\n",
       "      <td>1539413404000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96001</th>\n",
       "      <td>1539413424000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96015</th>\n",
       "      <td>1539413430000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>45.491322</td>\n",
       "      <td>1.94242</td>\n",
       "      <td>28975</td>\n",
       "      <td>186</td>\n",
       "      <td>445</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96183</th>\n",
       "      <td>1539413495000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96214</th>\n",
       "      <td>1539413504000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96341</th>\n",
       "      <td>1539413554000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96392</th>\n",
       "      <td>1539413574000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35699</td>\n",
       "      <td>0</td>\n",
       "      <td>229</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96460</th>\n",
       "      <td>1539413602000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>45.149490</td>\n",
       "      <td>1.88630</td>\n",
       "      <td>26850</td>\n",
       "      <td>186</td>\n",
       "      <td>421</td>\n",
       "      <td>-960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96488</th>\n",
       "      <td>1539413603000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35698</td>\n",
       "      <td>0</td>\n",
       "      <td>214</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96510</th>\n",
       "      <td>1539413609000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35698</td>\n",
       "      <td>0</td>\n",
       "      <td>314</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96536</th>\n",
       "      <td>1539413629000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35699</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96623</th>\n",
       "      <td>1539413659000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96747</th>\n",
       "      <td>1539413708000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96819</th>\n",
       "      <td>1539413728000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>132</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96873</th>\n",
       "      <td>1539413754000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>132</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96941</th>\n",
       "      <td>1539413774000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97130</th>\n",
       "      <td>1539413833000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97245</th>\n",
       "      <td>1539413873000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35700</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97401</th>\n",
       "      <td>1539413929000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35704</td>\n",
       "      <td>0</td>\n",
       "      <td>121</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97437</th>\n",
       "      <td>1539413952000</td>\n",
       "      <td>F-GTAP</td>\n",
       "      <td>3755023</td>\n",
       "      <td>44.502869</td>\n",
       "      <td>1.78176</td>\n",
       "      <td>18250</td>\n",
       "      <td>186</td>\n",
       "      <td>388</td>\n",
       "      <td>-1984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97457</th>\n",
       "      <td>1539413949000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35705</td>\n",
       "      <td>0</td>\n",
       "      <td>349</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97492</th>\n",
       "      <td>1539413963000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35704</td>\n",
       "      <td>0</td>\n",
       "      <td>311</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97521</th>\n",
       "      <td>1539413969000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35704</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97574</th>\n",
       "      <td>1539413993000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>311</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97638</th>\n",
       "      <td>1539414009000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0</td>\n",
       "      <td>167</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97766</th>\n",
       "      <td>1539414055000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>221</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97843</th>\n",
       "      <td>1539414099000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97889</th>\n",
       "      <td>1539414079000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>221</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97893</th>\n",
       "      <td>1539414084000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729351</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>309</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97909</th>\n",
       "      <td>1539414114000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729340</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2165 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             postime aircraftreg     icao        lat      lon    alt  course  \\\n",
       "59     1539414158000      F-GKXE  3746532  48.729340  2.35701      0     309   \n",
       "88     1539414164000      F-GKXE  3746532  48.729351  2.35702      0      89   \n",
       "109    1539414173000      F-GTAP  3755023  44.177410  1.55975  11375     215   \n",
       "153    1539414187000      F-GTAP  3755023  44.159729  1.54224  11025     215   \n",
       "157    1539414185000      F-GKXE  3746532  48.729340  2.35703      0     130   \n",
       "265    1539414230000      F-GKXE  3746532  48.729340  2.35702      0     271   \n",
       "373    1539414270000      F-GKXE  3746532  48.729340  2.35701      0     309   \n",
       "420    1539414284000      F-GKXE  3746532  48.729351  2.35702      0      89   \n",
       "435    1539414292000      F-GTAP  3755023  44.044289  1.41678   8775     227   \n",
       "457    1539414294000      F-GKXE  3746532  48.729340  2.35703      0      89   \n",
       "466    1539414304000      F-GTAP  3755023  44.033482  1.40050   8350     227   \n",
       "482    1539414305000      F-GKXE  3746532  48.729340  2.35702      0     247   \n",
       "534    1539414320000      F-GKXE  3746532  48.729340  2.35703      0      66   \n",
       "545    1539414328000      F-GTAP  3755023  44.012562  1.36888   7600     227   \n",
       "550    1539414329000      F-GKXE  3746532  48.729328  2.35703      0     180   \n",
       "685    1539414380000      F-GKXE  3746532  48.729321  2.35701      0     229   \n",
       "732    1539414399000      F-GKXE  3746532  48.729328  2.35702      0     229   \n",
       "783    1539414414000      F-GKXE  3746532  48.729321  2.35701      0     229   \n",
       "853    1539414440000      F-GKXE  3746532  48.729321  2.35703      0      89   \n",
       "877    1539414449000      F-GKXE  3746532  48.729321  2.35701      0     271   \n",
       "1016   1539414494000      F-GKXE  3746532  48.729321  2.35701      0     271   \n",
       "1042   1539414504000      F-GKXE  3746532  48.729321  2.35704      0      89   \n",
       "1146   1539414535000      F-GKXE  3746532  48.729328  2.35703      0      89   \n",
       "1167   1539414544000      F-GKXE  3746532  48.729321  2.35704      0      89   \n",
       "1198   1539414557000      F-GTAP  3755023  43.810131  1.18193   3350     183   \n",
       "1216   1539414557000      F-GTAP  3755023  43.810131  1.18193   3350     183   \n",
       "1228   1539414557000      F-GTAP  3755023  43.810131  1.18193   3350     183   \n",
       "1233   1539414557000      F-GTAP  3755023  43.810131  1.18193   3350     183   \n",
       "1237   1539414564000      F-GKXE  3746532  48.729321  2.35702      0     271   \n",
       "1381   1539414615000      F-GKXE  3746532  48.729321  2.35702      0      48   \n",
       "...              ...         ...      ...        ...      ...    ...     ...   \n",
       "95952  1539413404000      F-GKXE  3746532  48.729328  2.35702      0     229   \n",
       "96001  1539413424000      F-GKXE  3746532  48.729321  2.35701      0     229   \n",
       "96015  1539413430000      F-GTAP  3755023  45.491322  1.94242  28975     186   \n",
       "96183  1539413495000      F-GKXE  3746532  48.729340  2.35702      0      24   \n",
       "96214  1539413504000      F-GKXE  3746532  48.729340  2.35703      0      24   \n",
       "96341  1539413554000      F-GKXE  3746532  48.729328  2.35700      0     229   \n",
       "96392  1539413574000      F-GKXE  3746532  48.729328  2.35699      0     229   \n",
       "96460  1539413602000      F-GTAP  3755023  45.149490  1.88630  26850     186   \n",
       "96488  1539413603000      F-GKXE  3746532  48.729321  2.35698      0     214   \n",
       "96510  1539413609000      F-GKXE  3746532  48.729340  2.35698      0     314   \n",
       "96536  1539413629000      F-GKXE  3746532  48.729328  2.35699      0      89   \n",
       "96623  1539413659000      F-GKXE  3746532  48.729340  2.35700      0      89   \n",
       "96747  1539413708000      F-GKXE  3746532  48.729351  2.35700      0      89   \n",
       "96819  1539413728000      F-GKXE  3746532  48.729340  2.35700      0     132   \n",
       "96873  1539413754000      F-GKXE  3746532  48.729340  2.35700      0     132   \n",
       "96941  1539413774000      F-GKXE  3746532  48.729321  2.35701      0     180   \n",
       "97130  1539413833000      F-GKXE  3746532  48.729351  2.35701      0     180   \n",
       "97245  1539413873000      F-GKXE  3746532  48.729351  2.35700      0     180   \n",
       "97401  1539413929000      F-GKXE  3746532  48.729321  2.35704      0     121   \n",
       "97437  1539413952000      F-GTAP  3755023  44.502869  1.78176  18250     186   \n",
       "97457  1539413949000      F-GKXE  3746532  48.729351  2.35705      0     349   \n",
       "97492  1539413963000      F-GKXE  3746532  48.729351  2.35704      0     311   \n",
       "97521  1539413969000      F-GKXE  3746532  48.729340  2.35704      0     180   \n",
       "97574  1539413993000      F-GKXE  3746532  48.729351  2.35702      0     311   \n",
       "97638  1539414009000      F-GKXE  3746532  48.729328  2.35703      0     167   \n",
       "97766  1539414055000      F-GKXE  3746532  48.729328  2.35702      0     221   \n",
       "97843  1539414099000      F-GKXE  3746532  48.729340  2.35701      0     128   \n",
       "97889  1539414079000      F-GKXE  3746532  48.729328  2.35702      0     221   \n",
       "97893  1539414084000      F-GKXE  3746532  48.729351  2.35701      0     309   \n",
       "97909  1539414114000      F-GKXE  3746532  48.729340  2.35702      0     128   \n",
       "\n",
       "       speed  verticalspeed  \n",
       "59         0              0  \n",
       "88         0              0  \n",
       "109      342          -2496  \n",
       "153      336          -1088  \n",
       "157        0              0  \n",
       "265        0              0  \n",
       "373        0              0  \n",
       "420        0              0  \n",
       "435      283          -2112  \n",
       "457        0              0  \n",
       "466      283          -2304  \n",
       "482        0              0  \n",
       "534        0              0  \n",
       "545      283          -1472  \n",
       "550        0              0  \n",
       "685        0              0  \n",
       "732        0              0  \n",
       "783        0              0  \n",
       "853        0              0  \n",
       "877        0              0  \n",
       "1016       0              0  \n",
       "1042       0              0  \n",
       "1146       0              0  \n",
       "1167       0              0  \n",
       "1198     174          -1088  \n",
       "1216     174          -1088  \n",
       "1228     174          -1088  \n",
       "1233     174          -1088  \n",
       "1237       0              0  \n",
       "1381       0              0  \n",
       "...      ...            ...  \n",
       "95952      0              0  \n",
       "96001      0              0  \n",
       "96015    445              0  \n",
       "96183      0              0  \n",
       "96214      0              0  \n",
       "96341      0              0  \n",
       "96392      0              0  \n",
       "96460    421           -960  \n",
       "96488      0              0  \n",
       "96510      0              0  \n",
       "96536      0              0  \n",
       "96623      0              0  \n",
       "96747      0              0  \n",
       "96819      0              0  \n",
       "96873      0              0  \n",
       "96941      0              0  \n",
       "97130      0              0  \n",
       "97245      0              0  \n",
       "97401      0              0  \n",
       "97437    388          -1984  \n",
       "97457      0              0  \n",
       "97492      0              0  \n",
       "97521      0              0  \n",
       "97574      0              0  \n",
       "97638      0              0  \n",
       "97766      0              0  \n",
       "97843      0              0  \n",
       "97889      0              0  \n",
       "97893      0              0  \n",
       "97909      0              0  \n",
       "\n",
       "[2165 rows x 9 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check for Duplicates in the Dataframe\n",
    "flightpaths[flightpaths.duplicated(subset=['postime','icao','lat','lon'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 9739 entries, 1 to 97971\n",
      "Data columns (total 9 columns):\n",
      "postime          9739 non-null int64\n",
      "aircraftreg      9739 non-null object\n",
      "icao             9739 non-null int64\n",
      "lat              9739 non-null float64\n",
      "lon              9739 non-null float64\n",
      "alt              9739 non-null int64\n",
      "course           9739 non-null int64\n",
      "speed            9739 non-null int64\n",
      "verticalspeed    9739 non-null int64\n",
      "dtypes: float64(2), int64(6), object(1)\n",
      "memory usage: 760.9+ KB\n"
     ]
    }
   ],
   "source": [
    "flightpaths.drop_duplicates(keep='first', inplace=True)\n",
    "flightpaths.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>postime</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>icao</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>90170</th>\n",
       "      <td>1539411476000</td>\n",
       "      <td>OE-IJY</td>\n",
       "      <td>4457132</td>\n",
       "      <td>43.627281</td>\n",
       "      <td>1.37124</td>\n",
       "      <td>0</td>\n",
       "      <td>323</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90183</th>\n",
       "      <td>1539411482000</td>\n",
       "      <td>OE-IJY</td>\n",
       "      <td>4457132</td>\n",
       "      <td>43.627560</td>\n",
       "      <td>1.37093</td>\n",
       "      <td>0</td>\n",
       "      <td>323</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90394</th>\n",
       "      <td>1539411541000</td>\n",
       "      <td>OE-IJY</td>\n",
       "      <td>4457132</td>\n",
       "      <td>43.629581</td>\n",
       "      <td>1.37286</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90456</th>\n",
       "      <td>1539411557000</td>\n",
       "      <td>OE-IJY</td>\n",
       "      <td>4457132</td>\n",
       "      <td>43.629860</td>\n",
       "      <td>1.37323</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90444</th>\n",
       "      <td>1539411562000</td>\n",
       "      <td>OE-IJY</td>\n",
       "      <td>4457132</td>\n",
       "      <td>43.629890</td>\n",
       "      <td>1.37325</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             postime aircraftreg     icao        lat      lon  alt  course  \\\n",
       "90170  1539411476000      OE-IJY  4457132  43.627281  1.37124    0     323   \n",
       "90183  1539411482000      OE-IJY  4457132  43.627560  1.37093    0     323   \n",
       "90394  1539411541000      OE-IJY  4457132  43.629581  1.37286    0      42   \n",
       "90456  1539411557000      OE-IJY  4457132  43.629860  1.37323    0      42   \n",
       "90444  1539411562000      OE-IJY  4457132  43.629890  1.37325    0      42   \n",
       "\n",
       "       speed  verticalspeed  \n",
       "90170     33              0  \n",
       "90183     33              0  \n",
       "90394     21              0  \n",
       "90456     12              0  \n",
       "90444     11              0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Grouping and sorting by ICAO values and postime\n",
    "flightpaths.sort_values(by=['icao','postime'], inplace = True)\n",
    "flightpaths.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>postime</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>icao</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1539410352000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1539410362000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>212</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1539410381000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1539410386000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>131</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1539410398000</td>\n",
       "      <td>F-GKXE</td>\n",
       "      <td>3746532</td>\n",
       "      <td>48.729301</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>247</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         postime aircraftreg     icao        lat      lon  alt  course  speed  \\\n",
       "0  1539410352000      F-GKXE  3746532  48.729328  2.35702    0      31      0   \n",
       "1  1539410362000      F-GKXE  3746532  48.729309  2.35701    0     212      0   \n",
       "2  1539410381000      F-GKXE  3746532  48.729321  2.35702    0     128      0   \n",
       "3  1539410386000      F-GKXE  3746532  48.729309  2.35702    0     131      0   \n",
       "4  1539410398000      F-GKXE  3746532  48.729301  2.35701    0     247      0   \n",
       "\n",
       "   verticalspeed  \n",
       "0              0  \n",
       "1              0  \n",
       "2              0  \n",
       "3              0  \n",
       "4              0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "flightpaths.reset_index(drop=True, inplace = True)\n",
    "flightpaths.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No. of unique Aircraft Registrations: 12\n"
     ]
    }
   ],
   "source": [
    "print('No. of unique Aircraft Registrations:',flightpaths.aircraftreg.nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No. of Unique Airline Carriers flying from ORY to TLS: 12\n",
      "No. of Unique ICAOs flying from ORY to TLS: 12\n"
     ]
    }
   ],
   "source": [
    "print(\"No. of Unique Airline Carriers flying from ORY to TLS:\",flightpaths.aircraftreg.nunique())\n",
    "print(\"No. of Unique ICAOs flying from ORY to TLS:\",flightpaths.icao.nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique ICAOs flying from ORY to TLS: [3746532 3751299 3753188 3753192 3753697 3755023 3756228 3769765 3769768 3769770 4221197 4457132] \n",
      "\n",
      "Unique Airline Carriers flying from ORY to TLS: ['F-GKXE' 'F-GPMD' 'F-GRHE' 'F-GRHI' 'F-GRXB' 'F-GTAP' 'F-GUGE' 'F-HBNF' 'F-HBNI' 'F-HBNK' 'G-EZWL' 'OE-IJY']\n"
     ]
    }
   ],
   "source": [
    "print(\"Unique ICAOs flying from ORY to TLS:\",flightpaths.icao.unique(),\"\\n\")\n",
    "print(\"Unique Airline Carriers flying from ORY to TLS:\",flightpaths.aircraftreg.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>postime</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>icao</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [postime, aircraftreg, icao, lat, lon, alt, course, speed, verticalspeed]\n",
       "Index: []"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check for Duplicates in the Dataframe\n",
    "flightpaths[flightpaths.duplicated(subset=['postime','icao','lat','lon'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "postime          1539456573000\n",
       "aircraftreg             F-HBNI\n",
       "icao                   3769768\n",
       "lat                    43.6292\n",
       "lon                    1.37537\n",
       "alt                          0\n",
       "course                       8\n",
       "speed                        2\n",
       "verticalspeed                0\n",
       "Name: 6779, dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "flightpaths.iloc[6779]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Pre-Processing and Feature Extraction:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>postime</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1539410352000</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1539410362000</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>212</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1539410381000</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1539410386000</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0</td>\n",
       "      <td>131</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1539410398000</td>\n",
       "      <td>48.729301</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0</td>\n",
       "      <td>247</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         postime        lat      lon  alt  course  speed  verticalspeed\n",
       "0  1539410352000  48.729328  2.35702    0      31      0              0\n",
       "1  1539410362000  48.729309  2.35701    0     212      0              0\n",
       "2  1539410381000  48.729321  2.35702    0     128      0              0\n",
       "3  1539410386000  48.729309  2.35702    0     131      0              0\n",
       "4  1539410398000  48.729301  2.35701    0     247      0              0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop Categorical Columns not needed for the neural network\n",
    "df_nn = flightpaths.copy()\n",
    "df_nn.drop(columns=['aircraftreg','icao'],inplace = True)\n",
    "df_nn.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>postime</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [postime, lat, lon, alt, course, speed, verticalspeed]\n",
       "Index: []"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check for NULL values across on column or row\n",
    "df_nn[df_nn.isnull().any(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert series to supervised learning\n",
    "def series_to_supervised(data, n_in=1, n_out=1, dropnan=True):\n",
    "    \"\"\"\n",
    "    Frame a time series as a supervised learning dataset.\n",
    "    Arguments:\n",
    "        data: Sequence of observations as a list or NumPy array.\n",
    "        n_in: Number of lag (t-n) observations as input (X). Defaults to 1\n",
    "        n_out: Number of lead (t + n) observations as output (y). Defaults to 1, that is var(t)\n",
    "        dropnan: Boolean whether or not to drop rows with NaN values.\n",
    "    Returns:\n",
    "        Pandas DataFrame of series framed for supervised learning.\n",
    "    \"\"\"\n",
    "    n_vars = 1 if type(data) is list else data.shape[1]\n",
    "    df = pd.DataFrame(data)\n",
    "    cols, names = list(), list()\n",
    "    # input sequence (t-n, ... t-1)\n",
    "    for i in range(n_in, 0, -1):\n",
    "        cols.append(df.shift(i))\n",
    "        names += [('var%d(t-%d)' % (j+1, i)) for j in range(n_vars)]\n",
    "    # forecast sequence (t, t+1, ... t+n)\n",
    "    for i in range(0, n_out):\n",
    "        cols.append(df.shift(-i))\n",
    "        if i == 0:\n",
    "            names += [('var%d(t)' % (j+1)) for j in range(n_vars)]\n",
    "        else:\n",
    "            names += [('var%d(t+%d)' % (j+1, i)) for j in range(n_vars)]\n",
    "    # put it all together\n",
    "    agg = pd.concat(cols, axis=1)\n",
    "    agg.columns = names\n",
    "    # drop rows with NaN values\n",
    "    if dropnan:\n",
    "        agg.dropna(inplace=True)\n",
    "    return agg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>var1(t-10)</th>\n",
       "      <th>var2(t-10)</th>\n",
       "      <th>var3(t-10)</th>\n",
       "      <th>var4(t-10)</th>\n",
       "      <th>var5(t-10)</th>\n",
       "      <th>var6(t-10)</th>\n",
       "      <th>var7(t-10)</th>\n",
       "      <th>var1(t-9)</th>\n",
       "      <th>var2(t-9)</th>\n",
       "      <th>var3(t-9)</th>\n",
       "      <th>...</th>\n",
       "      <th>var6(t-2)</th>\n",
       "      <th>var7(t-2)</th>\n",
       "      <th>var1(t-1)</th>\n",
       "      <th>var2(t-1)</th>\n",
       "      <th>var3(t-1)</th>\n",
       "      <th>var4(t-1)</th>\n",
       "      <th>var5(t-1)</th>\n",
       "      <th>var6(t-1)</th>\n",
       "      <th>var7(t-1)</th>\n",
       "      <th>var1(t)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729328</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729301</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0.0</td>\n",
       "      <td>212.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0.0</td>\n",
       "      <td>128.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0.0</td>\n",
       "      <td>131.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729301</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729321</td>\n",
       "      <td>2.35703</td>\n",
       "      <td>0.0</td>\n",
       "      <td>311.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539411e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729301</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>0.0</td>\n",
       "      <td>247.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539410e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35701</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539411e+12</td>\n",
       "      <td>48.729309</td>\n",
       "      <td>2.35702</td>\n",
       "      <td>0.0</td>\n",
       "      <td>210.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.539411e+12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 71 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      var1(t-10)  var2(t-10)  var3(t-10)  var4(t-10)  var5(t-10)  var6(t-10)  \\\n",
       "10  1.539410e+12   48.729328     2.35702         0.0        31.0         0.0   \n",
       "11  1.539410e+12   48.729309     2.35701         0.0       212.0         0.0   \n",
       "12  1.539410e+12   48.729321     2.35702         0.0       128.0         0.0   \n",
       "13  1.539410e+12   48.729309     2.35702         0.0       131.0         0.0   \n",
       "14  1.539410e+12   48.729301     2.35701         0.0       247.0         0.0   \n",
       "\n",
       "    var7(t-10)     var1(t-9)  var2(t-9)  var3(t-9)      ...       var6(t-2)  \\\n",
       "10         0.0  1.539410e+12  48.729309    2.35701      ...             0.0   \n",
       "11         0.0  1.539410e+12  48.729321    2.35702      ...             0.0   \n",
       "12         0.0  1.539410e+12  48.729309    2.35702      ...             0.0   \n",
       "13         0.0  1.539410e+12  48.729301    2.35701      ...             0.0   \n",
       "14         0.0  1.539410e+12  48.729309    2.35701      ...             0.0   \n",
       "\n",
       "    var7(t-2)     var1(t-1)  var2(t-1)  var3(t-1)  var4(t-1)  var5(t-1)  \\\n",
       "10        0.0  1.539410e+12  48.729301    2.35703        0.0       66.0   \n",
       "11        0.0  1.539410e+12  48.729321    2.35702        0.0       66.0   \n",
       "12        0.0  1.539410e+12  48.729321    2.35703        0.0      130.0   \n",
       "13        0.0  1.539410e+12  48.729321    2.35703        0.0      311.0   \n",
       "14        0.0  1.539411e+12  48.729309    2.35702        0.0      210.0   \n",
       "\n",
       "    var6(t-1)  var7(t-1)       var1(t)  \n",
       "10        0.0        0.0  1.539410e+12  \n",
       "11        0.0        0.0  1.539410e+12  \n",
       "12        0.0        0.0  1.539410e+12  \n",
       "13        0.0        0.0  1.539411e+12  \n",
       "14        0.0        0.0  1.539411e+12  \n",
       "\n",
       "[5 rows x 71 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load dataset vales\n",
    "values = df_nn.values\n",
    "\n",
    "# frame as supervised learning with raw values\n",
    "reframed_raw = series_to_supervised(values, 10, 1)\n",
    "\n",
    "# drop columns we don't want to predict\n",
    "reframed_raw.drop(reframed_raw.columns[[71,72,73,74,75,76]], axis=1, inplace=True)\n",
    "reframed_raw.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9739, 7)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Using Label encoder for any categorical variables\n",
    "# encoder = LabelEncoder()\n",
    "# values[:,x] = encoder.fit_transform(values[:,x])\n",
    "\n",
    "# ensure all data is float\n",
    "values = values.astype('float32')\n",
    "values.shape # Has the same number of columns as flightpath columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Normalize features with separate MinMaxScaler for inputs and outputs\n",
    "scaler_x = MinMaxScaler(feature_range=(0, 1))\n",
    "scaler_y = MinMaxScaler(feature_range=(0, 1))\n",
    "scaled_inputs = scaler_x.fit_transform(values[:,1:])\n",
    "scaled_output = scaler_y.fit_transform(values[:, :1])\n",
    "scaled = np.concatenate((scaled_output,scaled_inputs), axis=1)"
   ]
  },
  {
   "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>var1(t-10)</th>\n",
       "      <th>var2(t-10)</th>\n",
       "      <th>var3(t-10)</th>\n",
       "      <th>var4(t-10)</th>\n",
       "      <th>var5(t-10)</th>\n",
       "      <th>var6(t-10)</th>\n",
       "      <th>var7(t-10)</th>\n",
       "      <th>var1(t-9)</th>\n",
       "      <th>var2(t-9)</th>\n",
       "      <th>var3(t-9)</th>\n",
       "      <th>...</th>\n",
       "      <th>var6(t-2)</th>\n",
       "      <th>var7(t-2)</th>\n",
       "      <th>var1(t-1)</th>\n",
       "      <th>var2(t-1)</th>\n",
       "      <th>var3(t-1)</th>\n",
       "      <th>var4(t-1)</th>\n",
       "      <th>var5(t-1)</th>\n",
       "      <th>var6(t-1)</th>\n",
       "      <th>var7(t-1)</th>\n",
       "      <th>var1(t)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997078</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.084270</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "      <td>0.997073</td>\n",
       "      <td>0.817614</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.182584</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817600</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.592697</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997077</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "      <td>0.997077</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.182584</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997077</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.356742</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "      <td>0.997077</td>\n",
       "      <td>0.817614</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.362360</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.365169</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997073</td>\n",
       "      <td>0.817600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "      <td>0.997077</td>\n",
       "      <td>0.817614</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.870787</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997073</td>\n",
       "      <td>0.817600</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.691011</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.058594</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "      <td>0.997074</td>\n",
       "      <td>0.817607</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.587079</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.563291</td>\n",
       "      <td>0.060547</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 71 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    var1(t-10)  var2(t-10)  var3(t-10)  var4(t-10)  var5(t-10)  var6(t-10)  \\\n",
       "10    0.058594    0.997078    0.817607         0.0    0.084270         0.0   \n",
       "11    0.058594    0.997074    0.817600         0.0    0.592697         0.0   \n",
       "12    0.058594    0.997077    0.817607         0.0    0.356742         0.0   \n",
       "13    0.058594    0.997074    0.817607         0.0    0.365169         0.0   \n",
       "14    0.058594    0.997073    0.817600         0.0    0.691011         0.0   \n",
       "\n",
       "    var7(t-10)  var1(t-9)  var2(t-9)  var3(t-9)    ...     var6(t-2)  \\\n",
       "10    0.563291   0.058594   0.997074   0.817600    ...           0.0   \n",
       "11    0.563291   0.058594   0.997077   0.817607    ...           0.0   \n",
       "12    0.563291   0.058594   0.997074   0.817607    ...           0.0   \n",
       "13    0.563291   0.058594   0.997073   0.817600    ...           0.0   \n",
       "14    0.563291   0.058594   0.997074   0.817600    ...           0.0   \n",
       "\n",
       "    var7(t-2)  var1(t-1)  var2(t-1)  var3(t-1)  var4(t-1)  var5(t-1)  \\\n",
       "10   0.563291   0.060547   0.997073   0.817614        0.0   0.182584   \n",
       "11   0.563291   0.060547   0.997077   0.817607        0.0   0.182584   \n",
       "12   0.563291   0.060547   0.997077   0.817614        0.0   0.362360   \n",
       "13   0.563291   0.060547   0.997077   0.817614        0.0   0.870787   \n",
       "14   0.563291   0.060547   0.997074   0.817607        0.0   0.587079   \n",
       "\n",
       "    var6(t-1)  var7(t-1)   var1(t)  \n",
       "10        0.0   0.563291  0.060547  \n",
       "11        0.0   0.563291  0.060547  \n",
       "12        0.0   0.563291  0.060547  \n",
       "13        0.0   0.563291  0.060547  \n",
       "14        0.0   0.563291  0.060547  \n",
       "\n",
       "[5 rows x 71 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# frame as supervised learning with scaled values time shifted to 10 steps\n",
    "lag_steps = 10 # n_in value given as input attribute to series_to_supervised function\n",
    "reframed = series_to_supervised(scaled, lag_steps, 1)\n",
    "\n",
    "# drop columns we don't want to predict\n",
    "reframed.drop(reframed.columns[[71,72,73,74,75,76]], axis=1, inplace=True)\n",
    "\n",
    "reframed.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9729, 71)\n"
     ]
    }
   ],
   "source": [
    "values = reframed.values\n",
    "print(values.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6779, 71) (2950, 71)\n"
     ]
    }
   ],
   "source": [
    "# Manually defining row split number which is equal last row of same flight and also roughly split at 70% of the data available\n",
    "train_test_split_row_nr = 6779 \n",
    "\n",
    "# split into train and test sets\n",
    "train = values[:train_test_split_row_nr, :]\n",
    "test = values[train_test_split_row_nr:, :]\n",
    "print(train.shape, test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6779, 70) (6779, 1) (2950, 70) (2950, 1)\n"
     ]
    }
   ],
   "source": [
    "# split into input and outputs\n",
    "train_X, train_y = train[:, :-1], train[:,-1:]\n",
    "test_X, test_y = test[:, :-1], test[:, -1:]\n",
    "print(train_X.shape,train_y.shape, test_X.shape, test_y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6779, 10, 7) (6779, 1) (2950, 10, 7) (2950, 1)\n"
     ]
    }
   ],
   "source": [
    "n_features = 7\n",
    "# reshape input to be 3D [samples, timesteps, features]\n",
    "train_X = train_X.reshape((train_X.shape[0], lag_steps, n_features))\n",
    "test_X = test_X.reshape((test_X.shape[0], lag_steps, n_features))\n",
    "print(train_X.shape, train_y.shape, test_X.shape, test_y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 6779 samples, validate on 2950 samples\n",
      "Epoch 1/500\n",
      "6779/6779 [==============================] - 2s 310us/step - loss: 0.1925 - val_loss: 0.3037\n",
      "Epoch 2/500\n",
      "6779/6779 [==============================] - 1s 145us/step - loss: 0.1654 - val_loss: 0.2183\n",
      "Epoch 3/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.1252 - val_loss: 0.1777\n",
      "Epoch 4/500\n",
      "6779/6779 [==============================] - 1s 134us/step - loss: 0.0819 - val_loss: 0.1439\n",
      "Epoch 5/500\n",
      "6779/6779 [==============================] - 1s 165us/step - loss: 0.0721 - val_loss: 0.0927\n",
      "Epoch 6/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0628 - val_loss: 0.0553\n",
      "Epoch 7/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0567 - val_loss: 0.0953\n",
      "Epoch 8/500\n",
      "6779/6779 [==============================] - 1s 153us/step - loss: 0.0479 - val_loss: 0.0360\n",
      "Epoch 9/500\n",
      "6779/6779 [==============================] - 2s 236us/step - loss: 0.0321 - val_loss: 0.1021\n",
      "Epoch 10/500\n",
      "6779/6779 [==============================] - 1s 155us/step - loss: 0.0451 - val_loss: 0.0550\n",
      "Epoch 11/500\n",
      "6779/6779 [==============================] - 1s 136us/step - loss: 0.0424 - val_loss: 0.0356\n",
      "Epoch 12/500\n",
      "6779/6779 [==============================] - 1s 143us/step - loss: 0.0274 - val_loss: 0.0750\n",
      "Epoch 13/500\n",
      "6779/6779 [==============================] - 1s 147us/step - loss: 0.0249 - val_loss: 0.0139\n",
      "Epoch 14/500\n",
      "6779/6779 [==============================] - 1s 138us/step - loss: 0.0140 - val_loss: 0.0356\n",
      "Epoch 15/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0222 - val_loss: 0.0485\n",
      "Epoch 16/500\n",
      "6779/6779 [==============================] - 1s 141us/step - loss: 0.0476 - val_loss: 0.1159\n",
      "Epoch 17/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0349 - val_loss: 0.0583\n",
      "Epoch 18/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0364 - val_loss: 0.0388\n",
      "Epoch 19/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0227 - val_loss: 0.0168\n",
      "Epoch 20/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0195 - val_loss: 0.0417\n",
      "Epoch 21/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0230 - val_loss: 0.0314\n",
      "Epoch 22/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0214 - val_loss: 0.0358\n",
      "Epoch 23/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0283 - val_loss: 0.0192\n",
      "Epoch 24/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0206 - val_loss: 0.0493\n",
      "Epoch 25/500\n",
      "6779/6779 [==============================] - ETA: 0s - loss: 0.026 - 1s 108us/step - loss: 0.0261 - val_loss: 0.0332\n",
      "Epoch 26/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0337 - val_loss: 0.1168\n",
      "Epoch 27/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0356 - val_loss: 0.0237\n",
      "Epoch 28/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0149 - val_loss: 0.0338\n",
      "Epoch 29/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0195 - val_loss: 0.0239\n",
      "Epoch 30/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0186 - val_loss: 0.0316\n",
      "Epoch 31/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0152 - val_loss: 0.0124\n",
      "Epoch 32/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0170 - val_loss: 0.0191\n",
      "Epoch 33/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0147 - val_loss: 0.0292\n",
      "Epoch 34/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0195 - val_loss: 0.0275\n",
      "Epoch 35/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0191 - val_loss: 0.0168\n",
      "Epoch 36/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0087 - val_loss: 0.0341\n",
      "Epoch 37/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0239 - val_loss: 0.0103\n",
      "Epoch 38/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0163 - val_loss: 0.0403\n",
      "Epoch 39/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0184 - val_loss: 0.0146\n",
      "Epoch 40/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0135 - val_loss: 0.0261\n",
      "Epoch 41/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0158 - val_loss: 0.0382\n",
      "Epoch 42/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0200 - val_loss: 0.0240\n",
      "Epoch 43/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0162 - val_loss: 0.0539\n",
      "Epoch 44/500\n",
      "6779/6779 [==============================] - 1s 138us/step - loss: 0.0158 - val_loss: 0.0245\n",
      "Epoch 45/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0211 - val_loss: 0.0306\n",
      "Epoch 46/500\n",
      "6779/6779 [==============================] - 1s 152us/step - loss: 0.0108 - val_loss: 0.0108\n",
      "Epoch 47/500\n",
      "6779/6779 [==============================] - 1s 152us/step - loss: 0.0179 - val_loss: 0.0472\n",
      "Epoch 48/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0182 - val_loss: 0.0217\n",
      "Epoch 49/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0190 - val_loss: 0.0275\n",
      "Epoch 50/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0142 - val_loss: 0.0525\n",
      "Epoch 51/500\n",
      "6779/6779 [==============================] - 1s 140us/step - loss: 0.0214 - val_loss: 0.0171\n",
      "Epoch 52/500\n",
      "6779/6779 [==============================] - 1s 152us/step - loss: 0.0141 - val_loss: 0.0282\n",
      "Epoch 53/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0114 - val_loss: 0.0274\n",
      "Epoch 54/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0158 - val_loss: 0.0250\n",
      "Epoch 55/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0141 - val_loss: 0.0469\n",
      "Epoch 56/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0192 - val_loss: 0.0196\n",
      "Epoch 57/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0143 - val_loss: 0.0105\n",
      "Epoch 58/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0145 - val_loss: 0.0294\n",
      "Epoch 59/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0154 - val_loss: 0.0357\n",
      "Epoch 60/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0200 - val_loss: 0.0178\n",
      "Epoch 61/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0131 - val_loss: 0.0229\n",
      "Epoch 62/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0134 - val_loss: 0.0241\n",
      "Epoch 63/500\n",
      "6779/6779 [==============================] - 1s 106us/step - loss: 0.0132 - val_loss: 0.0089\n",
      "Epoch 64/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0093 - val_loss: 0.0087\n",
      "Epoch 65/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0132 - val_loss: 0.0099\n",
      "Epoch 66/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0097 - val_loss: 0.0100\n",
      "Epoch 67/500\n",
      "6779/6779 [==============================] - 1s 155us/step - loss: 0.0114 - val_loss: 0.0126\n",
      "Epoch 68/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0172 - val_loss: 0.0265\n",
      "Epoch 69/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0151 - val_loss: 0.0163\n",
      "Epoch 70/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0118 - val_loss: 0.0121\n",
      "Epoch 71/500\n",
      "6779/6779 [==============================] - 1s 162us/step - loss: 0.0121 - val_loss: 0.0309\n",
      "Epoch 72/500\n",
      "6779/6779 [==============================] - 1s 191us/step - loss: 0.0169 - val_loss: 0.0163\n",
      "Epoch 73/500\n",
      "6779/6779 [==============================] - 1s 156us/step - loss: 0.0134 - val_loss: 0.0143\n",
      "Epoch 74/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0103 - val_loss: 0.0436\n",
      "Epoch 75/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0146 - val_loss: 0.0150\n",
      "Epoch 76/500\n",
      "6779/6779 [==============================] - 1s 135us/step - loss: 0.0110 - val_loss: 0.0206\n",
      "Epoch 77/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0147 - val_loss: 0.0107\n",
      "Epoch 78/500\n",
      "6779/6779 [==============================] - 1s 143us/step - loss: 0.0126 - val_loss: 0.0400\n",
      "Epoch 79/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0175 - val_loss: 0.0168\n",
      "Epoch 80/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0180 - val_loss: 0.0129\n",
      "Epoch 81/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0161 - val_loss: 0.0063\n",
      "Epoch 82/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0092 - val_loss: 0.0114\n",
      "Epoch 83/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0139 - val_loss: 0.0347\n",
      "Epoch 84/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0167 - val_loss: 0.0264\n",
      "Epoch 85/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0098 - val_loss: 0.0110\n",
      "Epoch 86/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0104 - val_loss: 0.0080\n",
      "Epoch 87/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0094 - val_loss: 0.0124\n",
      "Epoch 88/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0106 - val_loss: 0.0125\n",
      "Epoch 89/500\n",
      "6779/6779 [==============================] - 1s 147us/step - loss: 0.0114 - val_loss: 0.0097\n",
      "Epoch 90/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0116 - val_loss: 0.0231\n",
      "Epoch 91/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0099 - val_loss: 0.0094\n",
      "Epoch 92/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0085 - val_loss: 0.0164\n",
      "Epoch 93/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0107 - val_loss: 0.0093\n",
      "Epoch 94/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0078 - val_loss: 0.0163\n",
      "Epoch 95/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0130 - val_loss: 0.0079\n",
      "Epoch 96/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0135 - val_loss: 0.0134\n",
      "Epoch 97/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0106 - val_loss: 0.0107\n",
      "Epoch 98/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0092 - val_loss: 0.0043\n",
      "Epoch 99/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0096 - val_loss: 0.0199\n",
      "Epoch 100/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0101 - val_loss: 0.0153\n",
      "Epoch 101/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0097 - val_loss: 0.0085\n",
      "Epoch 102/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0126 - val_loss: 0.0234\n",
      "Epoch 103/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0105 - val_loss: 0.0118\n",
      "Epoch 104/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0086 - val_loss: 0.0099\n",
      "Epoch 105/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0108 - val_loss: 0.0406\n",
      "Epoch 106/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0162 - val_loss: 0.0118\n",
      "Epoch 107/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0113 - val_loss: 0.0161\n",
      "Epoch 108/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0111 - val_loss: 0.0093\n",
      "Epoch 109/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0086 - val_loss: 0.0294\n",
      "Epoch 110/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0173 - val_loss: 0.0075\n",
      "Epoch 111/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0115 - val_loss: 0.0133\n",
      "Epoch 112/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0112 - val_loss: 0.0078\n",
      "Epoch 113/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0099 - val_loss: 0.0079\n",
      "Epoch 114/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0114 - val_loss: 0.0141\n",
      "Epoch 115/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0133 - val_loss: 0.0136\n",
      "Epoch 116/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0119 - val_loss: 0.0371\n",
      "Epoch 117/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0100 - val_loss: 0.0169\n",
      "Epoch 118/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0101 - val_loss: 0.0280\n",
      "Epoch 119/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0132 - val_loss: 0.0063\n",
      "Epoch 120/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0073 - val_loss: 0.0087\n",
      "Epoch 121/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0114 - val_loss: 0.0222\n",
      "Epoch 122/500\n",
      "6779/6779 [==============================] - 1s 131us/step - loss: 0.0140 - val_loss: 0.0216\n",
      "Epoch 123/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0096 - val_loss: 0.0307\n",
      "Epoch 124/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0088 - val_loss: 0.0187\n",
      "Epoch 125/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0078 - val_loss: 0.0059\n",
      "Epoch 126/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0065 - val_loss: 0.0204\n",
      "Epoch 127/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0156 - val_loss: 0.0301\n",
      "Epoch 128/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0111 - val_loss: 0.0179\n",
      "Epoch 129/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0128 - val_loss: 0.0205\n",
      "Epoch 130/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0123 - val_loss: 0.0253\n",
      "Epoch 131/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0110 - val_loss: 0.0238\n",
      "Epoch 132/500\n",
      "6779/6779 [==============================] - 1s 136us/step - loss: 0.0098 - val_loss: 0.0083\n",
      "Epoch 133/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0080 - val_loss: 0.0084\n",
      "Epoch 134/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0063 - val_loss: 0.0050\n",
      "Epoch 135/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0050 - val_loss: 0.0107\n",
      "Epoch 136/500\n",
      "6779/6779 [==============================] - 1s 156us/step - loss: 0.0139 - val_loss: 0.0081\n",
      "Epoch 137/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0079 - val_loss: 0.0092\n",
      "Epoch 138/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0066 - val_loss: 0.0361\n",
      "Epoch 139/500\n",
      "6779/6779 [==============================] - 1s 142us/step - loss: 0.0107 - val_loss: 0.0144\n",
      "Epoch 140/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0119 - val_loss: 0.0056\n",
      "Epoch 141/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0057 - val_loss: 0.0050\n",
      "Epoch 142/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0058 - val_loss: 0.0073\n",
      "Epoch 143/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0073 - val_loss: 0.0041\n",
      "Epoch 144/500\n",
      "6779/6779 [==============================] - 1s 138us/step - loss: 0.0053 - val_loss: 0.0086\n",
      "Epoch 145/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0092 - val_loss: 0.0125\n",
      "Epoch 146/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0091 - val_loss: 0.0114\n",
      "Epoch 147/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0097 - val_loss: 0.0089\n",
      "Epoch 148/500\n",
      "6779/6779 [==============================] - 1s 144us/step - loss: 0.0089 - val_loss: 0.0224\n",
      "Epoch 149/500\n",
      "6779/6779 [==============================] - 1s 152us/step - loss: 0.0112 - val_loss: 0.0140\n",
      "Epoch 150/500\n",
      "6779/6779 [==============================] - 1s 159us/step - loss: 0.0078 - val_loss: 0.0032\n",
      "Epoch 151/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0069 - val_loss: 0.0122\n",
      "Epoch 152/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0072 - val_loss: 0.0041\n",
      "Epoch 153/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0052 - val_loss: 0.0274\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 154/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0108 - val_loss: 0.0146\n",
      "Epoch 155/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0088 - val_loss: 0.0279\n",
      "Epoch 156/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0102 - val_loss: 0.0133\n",
      "Epoch 157/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0104 - val_loss: 0.0180\n",
      "Epoch 158/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0101 - val_loss: 0.0066\n",
      "Epoch 159/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0105 - val_loss: 0.0110\n",
      "Epoch 160/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0074 - val_loss: 0.0191\n",
      "Epoch 161/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0125 - val_loss: 0.0161\n",
      "Epoch 162/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0073 - val_loss: 0.0092\n",
      "Epoch 163/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0096 - val_loss: 0.0101\n",
      "Epoch 164/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0125 - val_loss: 0.0092\n",
      "Epoch 165/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0115 - val_loss: 0.0026\n",
      "Epoch 166/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0070 - val_loss: 0.0087\n",
      "Epoch 167/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0068 - val_loss: 0.0083\n",
      "Epoch 168/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0102 - val_loss: 0.0103\n",
      "Epoch 169/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0094 - val_loss: 0.0046\n",
      "Epoch 170/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0060 - val_loss: 0.0145\n",
      "Epoch 171/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0080 - val_loss: 0.0122\n",
      "Epoch 172/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0092 - val_loss: 0.0095\n",
      "Epoch 173/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0133 - val_loss: 0.0251\n",
      "Epoch 174/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0105 - val_loss: 0.0114\n",
      "Epoch 175/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0068 - val_loss: 0.0051\n",
      "Epoch 176/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0057 - val_loss: 0.0122\n",
      "Epoch 177/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0150 - val_loss: 0.0065\n",
      "Epoch 178/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0065 - val_loss: 0.0131\n",
      "Epoch 179/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0094 - val_loss: 0.0109\n",
      "Epoch 180/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0097 - val_loss: 0.0173\n",
      "Epoch 181/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0068 - val_loss: 0.0115\n",
      "Epoch 182/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0117 - val_loss: 0.0091\n",
      "Epoch 183/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0062 - val_loss: 0.0056\n",
      "Epoch 184/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0094 - val_loss: 0.0106\n",
      "Epoch 185/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0072 - val_loss: 0.0057\n",
      "Epoch 186/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0085 - val_loss: 0.0198\n",
      "Epoch 187/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0067 - val_loss: 0.0124\n",
      "Epoch 188/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0107 - val_loss: 0.0110\n",
      "Epoch 189/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0074 - val_loss: 0.0231\n",
      "Epoch 190/500\n",
      "6779/6779 [==============================] - 1s 170us/step - loss: 0.0087 - val_loss: 0.0182\n",
      "Epoch 191/500\n",
      "6779/6779 [==============================] - 1s 192us/step - loss: 0.0117 - val_loss: 0.0068\n",
      "Epoch 192/500\n",
      "6779/6779 [==============================] - 1s 194us/step - loss: 0.0065 - val_loss: 0.0050\n",
      "Epoch 193/500\n",
      "6779/6779 [==============================] - 1s 204us/step - loss: 0.0081 - val_loss: 0.0159\n",
      "Epoch 194/500\n",
      "6779/6779 [==============================] - 1s 175us/step - loss: 0.0083 - val_loss: 0.0070\n",
      "Epoch 195/500\n",
      "6779/6779 [==============================] - 1s 144us/step - loss: 0.0071 - val_loss: 0.0086\n",
      "Epoch 196/500\n",
      "6779/6779 [==============================] - 1s 149us/step - loss: 0.0059 - val_loss: 0.0186\n",
      "Epoch 197/500\n",
      "6779/6779 [==============================] - 1s 140us/step - loss: 0.0108 - val_loss: 0.0064\n",
      "Epoch 198/500\n",
      "6779/6779 [==============================] - 1s 141us/step - loss: 0.0100 - val_loss: 0.0149\n",
      "Epoch 199/500\n",
      "6779/6779 [==============================] - 1s 132us/step - loss: 0.0113 - val_loss: 0.0080\n",
      "Epoch 200/500\n",
      "6779/6779 [==============================] - 1s 137us/step - loss: 0.0084 - val_loss: 0.0032\n",
      "Epoch 201/500\n",
      "6779/6779 [==============================] - 1s 144us/step - loss: 0.0050 - val_loss: 0.0047\n",
      "Epoch 202/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0049 - val_loss: 0.0171\n",
      "Epoch 203/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0107 - val_loss: 0.0091\n",
      "Epoch 204/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0087 - val_loss: 0.0042\n",
      "Epoch 205/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0074 - val_loss: 0.0215\n",
      "Epoch 206/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0100 - val_loss: 0.0068\n",
      "Epoch 207/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0057 - val_loss: 0.0125\n",
      "Epoch 208/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0072 - val_loss: 0.0067\n",
      "Epoch 209/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0077 - val_loss: 0.0110\n",
      "Epoch 210/500\n",
      "6779/6779 [==============================] - 1s 137us/step - loss: 0.0063 - val_loss: 0.0133\n",
      "Epoch 211/500\n",
      "6779/6779 [==============================] - 1s 134us/step - loss: 0.0068 - val_loss: 0.0056\n",
      "Epoch 212/500\n",
      "6779/6779 [==============================] - 1s 147us/step - loss: 0.0064 - val_loss: 0.0109\n",
      "Epoch 213/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0053 - val_loss: 0.0142\n",
      "Epoch 214/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0082 - val_loss: 0.0051\n",
      "Epoch 215/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0062 - val_loss: 0.0203\n",
      "Epoch 216/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0091 - val_loss: 0.0173\n",
      "Epoch 217/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0104 - val_loss: 0.0097\n",
      "Epoch 218/500\n",
      "6779/6779 [==============================] - 1s 136us/step - loss: 0.0067 - val_loss: 0.0089\n",
      "Epoch 219/500\n",
      "6779/6779 [==============================] - 1s 145us/step - loss: 0.0092 - val_loss: 0.0135\n",
      "Epoch 220/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0073 - val_loss: 0.0241\n",
      "Epoch 221/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0070 - val_loss: 0.0390\n",
      "Epoch 222/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0140 - val_loss: 0.0109\n",
      "Epoch 223/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0119 - val_loss: 0.0058\n",
      "Epoch 224/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0067 - val_loss: 0.0049\n",
      "Epoch 225/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0050 - val_loss: 0.0128\n",
      "Epoch 226/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0112 - val_loss: 0.0043\n",
      "Epoch 227/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0044 - val_loss: 0.0165\n",
      "Epoch 228/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0107 - val_loss: 0.0078\n",
      "Epoch 229/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0064 - val_loss: 0.0064\n",
      "Epoch 230/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0064 - val_loss: 0.0039\n",
      "Epoch 231/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0090 - val_loss: 0.0053\n",
      "Epoch 232/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0050 - val_loss: 0.0088\n",
      "Epoch 233/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0080 - val_loss: 0.0079\n",
      "Epoch 234/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0099 - val_loss: 0.0116\n",
      "Epoch 235/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0079 - val_loss: 0.0057\n",
      "Epoch 236/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0068 - val_loss: 0.0039\n",
      "Epoch 237/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0052 - val_loss: 0.0143\n",
      "Epoch 238/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0097 - val_loss: 0.0129\n",
      "Epoch 239/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0099 - val_loss: 0.0165\n",
      "Epoch 240/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0060 - val_loss: 0.0059\n",
      "Epoch 241/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0070 - val_loss: 0.0050\n",
      "Epoch 242/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0067 - val_loss: 0.0148\n",
      "Epoch 243/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0128 - val_loss: 0.0042\n",
      "Epoch 244/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0043 - val_loss: 0.0084\n",
      "Epoch 245/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0064 - val_loss: 0.0217\n",
      "Epoch 246/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0080 - val_loss: 0.0125\n",
      "Epoch 247/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0057 - val_loss: 0.0102\n",
      "Epoch 248/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0072 - val_loss: 0.0072\n",
      "Epoch 249/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0094 - val_loss: 0.0110\n",
      "Epoch 250/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0071 - val_loss: 0.0070\n",
      "Epoch 251/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0090 - val_loss: 0.0086\n",
      "Epoch 252/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0094 - val_loss: 0.0319\n",
      "Epoch 253/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0124 - val_loss: 0.0149\n",
      "Epoch 254/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0107 - val_loss: 0.0084\n",
      "Epoch 255/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0044 - val_loss: 0.0163\n",
      "Epoch 256/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0062 - val_loss: 0.0195\n",
      "Epoch 257/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0072 - val_loss: 0.0180\n",
      "Epoch 258/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0101 - val_loss: 0.0088\n",
      "Epoch 259/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0080 - val_loss: 0.0149\n",
      "Epoch 260/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0079 - val_loss: 0.0201\n",
      "Epoch 261/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0089 - val_loss: 0.0098\n",
      "Epoch 262/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0054 - val_loss: 0.0128\n",
      "Epoch 263/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0060 - val_loss: 0.0129\n",
      "Epoch 264/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0065 - val_loss: 0.0140\n",
      "Epoch 265/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0087 - val_loss: 0.0050\n",
      "Epoch 266/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0045 - val_loss: 0.0171\n",
      "Epoch 267/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0069 - val_loss: 0.0078\n",
      "Epoch 268/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0048 - val_loss: 0.0225\n",
      "Epoch 269/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0066 - val_loss: 0.0103\n",
      "Epoch 270/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0041 - val_loss: 0.0070\n",
      "Epoch 271/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0078 - val_loss: 0.0141\n",
      "Epoch 272/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0101 - val_loss: 0.0074\n",
      "Epoch 273/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0038 - val_loss: 0.0105\n",
      "Epoch 274/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0098 - val_loss: 0.0074\n",
      "Epoch 275/500\n",
      "6779/6779 [==============================] - 1s 134us/step - loss: 0.0042 - val_loss: 0.0115\n",
      "Epoch 276/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0133 - val_loss: 0.0154\n",
      "Epoch 277/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0100 - val_loss: 0.0092\n",
      "Epoch 278/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0060 - val_loss: 0.0082\n",
      "Epoch 279/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0062 - val_loss: 0.0150\n",
      "Epoch 280/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0115 - val_loss: 0.0221\n",
      "Epoch 281/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0060 - val_loss: 0.0118\n",
      "Epoch 282/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0066 - val_loss: 0.0039\n",
      "Epoch 283/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0044 - val_loss: 0.0084\n",
      "Epoch 284/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0060 - val_loss: 0.0060\n",
      "Epoch 285/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0053 - val_loss: 0.0054\n",
      "Epoch 286/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0044 - val_loss: 0.0124\n",
      "Epoch 287/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0073 - val_loss: 0.0091\n",
      "Epoch 288/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0068 - val_loss: 0.0254\n",
      "Epoch 289/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0114 - val_loss: 0.0142\n",
      "Epoch 290/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0114 - val_loss: 0.0150\n",
      "Epoch 291/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0061 - val_loss: 0.0218\n",
      "Epoch 292/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0083 - val_loss: 0.0143\n",
      "Epoch 293/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0068 - val_loss: 0.0072\n",
      "Epoch 294/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0062 - val_loss: 0.0061\n",
      "Epoch 295/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0042 - val_loss: 0.0156\n",
      "Epoch 296/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0089 - val_loss: 0.0052\n",
      "Epoch 297/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0078 - val_loss: 0.0173\n",
      "Epoch 298/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0059 - val_loss: 0.0059\n",
      "Epoch 299/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0057 - val_loss: 0.0179\n",
      "Epoch 300/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0075 - val_loss: 0.0181\n",
      "Epoch 301/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0066 - val_loss: 0.0161\n",
      "Epoch 302/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0080 - val_loss: 0.0051\n",
      "Epoch 303/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0063 - val_loss: 0.0051\n",
      "Epoch 304/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0076 - val_loss: 0.0109\n",
      "Epoch 305/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0060 - val_loss: 0.0144\n",
      "Epoch 306/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0060 - val_loss: 0.0051\n",
      "Epoch 307/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0049 - val_loss: 0.0053\n",
      "Epoch 308/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0056 - val_loss: 0.0102\n",
      "Epoch 309/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0074 - val_loss: 0.0183\n",
      "Epoch 310/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0097 - val_loss: 0.0046\n",
      "Epoch 311/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0055 - val_loss: 0.0107\n",
      "Epoch 312/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0058 - val_loss: 0.0129\n",
      "Epoch 313/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0061 - val_loss: 0.0078\n",
      "Epoch 314/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0071 - val_loss: 0.0144\n",
      "Epoch 315/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0089 - val_loss: 0.0071\n",
      "Epoch 316/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0072 - val_loss: 0.0135\n",
      "Epoch 317/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0102 - val_loss: 0.0064\n",
      "Epoch 318/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0062 - val_loss: 0.0095\n",
      "Epoch 319/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0071 - val_loss: 0.0069\n",
      "Epoch 320/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0048 - val_loss: 0.0144\n",
      "Epoch 321/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0120 - val_loss: 0.0059\n",
      "Epoch 322/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0045 - val_loss: 0.0049\n",
      "Epoch 323/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0061 - val_loss: 0.0152\n",
      "Epoch 324/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0052 - val_loss: 0.0078\n",
      "Epoch 325/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0058 - val_loss: 0.0186\n",
      "Epoch 326/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0103 - val_loss: 0.0151\n",
      "Epoch 327/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0108 - val_loss: 0.0080\n",
      "Epoch 328/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0066 - val_loss: 0.0054\n",
      "Epoch 329/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0081 - val_loss: 0.0097\n",
      "Epoch 330/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0080 - val_loss: 0.0123\n",
      "Epoch 331/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0066 - val_loss: 0.0156\n",
      "Epoch 332/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0077 - val_loss: 0.0060\n",
      "Epoch 333/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0049 - val_loss: 0.0049\n",
      "Epoch 334/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0070 - val_loss: 0.0120\n",
      "Epoch 335/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0060 - val_loss: 0.0094\n",
      "Epoch 336/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0058 - val_loss: 0.0112\n",
      "Epoch 337/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0094 - val_loss: 0.0142\n",
      "Epoch 338/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0105 - val_loss: 0.0095\n",
      "Epoch 339/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0069 - val_loss: 0.0068\n",
      "Epoch 340/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0067 - val_loss: 0.0107\n",
      "Epoch 341/500\n",
      "6779/6779 [==============================] - 1s 106us/step - loss: 0.0052 - val_loss: 0.0109\n",
      "Epoch 342/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0103 - val_loss: 0.0044\n",
      "Epoch 343/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0096 - val_loss: 0.0033\n",
      "Epoch 344/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0049 - val_loss: 0.0141\n",
      "Epoch 345/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0075 - val_loss: 0.0171\n",
      "Epoch 346/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0065 - val_loss: 0.0072\n",
      "Epoch 347/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0077 - val_loss: 0.0035\n",
      "Epoch 348/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0042 - val_loss: 0.0057\n",
      "Epoch 349/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0040 - val_loss: 0.0070\n",
      "Epoch 350/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0068 - val_loss: 0.0125\n",
      "Epoch 351/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0066 - val_loss: 0.0248\n",
      "Epoch 352/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0096 - val_loss: 0.0083\n",
      "Epoch 353/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0081 - val_loss: 0.0038\n",
      "Epoch 354/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0039 - val_loss: 0.0149\n",
      "Epoch 355/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0084 - val_loss: 0.0127\n",
      "Epoch 356/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0090 - val_loss: 0.0205\n",
      "Epoch 357/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0100 - val_loss: 0.0045\n",
      "Epoch 358/500\n",
      "6779/6779 [==============================] - 1s 107us/step - loss: 0.0072 - val_loss: 0.0080\n",
      "Epoch 359/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0080 - val_loss: 0.0035\n",
      "Epoch 360/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0048 - val_loss: 0.0176\n",
      "Epoch 361/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0083 - val_loss: 0.0101\n",
      "Epoch 362/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0075 - val_loss: 0.0086\n",
      "Epoch 363/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0056 - val_loss: 0.0054\n",
      "Epoch 364/500\n",
      "6779/6779 [==============================] - 1s 109us/step - loss: 0.0071 - val_loss: 0.0142\n",
      "Epoch 365/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0066 - val_loss: 0.0092\n",
      "Epoch 366/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0087 - val_loss: 0.0062\n",
      "Epoch 367/500\n",
      "6779/6779 [==============================] - 1s 108us/step - loss: 0.0051 - val_loss: 0.0125\n",
      "Epoch 368/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0049 - val_loss: 0.0134\n",
      "Epoch 369/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0060 - val_loss: 0.0065\n",
      "Epoch 370/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0050 - val_loss: 0.0077\n",
      "Epoch 371/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0072 - val_loss: 0.0056\n",
      "Epoch 372/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0045 - val_loss: 0.0048\n",
      "Epoch 373/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0043 - val_loss: 0.0053\n",
      "Epoch 374/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0053 - val_loss: 0.0116\n",
      "Epoch 375/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0076 - val_loss: 0.0123\n",
      "Epoch 376/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0062 - val_loss: 0.0075\n",
      "Epoch 377/500\n",
      "6779/6779 [==============================] - 1s 137us/step - loss: 0.0049 - val_loss: 0.0040\n",
      "Epoch 378/500\n",
      "6779/6779 [==============================] - 1s 136us/step - loss: 0.0038 - val_loss: 0.0071\n",
      "Epoch 379/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0078 - val_loss: 0.0056\n",
      "Epoch 380/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0069 - val_loss: 0.0036\n",
      "Epoch 381/500\n",
      "6779/6779 [==============================] - 1s 131us/step - loss: 0.0050 - val_loss: 0.0094\n",
      "Epoch 382/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0051 - val_loss: 0.0120\n",
      "Epoch 383/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0063 - val_loss: 0.0083\n",
      "Epoch 384/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0064 - val_loss: 0.0111\n",
      "Epoch 385/500\n",
      "6779/6779 [==============================] - 1s 131us/step - loss: 0.0080 - val_loss: 0.0053\n",
      "Epoch 386/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0040 - val_loss: 0.0061\n",
      "Epoch 387/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0045 - val_loss: 0.0208\n",
      "Epoch 388/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0090 - val_loss: 0.0198\n",
      "Epoch 389/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0070 - val_loss: 0.0111\n",
      "Epoch 390/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0070 - val_loss: 0.0041\n",
      "Epoch 391/500\n",
      "6779/6779 [==============================] - 1s 135us/step - loss: 0.0040 - val_loss: 0.0060\n",
      "Epoch 392/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0053 - val_loss: 0.0033\n",
      "Epoch 393/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0039 - val_loss: 0.0084\n",
      "Epoch 394/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0108 - val_loss: 0.0111\n",
      "Epoch 395/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0079 - val_loss: 0.0043\n",
      "Epoch 396/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0040 - val_loss: 0.0111\n",
      "Epoch 397/500\n",
      "6779/6779 [==============================] - 1s 139us/step - loss: 0.0067 - val_loss: 0.0130\n",
      "Epoch 398/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0070 - val_loss: 0.0139\n",
      "Epoch 399/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0101 - val_loss: 0.0033\n",
      "Epoch 400/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0049 - val_loss: 0.0051\n",
      "Epoch 401/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0048 - val_loss: 0.0040\n",
      "Epoch 402/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0065 - val_loss: 0.0039\n",
      "Epoch 403/500\n",
      "6779/6779 [==============================] - 1s 135us/step - loss: 0.0048 - val_loss: 0.0182\n",
      "Epoch 404/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0071 - val_loss: 0.0109\n",
      "Epoch 405/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0070 - val_loss: 0.0127\n",
      "Epoch 406/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0080 - val_loss: 0.0045\n",
      "Epoch 407/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0048 - val_loss: 0.0056\n",
      "Epoch 408/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0058 - val_loss: 0.0044\n",
      "Epoch 409/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0049 - val_loss: 0.0083\n",
      "Epoch 410/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0060 - val_loss: 0.0075\n",
      "Epoch 411/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0079 - val_loss: 0.0136\n",
      "Epoch 412/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0056 - val_loss: 0.0130\n",
      "Epoch 413/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0075 - val_loss: 0.0067\n",
      "Epoch 414/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0047 - val_loss: 0.0107\n",
      "Epoch 415/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0053 - val_loss: 0.0095\n",
      "Epoch 416/500\n",
      "6779/6779 [==============================] - 1s 140us/step - loss: 0.0069 - val_loss: 0.0164\n",
      "Epoch 417/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0056 - val_loss: 0.0056\n",
      "Epoch 418/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0053 - val_loss: 0.0086\n",
      "Epoch 419/500\n",
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0079 - val_loss: 0.0090\n",
      "Epoch 420/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0052 - val_loss: 0.0057\n",
      "Epoch 421/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0056 - val_loss: 0.0123\n",
      "Epoch 422/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0052 - val_loss: 0.0139\n",
      "Epoch 423/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0074 - val_loss: 0.0050\n",
      "Epoch 424/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0047 - val_loss: 0.0059\n",
      "Epoch 425/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0058 - val_loss: 0.0147\n",
      "Epoch 426/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0053 - val_loss: 0.0031\n",
      "Epoch 427/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0051 - val_loss: 0.0057\n",
      "Epoch 428/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0061 - val_loss: 0.0040\n",
      "Epoch 429/500\n",
      "6779/6779 [==============================] - 1s 137us/step - loss: 0.0046 - val_loss: 0.0105\n",
      "Epoch 430/500\n",
      "6779/6779 [==============================] - 1s 128us/step - loss: 0.0078 - val_loss: 0.0177\n",
      "Epoch 431/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0079 - val_loss: 0.0100\n",
      "Epoch 432/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0056 - val_loss: 0.0081\n",
      "Epoch 433/500\n",
      "6779/6779 [==============================] - 1s 122us/step - loss: 0.0078 - val_loss: 0.0222\n",
      "Epoch 434/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0071 - val_loss: 0.0061\n",
      "Epoch 435/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0045 - val_loss: 0.0105\n",
      "Epoch 436/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0061 - val_loss: 0.0039\n",
      "Epoch 437/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0059 - val_loss: 0.0050\n",
      "Epoch 438/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0086 - val_loss: 0.0120\n",
      "Epoch 439/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0087 - val_loss: 0.0155\n",
      "Epoch 440/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0062 - val_loss: 0.0049\n",
      "Epoch 441/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0055 - val_loss: 0.0036\n",
      "Epoch 442/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0042 - val_loss: 0.0143\n",
      "Epoch 443/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0076 - val_loss: 0.0048\n",
      "Epoch 444/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0051 - val_loss: 0.0101\n",
      "Epoch 445/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0068 - val_loss: 0.0050\n",
      "Epoch 446/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0038 - val_loss: 0.0172\n",
      "Epoch 447/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0057 - val_loss: 0.0109\n",
      "Epoch 448/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0080 - val_loss: 0.0057\n",
      "Epoch 449/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0072 - val_loss: 0.0045\n",
      "Epoch 450/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0052 - val_loss: 0.0110\n",
      "Epoch 451/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0058 - val_loss: 0.0083\n",
      "Epoch 452/500\n",
      "6779/6779 [==============================] - 1s 135us/step - loss: 0.0065 - val_loss: 0.0154\n",
      "Epoch 453/500\n",
      "6779/6779 [==============================] - 1s 120us/step - loss: 0.0084 - val_loss: 0.0037\n",
      "Epoch 454/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0069 - val_loss: 0.0050\n",
      "Epoch 455/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0077 - val_loss: 0.0103\n",
      "Epoch 456/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0047 - val_loss: 0.0075\n",
      "Epoch 457/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0068 - val_loss: 0.0044\n",
      "Epoch 458/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6779/6779 [==============================] - 1s 121us/step - loss: 0.0066 - val_loss: 0.0057\n",
      "Epoch 459/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0066 - val_loss: 0.0112\n",
      "Epoch 460/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0052 - val_loss: 0.0051\n",
      "Epoch 461/500\n",
      "6779/6779 [==============================] - 1s 112us/step - loss: 0.0040 - val_loss: 0.0086\n",
      "Epoch 462/500\n",
      "6779/6779 [==============================] - 1s 140us/step - loss: 0.0072 - val_loss: 0.0060\n",
      "Epoch 463/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0043 - val_loss: 0.0057\n",
      "Epoch 464/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0064 - val_loss: 0.0064\n",
      "Epoch 465/500\n",
      "6779/6779 [==============================] - 1s 113us/step - loss: 0.0044 - val_loss: 0.0046\n",
      "Epoch 466/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0049 - val_loss: 0.0070\n",
      "Epoch 467/500\n",
      "6779/6779 [==============================] - 1s 116us/step - loss: 0.0062 - val_loss: 0.0139\n",
      "Epoch 468/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0075 - val_loss: 0.0036\n",
      "Epoch 469/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0057 - val_loss: 0.0072\n",
      "Epoch 470/500\n",
      "6779/6779 [==============================] - 1s 152us/step - loss: 0.0057 - val_loss: 0.0033\n",
      "Epoch 471/500\n",
      "6779/6779 [==============================] - 1s 150us/step - loss: 0.0039 - val_loss: 0.0105\n",
      "Epoch 472/500\n",
      "6779/6779 [==============================] - 1s 127us/step - loss: 0.0048 - val_loss: 0.0107\n",
      "Epoch 473/500\n",
      "6779/6779 [==============================] - 1s 118us/step - loss: 0.0055 - val_loss: 0.0092\n",
      "Epoch 474/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0078 - val_loss: 0.0282\n",
      "Epoch 475/500\n",
      "6779/6779 [==============================] - 1s 126us/step - loss: 0.0071 - val_loss: 0.0045\n",
      "Epoch 476/500\n",
      "6779/6779 [==============================] - 1s 141us/step - loss: 0.0068 - val_loss: 0.0051\n",
      "Epoch 477/500\n",
      "6779/6779 [==============================] - 1s 136us/step - loss: 0.0045 - val_loss: 0.0125\n",
      "Epoch 478/500\n",
      "6779/6779 [==============================] - 1s 131us/step - loss: 0.0066 - val_loss: 0.0050\n",
      "Epoch 479/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0067 - val_loss: 0.0109\n",
      "Epoch 480/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0077 - val_loss: 0.0063\n",
      "Epoch 481/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0069 - val_loss: 0.0096\n",
      "Epoch 482/500\n",
      "6779/6779 [==============================] - 1s 119us/step - loss: 0.0065 - val_loss: 0.0130\n",
      "Epoch 483/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0076 - val_loss: 0.0253\n",
      "Epoch 484/500\n",
      "6779/6779 [==============================] - 1s 110us/step - loss: 0.0076 - val_loss: 0.0056\n",
      "Epoch 485/500\n",
      "6779/6779 [==============================] - 1s 117us/step - loss: 0.0049 - val_loss: 0.0067\n",
      "Epoch 486/500\n",
      "6779/6779 [==============================] - 1s 115us/step - loss: 0.0052 - val_loss: 0.0115\n",
      "Epoch 487/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0092 - val_loss: 0.0061\n",
      "Epoch 488/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0067 - val_loss: 0.0062\n",
      "Epoch 489/500\n",
      "6779/6779 [==============================] - 1s 140us/step - loss: 0.0047 - val_loss: 0.0098\n",
      "Epoch 490/500\n",
      "6779/6779 [==============================] - 1s 141us/step - loss: 0.0065 - val_loss: 0.0124\n",
      "Epoch 491/500\n",
      "6779/6779 [==============================] - 1s 125us/step - loss: 0.0066 - val_loss: 0.0145\n",
      "Epoch 492/500\n",
      "6779/6779 [==============================] - 1s 111us/step - loss: 0.0061 - val_loss: 0.0071\n",
      "Epoch 493/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0091 - val_loss: 0.0041\n",
      "Epoch 494/500\n",
      "6779/6779 [==============================] - 1s 130us/step - loss: 0.0032 - val_loss: 0.0101\n",
      "Epoch 495/500\n",
      "6779/6779 [==============================] - 1s 133us/step - loss: 0.0053 - val_loss: 0.0112\n",
      "Epoch 496/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0095 - val_loss: 0.0145\n",
      "Epoch 497/500\n",
      "6779/6779 [==============================] - 1s 123us/step - loss: 0.0110 - val_loss: 0.0055\n",
      "Epoch 498/500\n",
      "6779/6779 [==============================] - 1s 124us/step - loss: 0.0042 - val_loss: 0.0080\n",
      "Epoch 499/500\n",
      "6779/6779 [==============================] - 1s 129us/step - loss: 0.0042 - val_loss: 0.0152\n",
      "Epoch 500/500\n",
      "6779/6779 [==============================] - 1s 114us/step - loss: 0.0066 - val_loss: 0.0062\n",
      "Wall time: 6min 54s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# design network\n",
    "model = Sequential()\n",
    "model.add(LSTM(50, input_shape=(train_X.shape[1], train_X.shape[2])))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mae', optimizer='adam')\n",
    "# fit network\n",
    "history = model.fit(train_X, train_y, epochs=500, batch_size=72, validation_data=(test_X, test_y), verbose=1,shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3000x1800 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot history\n",
    "plt.figure(figsize=(20,12),dpi=150)\n",
    "plt.plot(history.history['loss'], label='train')\n",
    "plt.plot(history.history['val_loss'], label='test')\n",
    "plt.legend()\n",
    "plt.title('Training & Validation Loss (MAE) for predicting Lat and Long over each Training Iteration')\n",
    "plt.xlabel('Epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Make Predictions with Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2950, 10, 7)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2950, 1)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make a prediction\n",
    "yhat = model.predict(test_X)\n",
    "inv_yhat = scaler_y.inverse_transform(yhat)\n",
    "inv_yhat.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2950,)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inv_yhat = inv_yhat.flatten()\n",
    "inv_yhat.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2950, 1)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# invert scaling for actual\n",
    "test_y = test_y.reshape((len(test_y), 1))\n",
    "inv_y = scaler_y.inverse_transform(test_y)\n",
    "inv_y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2950,)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inv_y = inv_y.flatten()\n",
    "inv_y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test RMSE: 1242014.809 in milliseconds (UNIX Epoch Format )\n"
     ]
    }
   ],
   "source": [
    "# calculate RMSE\n",
    "rmse = sqrt(mean_squared_error(inv_y, inv_yhat))\n",
    "print('Test RMSE: %.3f' % rmse, 'in milliseconds (UNIX Epoch Format )')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test MAE: 314.000 seconds\n"
     ]
    }
   ],
   "source": [
    "# calculate MAE\n",
    "mae = mean_absolute_error(inv_y, inv_yhat)\n",
    "print('Test MAE: %.3f' % round(mae/1000), 'seconds')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>PredictedTimestamp(ms)</th>\n",
       "      <th>ActualTimestamp(ms)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1539437887488</td>\n",
       "      <td>1539437756416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1539437887488</td>\n",
       "      <td>1539437756416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1539437887488</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1539438018560</td>\n",
       "      <td>1539437887488</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PredictedTimestamp(ms)  ActualTimestamp(ms)\n",
       "0           1539437887488        1539437756416\n",
       "1           1539437887488        1539437756416\n",
       "2           1539437887488        1539437887488\n",
       "3           1539438018560        1539437887488\n",
       "4           1539438018560        1539437887488\n",
       "5           1539438018560        1539437887488\n",
       "6           1539438018560        1539437887488\n",
       "7           1539438018560        1539437887488\n",
       "8           1539438018560        1539437887488\n",
       "9           1539438018560        1539437887488"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pd.set_option('display.float_format', lambda x: '%.1f' % x)\n",
    "columns = ['PredictedTimestamp(ms)','ActualTimestamp(ms)']\n",
    "df_from_arr = pd.DataFrame({'PredictedTimestamp(ms)': inv_yhat, 'ActualTimestamp(ms)': inv_y})\n",
    "df_from_arr['PredictedTimestamp(ms)'] = df_from_arr['PredictedTimestamp(ms)'].apply(np.int64)\n",
    "df_from_arr['ActualTimestamp(ms)'] = df_from_arr['ActualTimestamp(ms)'].apply(np.int64)\n",
    "df_from_arr.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2950 entries, 0 to 2949\n",
      "Data columns (total 2 columns):\n",
      "PredictedTimestamp(ms)    2950 non-null int64\n",
      "ActualTimestamp(ms)       2950 non-null int64\n",
      "dtypes: int64(2)\n",
      "memory usage: 46.2 KB\n"
     ]
    }
   ],
   "source": [
    "df_from_arr.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2950 entries, 0 to 2949\n",
      "Data columns (total 9 columns):\n",
      "postime          2950 non-null int64\n",
      "aircraftreg      2950 non-null object\n",
      "icao             2950 non-null int64\n",
      "lat              2950 non-null float64\n",
      "lon              2950 non-null float64\n",
      "alt              2950 non-null int64\n",
      "course           2950 non-null int64\n",
      "speed            2950 non-null int64\n",
      "verticalspeed    2950 non-null int64\n",
      "dtypes: float64(2), int64(6), object(1)\n",
      "memory usage: 207.5+ KB\n"
     ]
    }
   ],
   "source": [
    "# Retrieve test data from original flightpaths dataset considering the lag = 10 introduced earlier\n",
    "flightpaths=flightpaths[(train_test_split_row_nr + 10):].reset_index(drop=True)\n",
    "flightpaths.info() #Check if it matches with the dimensions (no. of rows) of the predicted dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_from_arr['postime'] = flightpaths.postime.values\n",
    "df_from_arr['icao'] = flightpaths.icao.values\n",
    "df_from_arr['aircraftreg'] = flightpaths.aircraftreg.values\n",
    "df_from_arr['lat'] = flightpaths.lat.values\n",
    "df_from_arr['lon'] = flightpaths.lon.values\n",
    "df_from_arr['alt'] = flightpaths.alt.values\n",
    "df_from_arr['course'] = flightpaths.course.values\n",
    "df_from_arr['speed'] = flightpaths.speed.values\n",
    "df_from_arr['verticalspeed'] = flightpaths.verticalspeed.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>ActualTimestamp(ms)</th>\n",
       "      <th>PredictedTimestamp(ms)</th>\n",
       "      <th>Delta_(P-A)_secs</th>\n",
       "      <th>icao</th>\n",
       "      <th>aircraftreg</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>alt</th>\n",
       "      <th>course</th>\n",
       "      <th>speed</th>\n",
       "      <th>verticalspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1539437807000</td>\n",
       "      <td>1539437887488</td>\n",
       "      <td>80.0</td>\n",
       "      <td>3769770</td>\n",
       "      <td>F-HBNK</td>\n",
       "      <td>48.728519</td>\n",
       "      <td>2.35668</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1539437821000</td>\n",
       "      <td>1539437887488</td>\n",
       "      <td>66.0</td>\n",
       "      <td>3769770</td>\n",
       "      <td>F-HBNK</td>\n",
       "      <td>48.728519</td>\n",
       "      <td>2.35669</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1539437826000</td>\n",
       "      <td>1539437887488</td>\n",
       "      <td>61.0</td>\n",
       "      <td>3769770</td>\n",
       "      <td>F-HBNK</td>\n",
       "      <td>48.728519</td>\n",
       "      <td>2.35668</td>\n",
       "      <td>0</td>\n",
       "      <td>271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1539437837000</td>\n",
       "      <td>1539438018560</td>\n",
       "      <td>182.0</td>\n",
       "      <td>3769770</td>\n",
       "      <td>F-HBNK</td>\n",
       "      <td>48.728512</td>\n",
       "      <td>2.35669</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1539437841000</td>\n",
       "      <td>1539438018560</td>\n",
       "      <td>178.0</td>\n",
       "      <td>3769770</td>\n",
       "      <td>F-HBNK</td>\n",
       "      <td>48.728512</td>\n",
       "      <td>2.35668</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ActualTimestamp(ms)  PredictedTimestamp(ms)  Delta_(P-A)_secs     icao  \\\n",
       "0        1539437807000           1539437887488              80.0  3769770   \n",
       "1        1539437821000           1539437887488              66.0  3769770   \n",
       "2        1539437826000           1539437887488              61.0  3769770   \n",
       "3        1539437837000           1539438018560             182.0  3769770   \n",
       "4        1539437841000           1539438018560             178.0  3769770   \n",
       "\n",
       "  aircraftreg        lat      lon  alt  course  speed  verticalspeed  \n",
       "0      F-HBNK  48.728519  2.35668    0     180      0              0  \n",
       "1      F-HBNK  48.728519  2.35669    0     180      0              0  \n",
       "2      F-HBNK  48.728519  2.35668    0     271      0              0  \n",
       "3      F-HBNK  48.728512  2.35669    0      89      0              0  \n",
       "4      F-HBNK  48.728512  2.35668    0      89      0              0  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_from_arr.drop(columns=['ActualTimestamp(ms)'],inplace = True)\n",
    "df_from_arr.rename(columns={\"postime\": \"ActualTimestamp(ms)\"}, inplace = True)\n",
    "df_from_arr['Delta_(P-A)_secs'] = round((df_from_arr['PredictedTimestamp(ms)'] - df_from_arr['ActualTimestamp(ms)'])/1000)\n",
    "df_from_arr = df_from_arr[['ActualTimestamp(ms)','PredictedTimestamp(ms)','Delta_(P-A)_secs','icao','aircraftreg','lat','lon','alt','course','speed','verticalspeed']]\n",
    "df_from_arr.head()"
   ]
  },
  {
   "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}