{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Rhine level prediction\n",
    "\n",
    "Predict the rhine level in Bonn.\n",
    "\n",
    "[Feedback link for this task](https://beta.ins.uni-bonn.de/feedback/mllab?obj=Rhinelevel)\n",
    "\n",
    "Source for the data:\n",
    "\n",
    "    Wasserstraßen- und Schifffahrtsverwaltung des Bundes (WSV),\n",
    "    bereitgestellt durch die Bundesanstalt für Gewässerkunde (BfG).\n",
    "\n",
    "which (unofficially) translates to\n",
    "\n",
    "    German Federal Waterways and Shipping Administration (WSV),\n",
    "    provided by the German Federal Institute of Hydrology (BfG)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "% matplotlib inline\n",
    "import matplotlib.pyplot as plt \n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from mllab.rhinelevel import wsv, dwd\n",
    "CACHE = './cache'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We load the data 'levels' to plot the stations in a map, 'df' as the 'working' data, 'df_12' as the 'working data' for the 12h prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "levels = wsv.RiverLevelData('data/riverlevels.tar.bz2', CACHE)\n",
    "df = pd.read_pickle(\"riverlevels.pandas.pickle\")\n",
    "df_12 = pd.read_pickle(\"riverlevels.pandas.pickle\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We plot the stations in a map."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "53d05aebbfcb4a5a92116319d839a0c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map(basemap={'url': 'http://{s}.tile.openstreetmap.se/hydda/full/{z}/{x}/{y}.png', 'max_zoom': 18, 'attributio…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import ipyleaflet as leaf\n",
    "m = leaf.Map(center=(50.2, 7.5), zoom=7, basemap=leaf.basemaps.Hydda.Full)\n",
    "markers = []\n",
    "for station in levels.stations():\n",
    "    title = \"{}, river km {}, zero points over sea {}\".format(station.name, station.river_km, station.zero)\n",
    "    m += leaf.Marker(location=station.pos, draggable=False, title=title)\n",
    "m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Overview\n",
    "\n",
    "We use a neural network to get a 12h prediction. The notebook is organised as follows:\n",
    "\n",
    "At first we prepare our data in general by adding shifted values from the past and dropping all times where data from the station 'Bonn' are missing. Otherwise we would not have labels to train our neuronal network. (section 1)\n",
    "\n",
    "Then we look at two different things: Getting prediction from a single timestamp (section 2.1) or predicting succesivley for a time interval (section 2.2). In both scenarios we handling with missing data at our timestamp or at the time interval.\n",
    "\n",
    "If all data are prepared we train the neural networks and get a prediction.\n",
    "\n",
    "In the end we interpred our results and discuss advantages and disadvantages of our approach (section 3)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1) Shifting and dropping the times where data for 'Bonn' are missing\n",
    "\n",
    "We compute the time distance between each station and Bonn. To this end we use the mean flow speed of the river Rhine and the corresponding tributaries. This time is rounded to 15 minutes intervals, shifted by 12 hours and stored in 'df_shift'.\n",
    "\n",
    "The idea of 'df_12' is as follows: For each time index t we store the data of each station, which is not Bonn, at time t, the data of Bonn at time t+12h and the shifted data s = t+12h-rounded(time_distance) for each station where s<=t. The latter data are the data which correspond to the water level in Bonn at t+12h and are avaiable at time t."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def round_timedelta(td, period):\n",
    "    \"\"\"\n",
    "    Rounds the given timedelta by the given timedelta period\n",
    "    :param td: `timedelta` to round\n",
    "    :param period: `timedelta` period to round by.\n",
    "    \"\"\"\n",
    "    period_seconds = period.total_seconds()\n",
    "    half_period_seconds = period_seconds / 2\n",
    "    remainder = td.total_seconds() % period_seconds\n",
    "    if remainder >= half_period_seconds:\n",
    "        return pd.Timedelta(seconds=td.total_seconds() + (period_seconds - remainder))\n",
    "    else:\n",
    "        return pd.Timedelta(seconds=td.total_seconds() - remainder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['Andernach', 5.859999999999998],\n",
       " ['Bingen', 18.062857142857133],\n",
       " ['Bonn', 0.0],\n",
       " ['Frankfurt Osthafen', 116.57071428571429],\n",
       " ['Kalkofen Neu', 20.17142857142856],\n",
       " ['Kaub', 15.5],\n",
       " ['Koblenz', 9.044285714285706],\n",
       " ['Koblenz Up', 9.406071428571428],\n",
       " ['Mainz', 22.36142857142857],\n",
       " ['Oberwinter', 2.3728571428571286],\n",
       " ['Oestrich', 19.53142857142856],\n",
       " ['Raunheim', 53.120714285714286],\n",
       " ['Rockenau Ska', 119.08571428571429],\n",
       " ['Speyer', 36.3142857142857],\n",
       " ['Worms', 30.199999999999996]]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We do this by hand because of the tributary rivers and strange ways to count the river km (i.e. river Lahn)\n",
    "time_distance = [['Andernach', (654.8-levels.stations()[0].river_km)/7],\n",
    "['Bingen', (654.8-levels.stations()[1].river_km)/7],\n",
    "['Bonn', (654.8-levels.stations()[2].river_km)/7],\n",
    "['Frankfurt Osthafen', (654.8-496.63)/7+levels.stations()[3].river_km/0.4],\n",
    "['Kalkofen Neu', (654.8-585.7)/7+(137.3-levels.stations()[4].river_km)/3],\n",
    "['Kaub',(654.8-levels.stations()[5].river_km)/7],\n",
    "['Koblenz', (654.8-levels.stations()[6].river_km)/7],\n",
    "['Koblenz Up', (654.8-592.3)/7+levels.stations()[7].river_km/4],\n",
    "['Mainz', (654.8-levels.stations()[8].river_km)/7],\n",
    "['Oberwinter', (654.8-levels.stations()[9].river_km)/7],\n",
    "['Oestrich',(654.8-levels.stations()[10].river_km)/7],\n",
    "['Raunheim', (654.8-496.63)/7+levels.stations()[11].river_km/0.4],\n",
    "['Rockenau Ska', (654.8-428.2)/7+levels.stations()[12].river_km/0.7],\n",
    "['Speyer', (654.8-levels.stations()[13].river_km)/7],\n",
    "['Worms', (654.8-levels.stations()[14].river_km)/7]]\n",
    "time_distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_shift = pd.DataFrame()\n",
    "for i in range(15):\n",
    "    df_shift[time_distance[i][0]] = df[time_distance[i][0]].shift(freq = round_timedelta(pd.Timedelta(hours = time_distance[i][1]), pd.Timedelta(minutes=15))-pd.Timedelta(hours=12))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_12['Kaub shift'] = df_shift['Kaub']\n",
    "df_12['Kalkofen Neu shift'] = df_shift['Kalkofen Neu']\n",
    "df_12['Bingen shift'] = df_shift['Bingen']\n",
    "df_12['Oestrich shift'] = df_shift['Oestrich']\n",
    "df_12['Mainz shift'] = df_shift['Mainz']\n",
    "df_12['Raunheim shift'] = df_shift['Raunheim']\n",
    "df_12['Frankfurt Osthafen shift'] = df_shift['Frankfurt Osthafen']\n",
    "df_12['Rockenau Ska shift'] = df_shift['Rockenau Ska']\n",
    "df_12['Speyer shift'] = df_shift['Speyer']\n",
    "df_12['Worms shift'] = df_shift['Worms']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_12['Bonn'] = df['Bonn'].shift(freq = -pd.Timedelta(hours=12))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we drop all data in the beginning and the end where data for the station Bonn are missing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "first_valid_index = df_12['Bonn'].first_valid_index()\n",
    "last_valid_index = df_12['Bonn'].last_valid_index()\n",
    "df_12 = df_12[first_valid_index:]\n",
    "df_12 = df_12[:last_valid_index]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We fill gaps of length 24h and check afterwards if there are still missing data for Bonn (we need for each time index the data of Bonn the compare our prediction with the true water levels)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_12 = df_12.fillna(method='ffill', limit=4*24)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Andernach                        0\n",
       "Bingen                       58608\n",
       "Bonn                             0\n",
       "Frankfurt Osthafen          263184\n",
       "Kalkofen Neu                 59088\n",
       "Kaub                             0\n",
       "Koblenz                       2832\n",
       "Koblenz Up                   70704\n",
       "Mainz                         5914\n",
       "Oberwinter                  108240\n",
       "Oestrich                    201840\n",
       "Raunheim                    228048\n",
       "Rockenau Ska                 24432\n",
       "Speyer                      122928\n",
       "Worms                            0\n",
       "Kaub shift                       0\n",
       "Kalkofen Neu shift           59121\n",
       "Bingen shift                 58632\n",
       "Oestrich shift              201870\n",
       "Mainz shift                   5914\n",
       "Raunheim shift              228212\n",
       "Frankfurt Osthafen shift    263602\n",
       "Rockenau Ska shift           24860\n",
       "Speyer shift                123025\n",
       "Worms shift                      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_12.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the next cell we get the dates when the first values appear in each column. We do this to get an idea on how many data our neural network is trained."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Andernach                  1981-04-30 11:15:00+00:00\n",
       "Bingen                     1982-12-31 23:15:00+00:00\n",
       "Bonn                       1981-04-30 11:15:00+00:00\n",
       "Frankfurt Osthafen         1988-10-31 23:15:00+00:00\n",
       "Kalkofen Neu               1982-12-31 23:15:00+00:00\n",
       "Kaub                       1981-04-30 11:15:00+00:00\n",
       "Koblenz                    1981-04-30 23:15:00+00:00\n",
       "Koblenz Up                 1982-12-31 23:15:00+00:00\n",
       "Mainz                      1981-04-30 11:15:00+00:00\n",
       "Oberwinter                 1984-05-31 23:15:00+00:00\n",
       "Oestrich                   1987-01-31 23:15:00+00:00\n",
       "Raunheim                   1987-10-31 23:15:00+00:00\n",
       "Rockenau Ska               1981-12-31 23:15:00+00:00\n",
       "Speyer                     1984-10-31 23:15:00+00:00\n",
       "Worms                      1981-04-30 11:15:00+00:00\n",
       "Kaub shift                 1981-04-30 11:15:00+00:00\n",
       "Kalkofen Neu shift         1983-01-01 07:30:00+00:00\n",
       "Bingen shift               1983-01-01 05:15:00+00:00\n",
       "Oestrich shift             1987-02-01 06:45:00+00:00\n",
       "Mainz shift                1981-04-30 11:15:00+00:00\n",
       "Raunheim shift             1987-11-02 16:15:00+00:00\n",
       "Frankfurt Osthafen shift   1988-11-05 07:45:00+00:00\n",
       "Rockenau Ska shift         1982-01-05 10:15:00+00:00\n",
       "Speyer shift               1984-11-01 23:30:00+00:00\n",
       "Worms shift                1981-04-30 11:15:00+00:00\n",
       "dtype: datetime64[ns, UTC]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_12.apply(pd.Series.first_valid_index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the dataframe 'df_12a' we dropped the stations which are too near or too far away from Bonn to provide useful data, in 'df_12b' we used only the shifted data and in 'df_c' we used no shifted data at all. We will compare the results to 'df_12' and decide which approach is the best."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_12a = df_12.drop(columns=['Oberwinter', 'Andernach', 'Mainz', 'Raunheim', 'Frankfurt Osthafen', 'Speyer', 'Worms', 'Rockenau Ska'])\n",
    "df_12b = df_12.drop(columns=['Oberwinter', 'Andernach','Koblenz','Koblenz Up', 'Kalkofen Neu','Kaub','Bingen','Oestrich', 'Mainz', 'Raunheim', 'Frankfurt Osthafen', 'Speyer', 'Worms', 'Rockenau Ska'])\n",
    "df_12c = df_12.drop(columns=['Kaub shift', 'Kalkofen Neu shift', 'Bingen shift', 'Oestrich shift', 'Mainz shift', 'Raunheim shift', 'Frankfurt Osthafen shift', 'Rockenau Ska shift', 'Speyer shift', 'Worms shift'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2) Dealing with missing data\n",
    "\n",
    "### 2.1) Index time\n",
    "\n",
    "We choose an index time (here randomly) and check if there are missing data at this time. We drop the columns with the missing data for our training data. This is done to train the neural network with the same stations avaiable at time t."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1990-11-01 19:00:00\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2862: FutureWarning: Converting timezone-aware DatetimeArray to timezone-naive ndarray with 'datetime64[ns]' dtype. In the future, this will return an ndarray with 'object' dtype where each element is a 'pandas.Timestamp' with the correct 'tz'.\n",
      "\tTo accept the future behavior, pass 'dtype=object'.\n",
      "\tTo keep the old behavior, pass 'dtype=\"datetime64[ns]\"'.\n",
      "  exec(code_obj, self.user_global_ns, self.user_ns)\n"
     ]
    }
   ],
   "source": [
    "#gives a random time in the index of df_12\n",
    "index_random = pd.to_datetime(np.random.choice(df_12.index, replace=False))\n",
    "print(index_random)\n",
    "#time until we are allowed to train the model (otherwise we would train the model on future data)\n",
    "index_training = pd.to_datetime(index_random-pd.Timedelta(hours=12))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#exist at this time columns with missing data?\n",
    "nan_bool = df_12.iloc[df_12.index.get_loc(index_random)].isna().any()\n",
    "if nan_bool == True:\n",
    "    #columns with the missing data\n",
    "    nan_col = df_12.columns[df_12.iloc[df_12.index.get_loc(index_random)].isna()]\n",
    "    #drop these columns\n",
    "    df_tt = df_12.drop(columns=nan_col)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_test = df_tt.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_index = df_tt.drop(columns='Bonn').iloc[df_tt.index.get_loc(index_random)]\n",
    "    print('NaN columns')\n",
    "else: \n",
    "    df_test = df_12.dropna()\n",
    "    data_at_index = df_12.drop(columns='Bonn').iloc[df_12.index.get_loc(index_random)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#exist at this time columns with missing data?\n",
    "nan_boola = df_12a.iloc[df_12a.index.get_loc(index_random)].isna().any()\n",
    "if nan_boola == True:\n",
    "    #columns with the missing data\n",
    "    nan_cola = df_12a.columns[df_12a.iloc[df_12a.index.get_loc(index_random)].isna()]\n",
    "    #drop these columns\n",
    "    df_tta = df_12a.drop(columns=nan_cola)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testa = df_tta.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_indexa = df_tta.drop(columns='Bonn').iloc[df_tta.index.get_loc(index_random)]\n",
    "    print('NaN columns')\n",
    "else: \n",
    "    df_testa = df_12a.dropna()\n",
    "    data_at_indexa = df_12a.drop(columns='Bonn').iloc[df_12a.index.get_loc(index_random)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#exist at this time columns with missing data?\n",
    "nan_boolb = df_12b.iloc[df_12b.index.get_loc(index_random)].isna().any()\n",
    "if nan_boolb == True:\n",
    "    #columns with the missing data\n",
    "    nan_colb = df_12b.columns[df_12b.iloc[df_12b.index.get_loc(index_random)].isna()]\n",
    "    #drop these columns\n",
    "    df_ttb = df_12b.drop(columns=nan_colb)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testb = df_ttb.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_indexb = df_ttb.drop(columns='Bonn').iloc[df_ttb.index.get_loc(index_random)]\n",
    "    print('NaN columns')\n",
    "else: \n",
    "    df_testb = df_12b.dropna()\n",
    "    data_at_indexb = df_12b.drop(columns='Bonn').iloc[df_12b.index.get_loc(index_random)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#exist at this time columns with missing data?\n",
    "nan_boolc = df_12c.iloc[df_12c.index.get_loc(index_random)].isna().any()\n",
    "if nan_boolc == True:\n",
    "    #columns with the missing data\n",
    "    nan_colc = df_12c.columns[df_12c.iloc[df_12c.index.get_loc(index_random)].isna()]\n",
    "    #drop these columns\n",
    "    df_ttc = df_12c.drop(columns=nan_colc)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testc = df_ttc.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_indexc = df_ttc.drop(columns='Bonn').iloc[df_ttc.index.get_loc(index_random)]\n",
    "    print('NaN columns')\n",
    "else: \n",
    "    df_testc = df_12c.dropna()\n",
    "    data_at_indexc = df_12c.drop(columns='Bonn').iloc[df_12c.index.get_loc(index_random)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.1.1) Neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Flatten\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We prepare the test and training data using our preprocessed data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:1: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:3: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:12: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  if sys.path[0] == '':\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:17: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:19: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:21: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:26: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:28: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:30: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:35: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n"
     ]
    }
   ],
   "source": [
    "sc_Label = df_test['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_pred = df_test.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_train, X_test, y_train, y_test = train_test_split(df_pred, sc_Label)\n",
    "X_train_rs = X_train.reshape((X_train.shape[0], 1, X_train.shape[1]))\n",
    "X_test_rs = X_test.reshape((X_test.shape[0], 1, X_test.shape[1]))\n",
    "df_pred = df_pred.reshape((df_pred.shape[0], 1, df_pred.shape[1]))\n",
    "data_at_index = data_at_index.as_matrix().reshape(1,1, data_at_index.shape[0])\n",
    "\n",
    "sc_Labela = df_testa['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_preda = df_testa.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_traina, X_testa, y_traina, y_testa = train_test_split(df_preda, sc_Labela)\n",
    "X_train_rsa = X_traina.reshape((X_traina.shape[0], 1, X_traina.shape[1]))\n",
    "X_test_rsa = X_testa.reshape((X_testa.shape[0], 1, X_testa.shape[1]))\n",
    "df_preda = df_preda.reshape((df_preda.shape[0], 1, df_preda.shape[1]))\n",
    "data_at_indexa = data_at_indexa.as_matrix().reshape(1,1, data_at_indexa.shape[0])\n",
    "\n",
    "sc_Labelb = df_testb['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_predb = df_testb.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_trainb, X_testb, y_trainb, y_testb = train_test_split(df_predb, sc_Labelb)\n",
    "X_train_rsb = X_trainb.reshape((X_trainb.shape[0], 1, X_trainb.shape[1]))\n",
    "X_test_rsb = X_testb.reshape((X_testb.shape[0], 1, X_testb.shape[1]))\n",
    "df_predb = df_predb.reshape((df_predb.shape[0], 1, df_predb.shape[1]))\n",
    "data_at_indexb = data_at_indexb.as_matrix().reshape(1,1, data_at_indexb.shape[0])\n",
    "\n",
    "sc_Labelc = df_testc['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_predc = df_testc.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_trainc, X_testc, y_trainc, y_testc = train_test_split(df_predc, sc_Labelc)\n",
    "X_train_rsc = X_trainc.reshape((X_trainc.shape[0], 1, X_trainc.shape[1]))\n",
    "X_test_rsc = X_testc.reshape((X_testc.shape[0], 1, X_testc.shape[1]))\n",
    "df_predc = df_predc.reshape((df_predc.shape[0], 1, df_predc.shape[1]))\n",
    "data_at_indexc = data_at_indexc.as_matrix().reshape(1,1, data_at_indexc.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We chose a very simple neural network (linear regression) because it seemed quite reasonable in the case of high water prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    }
   ],
   "source": [
    "#build neural network\n",
    "model = Sequential()\n",
    "model.add(Dense(units=28, activation='relu', input_shape=(X_train_rs.shape[1], X_train_rs.shape[2])))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "modela = Sequential()\n",
    "modela.add(Dense(units=28, activation='relu', input_shape=(X_train_rsa.shape[1], X_train_rsa.shape[2])))\n",
    "modela.add(Flatten())\n",
    "modela.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "modelb = Sequential()\n",
    "modelb.add(Dense(units=28, activation='relu', input_shape=(X_train_rsb.shape[1], X_train_rsb.shape[2])))\n",
    "modelb.add(Flatten())\n",
    "modelb.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "modelc = Sequential()\n",
    "modelc.add(Dense(units=28, activation='relu', input_shape=(X_train_rsc.shape[1], X_train_rsc.shape[2])))\n",
    "modelc.add(Flatten())\n",
    "modelc.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "model.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "modela.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "modelb.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "modelc.compile(optimizer='Nadam', loss='mse', metrics=['mse'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Train on 52270 samples, validate on 17424 samples\n",
      "Epoch 1/15\n",
      "52270/52270 [==============================] - 4s 83us/step - loss: 89.4119 - mean_squared_error: 89.4119 - val_loss: 37.2347 - val_mean_squared_error: 37.2347\n",
      "Epoch 2/15\n",
      "52270/52270 [==============================] - 4s 75us/step - loss: 43.9562 - mean_squared_error: 43.9562 - val_loss: 37.3822 - val_mean_squared_error: 37.3822\n",
      "Epoch 3/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 35.2152 - mean_squared_error: 35.2152 - val_loss: 44.4825 - val_mean_squared_error: 44.4825\n",
      "Epoch 4/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 29.9298 - mean_squared_error: 29.9298 - val_loss: 26.4250 - val_mean_squared_error: 26.4250\n",
      "Epoch 5/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 28.6022 - mean_squared_error: 28.6022 - val_loss: 24.8810 - val_mean_squared_error: 24.8810\n",
      "Epoch 6/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 26.7358 - mean_squared_error: 26.7358 - val_loss: 31.7776 - val_mean_squared_error: 31.7776\n",
      "Epoch 7/15\n",
      "52270/52270 [==============================] - 4s 73us/step - loss: 26.0367 - mean_squared_error: 26.0367 - val_loss: 30.4591 - val_mean_squared_error: 30.4591\n",
      "Epoch 8/15\n",
      "52270/52270 [==============================] - 4s 77us/step - loss: 24.6864 - mean_squared_error: 24.6864 - val_loss: 21.0774 - val_mean_squared_error: 21.0774\n",
      "Epoch 9/15\n",
      "52270/52270 [==============================] - 4s 81us/step - loss: 23.6790 - mean_squared_error: 23.6790 - val_loss: 21.9277 - val_mean_squared_error: 21.9277\n",
      "Epoch 10/15\n",
      "52270/52270 [==============================] - 4s 84us/step - loss: 22.6281 - mean_squared_error: 22.6281 - val_loss: 48.6931 - val_mean_squared_error: 48.6931\n",
      "Epoch 11/15\n",
      "52270/52270 [==============================] - 4s 81us/step - loss: 21.9886 - mean_squared_error: 21.9886 - val_loss: 29.5959 - val_mean_squared_error: 29.5959\n",
      "Epoch 12/15\n",
      "52270/52270 [==============================] - 4s 73us/step - loss: 21.6359 - mean_squared_error: 21.6359 - val_loss: 25.9901 - val_mean_squared_error: 25.9901\n",
      "Epoch 13/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 21.3130 - mean_squared_error: 21.3130 - val_loss: 20.8832 - val_mean_squared_error: 20.8832\n",
      "Epoch 14/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 20.9960 - mean_squared_error: 20.9960 - val_loss: 21.5841 - val_mean_squared_error: 21.5841\n",
      "Epoch 15/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 20.6767 - mean_squared_error: 20.6767 - val_loss: 18.4834 - val_mean_squared_error: 18.4834\n"
     ]
    }
   ],
   "source": [
    "history_index = model.fit(X_train_rs, y_train, epochs=15, batch_size=15, validation_data=(X_test_rs, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 52270 samples, validate on 17424 samples\n",
      "Epoch 1/15\n",
      "52270/52270 [==============================] - 4s 77us/step - loss: 385.5698 - mean_squared_error: 385.5698 - val_loss: 56.9802 - val_mean_squared_error: 56.9802\n",
      "Epoch 2/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 58.9054 - mean_squared_error: 58.9054 - val_loss: 36.8480 - val_mean_squared_error: 36.8480\n",
      "Epoch 3/15\n",
      "52270/52270 [==============================] - 4s 77us/step - loss: 46.3177 - mean_squared_error: 46.3177 - val_loss: 38.9378 - val_mean_squared_error: 38.9378\n",
      "Epoch 4/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 39.9444 - mean_squared_error: 39.9444 - val_loss: 43.0281 - val_mean_squared_error: 43.0281\n",
      "Epoch 5/15\n",
      "52270/52270 [==============================] - 4s 77us/step - loss: 37.2752 - mean_squared_error: 37.2752 - val_loss: 28.1650 - val_mean_squared_error: 28.1650\n",
      "Epoch 6/15\n",
      "52270/52270 [==============================] - 4s 80us/step - loss: 34.8010 - mean_squared_error: 34.8010 - val_loss: 64.0022 - val_mean_squared_error: 64.0022\n",
      "Epoch 7/15\n",
      "52270/52270 [==============================] - 4s 75us/step - loss: 32.8977 - mean_squared_error: 32.8977 - val_loss: 29.2914 - val_mean_squared_error: 29.2914\n",
      "Epoch 8/15\n",
      "52270/52270 [==============================] - 4s 76us/step - loss: 31.9047 - mean_squared_error: 31.9047 - val_loss: 28.2222 - val_mean_squared_error: 28.2222\n",
      "Epoch 9/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 31.1354 - mean_squared_error: 31.1354 - val_loss: 30.2915 - val_mean_squared_error: 30.2915\n",
      "Epoch 10/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 30.3424 - mean_squared_error: 30.3424 - val_loss: 26.5515 - val_mean_squared_error: 26.5515\n",
      "Epoch 11/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 29.7302 - mean_squared_error: 29.7302 - val_loss: 26.3825 - val_mean_squared_error: 26.3825\n",
      "Epoch 12/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 28.8459 - mean_squared_error: 28.8459 - val_loss: 31.3693 - val_mean_squared_error: 31.3693\n",
      "Epoch 13/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 28.6455 - mean_squared_error: 28.6455 - val_loss: 24.8838 - val_mean_squared_error: 24.8838\n",
      "Epoch 14/15\n",
      "52270/52270 [==============================] - 4s 71us/step - loss: 27.9708 - mean_squared_error: 27.9708 - val_loss: 26.6426 - val_mean_squared_error: 26.6426\n",
      "Epoch 15/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 27.4812 - mean_squared_error: 27.4812 - val_loss: 24.4904 - val_mean_squared_error: 24.4904\n"
     ]
    }
   ],
   "source": [
    "history_indexa = modela.fit(X_train_rsa, y_traina, epochs=15, batch_size=15, validation_data=(X_test_rsa, y_testa))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 52270 samples, validate on 17424 samples\n",
      "Epoch 1/15\n",
      "52270/52270 [==============================] - 4s 78us/step - loss: 1125.1165 - mean_squared_error: 1125.1165 - val_loss: 311.1348 - val_mean_squared_error: 311.1348\n",
      "Epoch 2/15\n",
      "52270/52270 [==============================] - 4s 72us/step - loss: 335.9710 - mean_squared_error: 335.9710 - val_loss: 434.6228 - val_mean_squared_error: 434.6228\n",
      "Epoch 3/15\n",
      "52270/52270 [==============================] - 4s 82us/step - loss: 286.0188 - mean_squared_error: 286.0188 - val_loss: 222.0398 - val_mean_squared_error: 222.0398\n",
      "Epoch 4/15\n",
      "52270/52270 [==============================] - 4s 82us/step - loss: 239.6312 - mean_squared_error: 239.6312 - val_loss: 205.2672 - val_mean_squared_error: 205.2672\n",
      "Epoch 5/15\n",
      "52270/52270 [==============================] - 4s 83us/step - loss: 227.3799 - mean_squared_error: 227.3799 - val_loss: 257.3425 - val_mean_squared_error: 257.3425\n",
      "Epoch 6/15\n",
      "52270/52270 [==============================] - 4s 81us/step - loss: 216.1712 - mean_squared_error: 216.1712 - val_loss: 192.8787 - val_mean_squared_error: 192.8787\n",
      "Epoch 7/15\n",
      "52270/52270 [==============================] - 4s 79us/step - loss: 208.8647 - mean_squared_error: 208.8647 - val_loss: 185.1783 - val_mean_squared_error: 185.1783\n",
      "Epoch 8/15\n",
      "52270/52270 [==============================] - 4s 83us/step - loss: 204.2803 - mean_squared_error: 204.2803 - val_loss: 193.9376 - val_mean_squared_error: 193.9376\n",
      "Epoch 9/15\n",
      "52270/52270 [==============================] - 4s 75us/step - loss: 201.5942 - mean_squared_error: 201.5942 - val_loss: 202.5529 - val_mean_squared_error: 202.5529\n",
      "Epoch 10/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 198.9957 - mean_squared_error: 198.9957 - val_loss: 191.8814 - val_mean_squared_error: 191.8814\n",
      "Epoch 11/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 195.9183 - mean_squared_error: 195.9183 - val_loss: 191.8872 - val_mean_squared_error: 191.8872\n",
      "Epoch 12/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 187.8378 - mean_squared_error: 187.8378 - val_loss: 197.6393 - val_mean_squared_error: 197.6393\n",
      "Epoch 13/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 185.0694 - mean_squared_error: 185.0694 - val_loss: 166.6548 - val_mean_squared_error: 166.6548\n",
      "Epoch 14/15\n",
      "52270/52270 [==============================] - 4s 74us/step - loss: 184.3189 - mean_squared_error: 184.3189 - val_loss: 188.5201 - val_mean_squared_error: 188.5201\n",
      "Epoch 15/15\n",
      "52270/52270 [==============================] - 4s 73us/step - loss: 182.2890 - mean_squared_error: 182.2890 - val_loss: 165.7454 - val_mean_squared_error: 165.7454\n"
     ]
    }
   ],
   "source": [
    "history_indexb = modelb.fit(X_train_rsb, y_trainb, epochs=15, batch_size=15, validation_data=(X_test_rsb, y_testb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 52584 samples, validate on 17528 samples\n",
      "Epoch 1/15\n",
      "52584/52584 [==============================] - 4s 82us/step - loss: 521.6134 - mean_squared_error: 521.6134 - val_loss: 34.3654 - val_mean_squared_error: 34.3654\n",
      "Epoch 2/15\n",
      "52584/52584 [==============================] - 4s 80us/step - loss: 47.9828 - mean_squared_error: 47.9828 - val_loss: 100.7452 - val_mean_squared_error: 100.7452\n",
      "Epoch 3/15\n",
      "52584/52584 [==============================] - 4s 76us/step - loss: 41.3564 - mean_squared_error: 41.3564 - val_loss: 31.2887 - val_mean_squared_error: 31.2887\n",
      "Epoch 4/15\n",
      "52584/52584 [==============================] - 4s 72us/step - loss: 37.4596 - mean_squared_error: 37.4596 - val_loss: 76.0630 - val_mean_squared_error: 76.0630\n",
      "Epoch 5/15\n",
      "52584/52584 [==============================] - 4s 73us/step - loss: 34.7775 - mean_squared_error: 34.7775 - val_loss: 30.8677 - val_mean_squared_error: 30.8677\n",
      "Epoch 6/15\n",
      "52584/52584 [==============================] - 4s 77us/step - loss: 33.4990 - mean_squared_error: 33.4990 - val_loss: 41.4452 - val_mean_squared_error: 41.4452\n",
      "Epoch 7/15\n",
      "52584/52584 [==============================] - 4s 78us/step - loss: 30.5525 - mean_squared_error: 30.5525 - val_loss: 24.8309 - val_mean_squared_error: 24.8309\n",
      "Epoch 8/15\n",
      "52584/52584 [==============================] - 4s 77us/step - loss: 27.6282 - mean_squared_error: 27.6282 - val_loss: 32.7727 - val_mean_squared_error: 32.7727\n",
      "Epoch 9/15\n",
      "52584/52584 [==============================] - 4s 77us/step - loss: 26.3901 - mean_squared_error: 26.3901 - val_loss: 22.5594 - val_mean_squared_error: 22.5594\n",
      "Epoch 10/15\n",
      "52584/52584 [==============================] - 4s 75us/step - loss: 25.9981 - mean_squared_error: 25.9981 - val_loss: 20.6148 - val_mean_squared_error: 20.6148\n",
      "Epoch 11/15\n",
      "52584/52584 [==============================] - 4s 73us/step - loss: 24.8991 - mean_squared_error: 24.8991 - val_loss: 30.9358 - val_mean_squared_error: 30.9358\n",
      "Epoch 12/15\n",
      "52584/52584 [==============================] - 4s 73us/step - loss: 24.4324 - mean_squared_error: 24.4324 - val_loss: 21.1702 - val_mean_squared_error: 21.1702\n",
      "Epoch 13/15\n",
      "52584/52584 [==============================] - 4s 76us/step - loss: 23.5188 - mean_squared_error: 23.5188 - val_loss: 21.3617 - val_mean_squared_error: 21.3617\n",
      "Epoch 14/15\n",
      "52584/52584 [==============================] - 4s 79us/step - loss: 22.9069 - mean_squared_error: 22.9069 - val_loss: 49.0700 - val_mean_squared_error: 49.0700\n",
      "Epoch 15/15\n",
      "52584/52584 [==============================] - 4s 78us/step - loss: 22.8639 - mean_squared_error: 22.8639 - val_loss: 48.0084 - val_mean_squared_error: 48.0084\n"
     ]
    }
   ],
   "source": [
    "history_indexc = modelc.fit(X_train_rsc, y_trainc, epochs=15, batch_size=15, validation_data=(X_test_rsc, y_testc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#get prediction\n",
    "y_pred_index = model.predict(data_at_index)\n",
    "y_pred_indexa = modela.predict(data_at_indexa)\n",
    "y_pred_indexb = modelb.predict(data_at_indexb)\n",
    "y_pred_indexc = modelc.predict(data_at_indexc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date: 1990-11-01 19:00:00\n",
      "True value: 301.0\n",
      "Prediction: 314.1676330566406\n",
      "Prediction a: 309.4034729003906\n",
      "Prediction b: 298.2431945800781\n",
      "Prediction c: 320.61383056640625\n",
      "Number of missing stations/ shifts: 0\n",
      "Number of used stations/ shifts: 25\n",
      "Number of missing stations/ shifts a: 0\n",
      "Number of used stations/ shifts a: 17\n",
      "Number of missing stations/ shifts b: 0\n",
      "Number of used stations/ shifts b: 11\n",
      "Number of missing stations/ shifts c: 0\n",
      "Number of used stations/ shifts c: 15\n"
     ]
    }
   ],
   "source": [
    "print('Date: {}'.format(index_random))\n",
    "print('True value: {}'.format(df_test.iloc[df_test.index.get_loc(index_random)]['Bonn']))\n",
    "print('Prediction: {}'.format(y_pred_index[0][0]))\n",
    "print('Prediction a: {}'.format(y_pred_indexa[0][0]))\n",
    "print('Prediction b: {}'.format(y_pred_indexb[0][0]))\n",
    "print('Prediction c: {}'.format(y_pred_indexc[0][0]))\n",
    "print('Number of missing stations/ shifts: {}'.format(df_12.iloc[df_12.index.get_loc(index_random)].isna().sum()))\n",
    "print('Number of used stations/ shifts: {}'.format(len(df_test.columns)))\n",
    "print('Number of missing stations/ shifts a: {}'.format(df_12a.iloc[df_12a.index.get_loc(index_random)].isna().sum()))\n",
    "print('Number of used stations/ shifts a: {}'.format(len(df_testa.columns)))\n",
    "print('Number of missing stations/ shifts b: {}'.format(df_12b.iloc[df_12b.index.get_loc(index_random)].isna().sum()))\n",
    "print('Number of used stations/ shifts b: {}'.format(len(df_testb.columns)))\n",
    "print('Number of missing stations/ shifts c: {}'.format(df_12c.iloc[df_12c.index.get_loc(index_random)].isna().sum()))\n",
    "print('Number of used stations/ shifts c: {}'.format(len(df_testc.columns)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2) Index interval\n",
    "\n",
    "Here we choose a random time interval for prediction. The prediction uses the actual data, but the neural network is trained only on the times before the interval starts. The code is nearly the same (and uses the same ideas) as in in section 2.1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2001-07-29 16:30:00 2012-08-01 16:30:00\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2862: FutureWarning: Converting timezone-aware DatetimeArray to timezone-naive ndarray with 'datetime64[ns]' dtype. In the future, this will return an ndarray with 'object' dtype where each element is a 'pandas.Timestamp' with the correct 'tz'.\n",
      "\tTo accept the future behavior, pass 'dtype=object'.\n",
      "\tTo keep the old behavior, pass 'dtype=\"datetime64[ns]\"'.\n",
      "  exec(code_obj, self.user_global_ns, self.user_ns)\n"
     ]
    }
   ],
   "source": [
    "index_random = pd.to_datetime(np.random.choice(df_12.index, replace=False))\n",
    "index_random2 = pd.to_datetime(np.random.choice(df_12.index[df_12.index.get_loc(index_random):], replace=False))\n",
    "index_training = pd.to_datetime(index_random-pd.Timedelta(hours=12))\n",
    "print(index_random, index_random2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_sub = df_12.iloc[np.arange(df_12.index.get_loc(index_random),df_12.index.get_loc(index_random2)+1)]\n",
    "nan = df_sub.isna().any().sum()\n",
    "if nan > 0:\n",
    "    #columns with the missing data\n",
    "    nan_col = df_sub.columns[df_sub.isna().any()]\n",
    "    #drop these columns\n",
    "    df_tt = df_12.drop(columns=nan_col)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_test = df_tt.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_interval = df_tt.drop(columns='Bonn').iloc[np.arange(df_tt.index.get_loc(index_random),\n",
    "                                                                 df_tt.index.get_loc(index_random2)+1)]\n",
    "    print('NaN columns')\n",
    "else:\n",
    "    df_test = df_12.dropna()\n",
    "    data_at_interval = df_12.drop(columns='Bonn').iloc[np.arange(df_12.index.get_loc(index_random),\n",
    "                                                                 df_12.index.get_loc(index_random2)+1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_suba = df_12a.iloc[np.arange(df_12a.index.get_loc(index_random),df_12a.index.get_loc(index_random2)+1)]\n",
    "nana = df_suba.isna().any().sum()\n",
    "if nana > 0:\n",
    "    #columns with the missing data\n",
    "    nan_cola = df_suba.columns[df_suba.isna().any()]\n",
    "    #drop these columns\n",
    "    df_tta = df_12a.drop(columns=nan_cola)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testa = df_tta.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_intervala = df_tta.drop(columns='Bonn').iloc[np.arange(df_tta.index.get_loc(index_random),\n",
    "                                                                 df_tta.index.get_loc(index_random2)+1)]\n",
    "    print('NaN columns')\n",
    "else:\n",
    "    df_testa = df_12a.dropna()\n",
    "    data_at_intervala = df_12a.drop(columns='Bonn').iloc[np.arange(df_12a.index.get_loc(index_random),\n",
    "                                                                 df_12a.index.get_loc(index_random2)+1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_subb = df_12b.iloc[np.arange(df_12b.index.get_loc(index_random),df_12b.index.get_loc(index_random2)+1)]\n",
    "nanb = df_subb.isna().any().sum()\n",
    "if nanb > 0:\n",
    "    #columns with the missing data\n",
    "    nan_colb = df_subb.columns[df_subb.isna().any()]\n",
    "    #drop these columns\n",
    "    df_ttb = df_12b.drop(columns=nan_colb)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testb = df_ttb.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_intervalb = df_ttb.drop(columns='Bonn').iloc[np.arange(df_ttb.index.get_loc(index_random),\n",
    "                                                                 df_ttb.index.get_loc(index_random2)+1)]\n",
    "    print('NaN columns')\n",
    "else:\n",
    "    df_testb = df_12b.dropna()\n",
    "    data_at_intervalb = df_12b.drop(columns='Bonn').iloc[np.arange(df_12b.index.get_loc(index_random),\n",
    "                                                                 df_12b.index.get_loc(index_random2)+1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_subc = df_12c.iloc[np.arange(df_12c.index.get_loc(index_random),df_12c.index.get_loc(index_random2)+1)]\n",
    "nanc = df_subc.isna().any().sum()\n",
    "if nanc > 0:\n",
    "    #columns with the missing data\n",
    "    nan_colc = df_subc.columns[df_subc.isna().any()]\n",
    "    #drop these columns\n",
    "    df_ttc = df_12c.drop(columns=nan_colc)\n",
    "    #drop all times before the random time which contain missing data\n",
    "    df_testc = df_ttc.dropna()\n",
    "    #series without all NaN data at the given time\n",
    "    data_at_intervalc = df_ttc.drop(columns='Bonn').iloc[np.arange(df_ttc.index.get_loc(index_random),\n",
    "                                                                 df_ttc.index.get_loc(index_random2)+1)]\n",
    "    print('NaN columns')\n",
    "else:\n",
    "    df_testc = df_12c.dropna()\n",
    "    data_at_intervalc = df_12c.drop(columns='Bonn').iloc[np.arange(df_12c.index.get_loc(index_random),\n",
    "                                                                 df_12c.index.get_loc(index_random2)+1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.2.1) Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:1: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:3: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:12: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  if sys.path[0] == '':\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:17: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:19: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:21: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:26: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:28: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:30: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:35: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n"
     ]
    }
   ],
   "source": [
    "sc_Label2 = df_test['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_pred2 = df_test.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_train2, X_test2, y_train2, y_test2 = train_test_split(df_pred2, sc_Label2)\n",
    "X_train_rs2 = X_train2.reshape((X_train2.shape[0], 1, X_train2.shape[1]))\n",
    "X_test_rs2 = X_test2.reshape((X_test2.shape[0], 1, X_test2.shape[1]))\n",
    "df_pred2 = df_pred2.reshape((df_pred2.shape[0], 1, df_pred2.shape[1]))\n",
    "data_at_interval = data_at_interval.as_matrix().reshape(data_at_interval.shape[0],1, data_at_interval.shape[1])\n",
    "\n",
    "sc_Label2a = df_testa['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_pred2a = df_testa.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_train2a, X_test2a, y_train2a, y_test2a = train_test_split(df_pred2a, sc_Label2a)\n",
    "X_train_rs2a = X_train2a.reshape((X_train2a.shape[0], 1, X_train2a.shape[1]))\n",
    "X_test_rs2a = X_test2a.reshape((X_test2a.shape[0], 1, X_test2a.shape[1]))\n",
    "df_pred2a = df_pred2a.reshape((df_pred2a.shape[0], 1, df_pred2a.shape[1]))\n",
    "data_at_intervala = data_at_intervala.as_matrix().reshape(data_at_intervala.shape[0],1, data_at_intervala.shape[1])\n",
    "\n",
    "sc_Label2b = df_testb['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_pred2b = df_testb.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_train2b, X_test2b, y_train2b, y_test2b = train_test_split(df_pred2b, sc_Label2b)\n",
    "X_train_rs2b = X_train2b.reshape((X_train2b.shape[0], 1, X_train2b.shape[1]))\n",
    "X_test_rs2b = X_test2b.reshape((X_test2b.shape[0], 1, X_test2b.shape[1]))\n",
    "df_pred2b = df_pred2b.reshape((df_pred2b.shape[0], 1, df_pred2b.shape[1]))\n",
    "data_at_intervalb = data_at_intervalb.as_matrix().reshape(data_at_intervalb.shape[0],1, data_at_intervalb.shape[1])\n",
    "\n",
    "sc_Label2c = df_testc['Bonn'][:index_training].as_matrix()\n",
    "\n",
    "df_pred2c = df_testc.drop(columns='Bonn')[:][:index_training].as_matrix()\n",
    "X_train2c, X_test2c, y_train2c, y_test2c = train_test_split(df_pred2c, sc_Label2c)\n",
    "X_train_rs2c = X_train2c.reshape((X_train2c.shape[0], 1, X_train2c.shape[1]))\n",
    "X_test_rs2c = X_test2c.reshape((X_test2c.shape[0], 1, X_test2c.shape[1]))\n",
    "df_pred2c = df_pred2c.reshape((df_pred2c.shape[0], 1, df_pred2c.shape[1]))\n",
    "data_at_intervalc = data_at_intervalc.as_matrix().reshape(data_at_intervalc.shape[0],1, data_at_intervalc.shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model2 = Sequential()\n",
    "model2.add(Dense(units=28, activation='relu', input_shape=(X_train_rs2.shape[1], X_train_rs2.shape[2])))\n",
    "model2.add(Flatten())\n",
    "model2.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "model2a = Sequential()\n",
    "model2a.add(Dense(units=28, activation='relu', input_shape=(X_train_rs2a.shape[1], X_train_rs2a.shape[2])))\n",
    "model2a.add(Flatten())\n",
    "model2a.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "model2b = Sequential()\n",
    "model2b.add(Dense(units=28, activation='relu', input_shape=(X_train_rs2b.shape[1], X_train_rs2b.shape[2])))\n",
    "model2b.add(Flatten())\n",
    "model2b.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "model2c = Sequential()\n",
    "model2c.add(Dense(units=28, activation='relu', input_shape=(X_train_rs2c.shape[1], X_train_rs2c.shape[2])))\n",
    "model2c.add(Flatten())\n",
    "model2c.add(Dense(units=1, activation='linear'))\n",
    "\n",
    "model2.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "model2a.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "model2b.compile(optimizer='Nadam', loss='mse', metrics=['mse'])\n",
    "model2c.compile(optimizer='Nadam', loss='mse', metrics=['mse'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 324966 samples, validate on 108322 samples\n",
      "Epoch 1/15\n",
      "324966/324966 [==============================] - 25s 76us/step - loss: 138.1114 - mean_squared_error: 138.1114 - val_loss: 32.6469 - val_mean_squared_error: 32.6469\n",
      "Epoch 2/15\n",
      "324966/324966 [==============================] - 24s 75us/step - loss: 30.3755 - mean_squared_error: 30.3755 - val_loss: 29.6900 - val_mean_squared_error: 29.6900\n",
      "Epoch 3/15\n",
      "324966/324966 [==============================] - 25s 76us/step - loss: 28.0541 - mean_squared_error: 28.0541 - val_loss: 25.3061 - val_mean_squared_error: 25.3061\n",
      "Epoch 4/15\n",
      "324966/324966 [==============================] - 24s 75us/step - loss: 27.0100 - mean_squared_error: 27.0100 - val_loss: 26.5050 - val_mean_squared_error: 26.5050\n",
      "Epoch 5/15\n",
      "324966/324966 [==============================] - 24s 75us/step - loss: 26.2240 - mean_squared_error: 26.2240 - val_loss: 35.4440 - val_mean_squared_error: 35.4440\n",
      "Epoch 6/15\n",
      "324966/324966 [==============================] - 24s 75us/step - loss: 25.5941 - mean_squared_error: 25.5941 - val_loss: 22.2930 - val_mean_squared_error: 22.2930\n",
      "Epoch 7/15\n",
      "324966/324966 [==============================] - 25s 78us/step - loss: 25.1984 - mean_squared_error: 25.1984 - val_loss: 50.7207 - val_mean_squared_error: 50.7207\n",
      "Epoch 8/15\n",
      "324966/324966 [==============================] - 25s 76us/step - loss: 24.9479 - mean_squared_error: 24.9479 - val_loss: 29.2279 - val_mean_squared_error: 29.2279\n",
      "Epoch 9/15\n",
      "324966/324966 [==============================] - 25s 78us/step - loss: 24.6591 - mean_squared_error: 24.6591 - val_loss: 45.0504 - val_mean_squared_error: 45.0504\n",
      "Epoch 10/15\n",
      "324966/324966 [==============================] - 25s 78us/step - loss: 24.4209 - mean_squared_error: 24.4209 - val_loss: 21.6027 - val_mean_squared_error: 21.6027\n",
      "Epoch 11/15\n",
      "324966/324966 [==============================] - 26s 79us/step - loss: 24.1611 - mean_squared_error: 24.1611 - val_loss: 22.1035 - val_mean_squared_error: 22.1035\n",
      "Epoch 12/15\n",
      "324966/324966 [==============================] - 26s 79us/step - loss: 23.8343 - mean_squared_error: 23.8343 - val_loss: 22.0482 - val_mean_squared_error: 22.0482\n",
      "Epoch 13/15\n",
      "324966/324966 [==============================] - 25s 78us/step - loss: 23.5903 - mean_squared_error: 23.5903 - val_loss: 44.2795 - val_mean_squared_error: 44.2795\n",
      "Epoch 14/15\n",
      "324966/324966 [==============================] - 25s 77us/step - loss: 23.3718 - mean_squared_error: 23.3718 - val_loss: 20.9021 - val_mean_squared_error: 20.9021\n",
      "Epoch 15/15\n",
      "324966/324966 [==============================] - 25s 78us/step - loss: 23.2722 - mean_squared_error: 23.2722 - val_loss: 23.0152 - val_mean_squared_error: 23.0152\n"
     ]
    }
   ],
   "source": [
    "history_interval = model2.fit(X_train_rs2, y_train2, epochs=15, batch_size=15, validation_data=(X_test_rs2, y_test2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 325287 samples, validate on 108429 samples\n",
      "Epoch 1/15\n",
      "325287/325287 [==============================] - 26s 78us/step - loss: 140.2673 - mean_squared_error: 140.2673 - val_loss: 30.2312 - val_mean_squared_error: 30.2312\n",
      "Epoch 2/15\n",
      "325287/325287 [==============================] - 25s 77us/step - loss: 33.4585 - mean_squared_error: 33.4585 - val_loss: 33.7881 - val_mean_squared_error: 33.7881\n",
      "Epoch 3/15\n",
      "325287/325287 [==============================] - 25s 77us/step - loss: 31.3960 - mean_squared_error: 31.3960 - val_loss: 30.5772 - val_mean_squared_error: 30.5772\n",
      "Epoch 4/15\n",
      "325287/325287 [==============================] - 26s 81us/step - loss: 30.2081 - mean_squared_error: 30.2081 - val_loss: 29.1489 - val_mean_squared_error: 29.1489\n",
      "Epoch 5/15\n",
      "325287/325287 [==============================] - 27s 82us/step - loss: 29.2683 - mean_squared_error: 29.2683 - val_loss: 26.5638 - val_mean_squared_error: 26.5638\n",
      "Epoch 6/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 28.5575 - mean_squared_error: 28.5575 - val_loss: 27.7057 - val_mean_squared_error: 27.7057\n",
      "Epoch 7/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 28.1925 - mean_squared_error: 28.1925 - val_loss: 28.7318 - val_mean_squared_error: 28.7318\n",
      "Epoch 8/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 27.8315 - mean_squared_error: 27.8315 - val_loss: 28.4734 - val_mean_squared_error: 28.4734\n",
      "Epoch 9/15\n",
      "325287/325287 [==============================] - 26s 80us/step - loss: 27.6223 - mean_squared_error: 27.6223 - val_loss: 28.6503 - val_mean_squared_error: 28.6503\n",
      "Epoch 10/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 27.3624 - mean_squared_error: 27.3624 - val_loss: 26.6216 - val_mean_squared_error: 26.6216\n",
      "Epoch 11/15\n",
      "325287/325287 [==============================] - 26s 80us/step - loss: 27.2393 - mean_squared_error: 27.2393 - val_loss: 25.2475 - val_mean_squared_error: 25.2475\n",
      "Epoch 12/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 27.0475 - mean_squared_error: 27.0475 - val_loss: 24.6912 - val_mean_squared_error: 24.6912\n",
      "Epoch 13/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 26.9596 - mean_squared_error: 26.9596 - val_loss: 25.7362 - val_mean_squared_error: 25.7362\n",
      "Epoch 14/15\n",
      "325287/325287 [==============================] - 26s 79us/step - loss: 26.8573 - mean_squared_error: 26.8573 - val_loss: 25.2886 - val_mean_squared_error: 25.2886\n",
      "Epoch 15/15\n",
      "325287/325287 [==============================] - 25s 77us/step - loss: 26.7911 - mean_squared_error: 26.7911 - val_loss: 24.8957 - val_mean_squared_error: 24.8957\n"
     ]
    }
   ],
   "source": [
    "history_intervala = model2a.fit(X_train_rs2a, y_train2a, epochs=15, batch_size=15, validation_data=(X_test_rs2a, y_test2a))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 334359 samples, validate on 111453 samples\n",
      "Epoch 1/15\n",
      "334359/334359 [==============================] - 26s 79us/step - loss: 403.4879 - mean_squared_error: 403.4879 - val_loss: 311.4679 - val_mean_squared_error: 311.4679\n",
      "Epoch 2/15\n",
      "334359/334359 [==============================] - 26s 77us/step - loss: 318.9699 - mean_squared_error: 318.9699 - val_loss: 386.7357 - val_mean_squared_error: 386.7357\n",
      "Epoch 3/15\n",
      "334359/334359 [==============================] - 27s 81us/step - loss: 309.1626 - mean_squared_error: 309.1626 - val_loss: 305.1217 - val_mean_squared_error: 305.1217\n",
      "Epoch 4/15\n",
      "334359/334359 [==============================] - 26s 79us/step - loss: 301.8756 - mean_squared_error: 301.8756 - val_loss: 298.6228 - val_mean_squared_error: 298.6228\n",
      "Epoch 5/15\n",
      "334359/334359 [==============================] - 26s 77us/step - loss: 295.5890 - mean_squared_error: 295.5890 - val_loss: 322.4487 - val_mean_squared_error: 322.4487\n",
      "Epoch 6/15\n",
      "334359/334359 [==============================] - 26s 77us/step - loss: 291.6822 - mean_squared_error: 291.6822 - val_loss: 276.8433 - val_mean_squared_error: 276.8433\n",
      "Epoch 7/15\n",
      "334359/334359 [==============================] - 26s 77us/step - loss: 288.0823 - mean_squared_error: 288.0823 - val_loss: 276.9344 - val_mean_squared_error: 276.9344\n",
      "Epoch 8/15\n",
      "334359/334359 [==============================] - 26s 78us/step - loss: 285.7671 - mean_squared_error: 285.7671 - val_loss: 275.1520 - val_mean_squared_error: 275.1520\n",
      "Epoch 9/15\n",
      "334359/334359 [==============================] - 27s 82us/step - loss: 283.9456 - mean_squared_error: 283.9456 - val_loss: 270.2828 - val_mean_squared_error: 270.2828\n",
      "Epoch 10/15\n",
      "334359/334359 [==============================] - 28s 83us/step - loss: 282.4106 - mean_squared_error: 282.4106 - val_loss: 297.2377 - val_mean_squared_error: 297.2377\n",
      "Epoch 11/15\n",
      "334359/334359 [==============================] - 26s 78us/step - loss: 280.6361 - mean_squared_error: 280.6361 - val_loss: 283.2807 - val_mean_squared_error: 283.2807\n",
      "Epoch 12/15\n",
      "334359/334359 [==============================] - 26s 79us/step - loss: 279.7924 - mean_squared_error: 279.7924 - val_loss: 267.4940 - val_mean_squared_error: 267.4940\n",
      "Epoch 13/15\n",
      "334359/334359 [==============================] - 27s 80us/step - loss: 278.4845 - mean_squared_error: 278.4845 - val_loss: 268.8325 - val_mean_squared_error: 268.8325\n",
      "Epoch 14/15\n",
      "334359/334359 [==============================] - 27s 81us/step - loss: 277.3335 - mean_squared_error: 277.3335 - val_loss: 265.9010 - val_mean_squared_error: 265.9010\n",
      "Epoch 15/15\n",
      "334359/334359 [==============================] - 26s 77us/step - loss: 275.9820 - mean_squared_error: 275.9820 - val_loss: 265.3528 - val_mean_squared_error: 265.3528\n"
     ]
    }
   ],
   "source": [
    "history_intervalb = model2b.fit(X_train_rs2b, y_train2b, epochs=15, batch_size=15, validation_data=(X_test_rs2b, y_test2b))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 325600 samples, validate on 108534 samples\n",
      "Epoch 1/15\n",
      "325600/325600 [==============================] - 26s 80us/step - loss: 108.0437 - mean_squared_error: 108.0437 - val_loss: 31.4778 - val_mean_squared_error: 31.4778\n",
      "Epoch 2/15\n",
      "325600/325600 [==============================] - 26s 79us/step - loss: 35.6667 - mean_squared_error: 35.6667 - val_loss: 46.2031 - val_mean_squared_error: 46.2031\n",
      "Epoch 3/15\n",
      "325600/325600 [==============================] - 25s 77us/step - loss: 32.4826 - mean_squared_error: 32.4826 - val_loss: 29.9804 - val_mean_squared_error: 29.9804\n",
      "Epoch 4/15\n",
      "325600/325600 [==============================] - 25s 77us/step - loss: 31.2408 - mean_squared_error: 31.2408 - val_loss: 27.4747 - val_mean_squared_error: 27.4747\n",
      "Epoch 5/15\n",
      "325600/325600 [==============================] - 25s 77us/step - loss: 30.6439 - mean_squared_error: 30.6439 - val_loss: 28.8556 - val_mean_squared_error: 28.8556\n",
      "Epoch 6/15\n",
      "325600/325600 [==============================] - 26s 80us/step - loss: 30.3594 - mean_squared_error: 30.3594 - val_loss: 27.4877 - val_mean_squared_error: 27.4877\n",
      "Epoch 7/15\n",
      "325600/325600 [==============================] - 27s 82us/step - loss: 30.0577 - mean_squared_error: 30.0577 - val_loss: 27.6359 - val_mean_squared_error: 27.6359\n",
      "Epoch 8/15\n",
      "325600/325600 [==============================] - 26s 80us/step - loss: 29.8191 - mean_squared_error: 29.8191 - val_loss: 29.2219 - val_mean_squared_error: 29.2219\n",
      "Epoch 9/15\n",
      "325600/325600 [==============================] - 26s 79us/step - loss: 29.6445 - mean_squared_error: 29.6445 - val_loss: 29.3320 - val_mean_squared_error: 29.3320\n",
      "Epoch 10/15\n",
      "325600/325600 [==============================] - 25s 78us/step - loss: 29.5249 - mean_squared_error: 29.5249 - val_loss: 29.1952 - val_mean_squared_error: 29.1952\n",
      "Epoch 11/15\n",
      "325600/325600 [==============================] - 26s 80us/step - loss: 29.4554 - mean_squared_error: 29.4554 - val_loss: 26.6542 - val_mean_squared_error: 26.6542\n",
      "Epoch 12/15\n",
      "325600/325600 [==============================] - 26s 79us/step - loss: 29.3065 - mean_squared_error: 29.3065 - val_loss: 29.1901 - val_mean_squared_error: 29.1901\n",
      "Epoch 13/15\n",
      "325600/325600 [==============================] - 26s 79us/step - loss: 29.1565 - mean_squared_error: 29.1565 - val_loss: 27.1787 - val_mean_squared_error: 27.1787\n",
      "Epoch 14/15\n",
      "325600/325600 [==============================] - 25s 77us/step - loss: 29.1476 - mean_squared_error: 29.1476 - val_loss: 26.4658 - val_mean_squared_error: 26.4658\n",
      "Epoch 15/15\n",
      "325600/325600 [==============================] - 25s 77us/step - loss: 29.0124 - mean_squared_error: 29.0124 - val_loss: 27.2593 - val_mean_squared_error: 27.2593\n"
     ]
    }
   ],
   "source": [
    "history_intervalc = model2c.fit(X_train_rs2c, y_train2c, epochs=15, batch_size=15, validation_data=(X_test_rs2c, y_test2c))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# get prediction\n",
    "y_pred_interval = model2.predict(data_at_interval)\n",
    "y_pred_intervala = model2a.predict(data_at_intervala)\n",
    "y_pred_intervalb = model2b.predict(data_at_intervalb)\n",
    "y_pred_intervalc = model2c.predict(data_at_intervalc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_store = pd.DataFrame(index = df_sub.index)\n",
    "df_store['Prediction'] = y_pred_interval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_storea = pd.DataFrame(index = df_suba.index)\n",
    "df_storea['Prediction a'] = y_pred_intervala"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_storeb = pd.DataFrame(index = df_subb.index)\n",
    "df_storeb['Prediction b'] = y_pred_intervalb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_storec = pd.DataFrame(index = df_subc.index)\n",
    "df_storec['Prediction c'] = y_pred_intervalc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We plot the results to evaluate our predictions. The first plot shows the whole interval, the second one a flood and the last one a 'normal' month."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n",
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n",
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n",
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n",
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x124908320>"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI4AAAFpCAYAAAD6EmMEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niHT9P30Grh+eoErPAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11c108400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_sub['Bonn'].plot(figsize=(20,6))\n",
    "df_store['Prediction'].plot(figsize=(20,6))\n",
    "df_storea['Prediction a'].plot(figsize=(20,6))\n",
    "df_storeb['Prediction b'].plot(figsize=(20,6))\n",
    "df_storec['Prediction c'].plot(figsize=(20,6))\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x14f31fe80>"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAGACAYAAAA3TAKEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIiIiPRi/x9dex/O7wSSkwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x14f2bbda0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_sub['Bonn']['2011-01-03':'2011-01-24'].plot(figsize=(20,6))\n",
    "df_store['Prediction']['2011-01-03':'2011-01-24'].plot(figsize=(20,6))\n",
    "df_storea['Prediction a']['2011-01-03':'2011-01-24'].plot(figsize=(20,6))\n",
    "df_storeb['Prediction b']['2011-01-03':'2011-01-24'].plot(figsize=(20,6))\n",
    "df_storec['Prediction c']['2011-01-03':'2011-01-24'].plot(figsize=(20,6))\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/pandas/core/arrays/datetimes.py:1172: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  \"will drop timezone information.\", UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x200cbc630>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAGACAYAAAA3TAKEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nP3ou2uyVHmvLAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x200cbcb38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_sub['Bonn']['2008-10-01':'2008-10-20'].plot(figsize=(20,6))\n",
    "df_store['Prediction']['2008-10-01':'2008-10-20'].plot(figsize=(20,6))\n",
    "df_storea['Prediction a']['2008-10-01':'2008-10-20'].plot(figsize=(20,6))\n",
    "#df_storeb['Prediction b']['2008-10-01':'2008-10-20'].plot(figsize=(20,6))\n",
    "df_storec['Prediction c']['2008-10-01':'2008-10-20'].plot(figsize=(20,6))\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## 3) Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "- Data: We see that we get the best results if we use the normal data and the shifted data together. Dropping the shifted data or the stations to near / to far away does not effect the prediction to much. Just the prediction using only the shifted data is not this good. Maybe this comes from the fact that we used just the mean flow velocity of the different rivers.\n",
    "\n",
    "- Model: Our model seems to be very good - usually we get a mean squared error of 30 cm!\n",
    "\n",
    "- Problems: \n",
    "\n",
    "1) Since we use only data from the past and the fact that there are many missing data in the 1980's, we get bad results for timestamps in the 1980's and early 1990's. The missing data force us to train our neural network only on the data of a few month instead of many years (if we decide to allow time travels the problem vanishes).\n",
    "\n",
    "2) Floods are extreme and rare events. It is reasonable that our predictions are not as good as in the 'normal' time periods. But the deviation are only around 50 cm and the more important thing: The predictions are higher than the true values. So the flood prevention will be better than needed. This is much better than flood prevention that is not as good as needed. Nevertheless: The prediction for the arrival time of the flood wave is too late which is still a big problem. Here we are confronted with the uncertainty principle of communications engineering (tradeoff between time and frequency)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}