7_lstm_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "eztdNRTDvk_n"
   },
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "UQcatIbyEorJ"
   },
   "outputs": [],
   "source": [
    "# import libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn\n",
    "import tensorflow as tf\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numpy: 1.19.5\n",
      "Pandas: 1.1.5\n",
      "Tensorflow: 2.4.1\n",
      "Science Kit Learn: 0.24.1\n"
     ]
    }
   ],
   "source": [
    "# print versions of modules\n",
    "print(\"Numpy: \" + np.__version__)\n",
    "print(\"Pandas: \" + pd.__version__)\n",
    "print(\"Tensorflow: \" + tf.__version__)\n",
    "print(\"Science Kit Learn: \" + sklearn.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "gMa_b4s9E_fb"
   },
   "outputs": [],
   "source": [
    "# import additional modules\n",
    "from statistics import mean\n",
    "from math import sqrt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.layers import concatenate\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, LSTM\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.regularizers import l2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Area of study"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Ansongo-Niamey basin is a catchment located in the middle region of the Niger basin between the cities of Ansongo (Mali) and Niamey (Niger). This basin was chosen as study area because it would be challenging to collect data in the upstream of the Niger basin: the upper Niger basin. Regarding that, Ansongo being the downstream of the upper delta was considered as the upstream of our study area and Niamey the outlet of the basin."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Ansongo-Niamey map](./images/map_study_area.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Ansongo-Niger basin runoff regime is affected by different types of reoccurring floods, which result from the geographic locations and characteristics of their main source areas. \n",
    "\n",
    "The first one is the **Guinean Flood**, which originates from the headwaters of the Niger in the Guinean highlands during the rainy season between July and November. The flood originating in the Guinean highlands experiences its peak usually around October. From here, the Niger flows into a vast wetland that covers an area and delays there for approximately three months. This upper flood arrives in the Ansongo around January, although rainfall in the Sahelian region falls at\n",
    "the same time as in the Guinean highlands. \n",
    "\n",
    "The second one is the annual peak during the rainy season (July to November) in the Ansongo-Niamey basin called the **“Red Flood”** or “Sahelian Flood”.\n",
    "\n",
    "Figure below illustrates well the basin’s runoff regime by distinguishing red flood from the Guinean flood."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Hydrographs of Ansongo, Kandadji and Niamey for June 1991 to Mai 1992](./images/q_day_1991_1993_sub.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SWug2WWiv1DF"
   },
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "wUfHY79iFQBB"
   },
   "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>Date</th>\n",
       "      <th>pr</th>\n",
       "      <th>tmax</th>\n",
       "      <th>tmin</th>\n",
       "      <th>Ansongo</th>\n",
       "      <th>Kandadji</th>\n",
       "      <th>Niamey</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1981-06-01</td>\n",
       "      <td>0.525346</td>\n",
       "      <td>38.824417</td>\n",
       "      <td>27.352827</td>\n",
       "      <td>14.0</td>\n",
       "      <td>20.400000</td>\n",
       "      <td>5.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1981-06-02</td>\n",
       "      <td>0.732719</td>\n",
       "      <td>39.789036</td>\n",
       "      <td>27.593908</td>\n",
       "      <td>13.0</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>4.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1981-06-03</td>\n",
       "      <td>0.341014</td>\n",
       "      <td>39.171715</td>\n",
       "      <td>26.769451</td>\n",
       "      <td>13.0</td>\n",
       "      <td>14.600000</td>\n",
       "      <td>4.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1981-06-04</td>\n",
       "      <td>1.986175</td>\n",
       "      <td>39.452515</td>\n",
       "      <td>28.201244</td>\n",
       "      <td>13.0</td>\n",
       "      <td>14.100000</td>\n",
       "      <td>3.68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1981-06-05</td>\n",
       "      <td>0.271889</td>\n",
       "      <td>39.426506</td>\n",
       "      <td>26.821142</td>\n",
       "      <td>13.0</td>\n",
       "      <td>12.900000</td>\n",
       "      <td>2.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10801</th>\n",
       "      <td>2010-12-27</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.876694</td>\n",
       "      <td>19.051850</td>\n",
       "      <td>1751.0</td>\n",
       "      <td>1841.915039</td>\n",
       "      <td>1669.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10802</th>\n",
       "      <td>2010-12-28</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.088223</td>\n",
       "      <td>18.405252</td>\n",
       "      <td>1753.0</td>\n",
       "      <td>1851.276978</td>\n",
       "      <td>1677.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10803</th>\n",
       "      <td>2010-12-29</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.650715</td>\n",
       "      <td>17.136805</td>\n",
       "      <td>1752.0</td>\n",
       "      <td>1851.276978</td>\n",
       "      <td>1684.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10804</th>\n",
       "      <td>2010-12-30</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.456808</td>\n",
       "      <td>16.068770</td>\n",
       "      <td>1749.0</td>\n",
       "      <td>1851.276978</td>\n",
       "      <td>1684.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10805</th>\n",
       "      <td>2010-12-31</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.198921</td>\n",
       "      <td>16.231855</td>\n",
       "      <td>1743.0</td>\n",
       "      <td>1860.637939</td>\n",
       "      <td>1684.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10806 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Date        pr       tmax       tmin  Ansongo     Kandadji  \\\n",
       "0      1981-06-01  0.525346  38.824417  27.352827     14.0    20.400000   \n",
       "1      1981-06-02  0.732719  39.789036  27.593908     13.0    17.000000   \n",
       "2      1981-06-03  0.341014  39.171715  26.769451     13.0    14.600000   \n",
       "3      1981-06-04  1.986175  39.452515  28.201244     13.0    14.100000   \n",
       "4      1981-06-05  0.271889  39.426506  26.821142     13.0    12.900000   \n",
       "...           ...       ...        ...        ...      ...          ...   \n",
       "10801  2010-12-27  0.000000  33.876694  19.051850   1751.0  1841.915039   \n",
       "10802  2010-12-28  0.000000  33.088223  18.405252   1753.0  1851.276978   \n",
       "10803  2010-12-29  0.000000  30.650715  17.136805   1752.0  1851.276978   \n",
       "10804  2010-12-30  0.000000  29.456808  16.068770   1749.0  1851.276978   \n",
       "10805  2010-12-31  0.000000  29.198921  16.231855   1743.0  1860.637939   \n",
       "\n",
       "        Niamey  \n",
       "0         5.03  \n",
       "1         4.61  \n",
       "2         4.14  \n",
       "3         3.68  \n",
       "4         2.99  \n",
       "...        ...  \n",
       "10801  1669.00  \n",
       "10802  1677.00  \n",
       "10803  1684.00  \n",
       "10804  1684.00  \n",
       "10805  1684.00  \n",
       "\n",
       "[10806 rows x 7 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load data\n",
    "df = pd.read_csv('./model_data/model_data_interpolated.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "mGCklzx5FQDz"
   },
   "outputs": [],
   "source": [
    "# drop time column\n",
    "date_time = pd.to_datetime(df.pop('Date'), format='%Y-%m-%d') #or infer_datetime_format=True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot:xlabel='Date'>, <AxesSubplot:xlabel='Date'>,\n",
       "       <AxesSubplot:xlabel='Date'>, <AxesSubplot:xlabel='Date'>,\n",
       "       <AxesSubplot:xlabel='Date'>, <AxesSubplot:xlabel='Date'>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# view dataset\n",
    "df.set_index(date_time)[['pr', 'tmax', 'tmin', 'Ansongo', 'Kandadji', 'Niamey']].plot(subplots=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pr</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>1.375334</td>\n",
       "      <td>2.311514</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.414747</td>\n",
       "      <td>1.824885</td>\n",
       "      <td>24.069124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tmax</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>35.124774</td>\n",
       "      <td>3.299601</td>\n",
       "      <td>2.389276e+01</td>\n",
       "      <td>32.604908</td>\n",
       "      <td>35.198283</td>\n",
       "      <td>37.532277</td>\n",
       "      <td>43.912750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tmin</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>22.947793</td>\n",
       "      <td>3.614374</td>\n",
       "      <td>1.133926e+01</td>\n",
       "      <td>20.857830</td>\n",
       "      <td>23.514691</td>\n",
       "      <td>25.549257</td>\n",
       "      <td>32.082210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ansongo</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>685.853843</td>\n",
       "      <td>599.710296</td>\n",
       "      <td>5.530694e-07</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>538.000000</td>\n",
       "      <td>1240.750000</td>\n",
       "      <td>1904.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Kandadji</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>695.545463</td>\n",
       "      <td>607.054939</td>\n",
       "      <td>1.120000e+00</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>542.094513</td>\n",
       "      <td>1260.000000</td>\n",
       "      <td>2090.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Niamey</th>\n",
       "      <td>10806.0</td>\n",
       "      <td>705.412916</td>\n",
       "      <td>565.640158</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>116.000000</td>\n",
       "      <td>688.500000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1993.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count        mean         std           min         25%  \\\n",
       "pr        10806.0    1.375334    2.311514  0.000000e+00    0.000000   \n",
       "tmax      10806.0   35.124774    3.299601  2.389276e+01   32.604908   \n",
       "tmin      10806.0   22.947793    3.614374  1.133926e+01   20.857830   \n",
       "Ansongo   10806.0  685.853843  599.710296  5.530694e-07   99.000000   \n",
       "Kandadji  10806.0  695.545463  607.054939  1.120000e+00  104.000000   \n",
       "Niamey    10806.0  705.412916  565.640158  0.000000e+00  116.000000   \n",
       "\n",
       "                 50%          75%          max  \n",
       "pr          0.414747     1.824885    24.069124  \n",
       "tmax       35.198283    37.532277    43.912750  \n",
       "tmin       23.514691    25.549257    32.082210  \n",
       "Ansongo   538.000000  1240.750000  1904.000000  \n",
       "Kandadji  542.094513  1260.000000  2090.000000  \n",
       "Niamey    688.500000  1209.000000  1993.000000  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# inspection\n",
    "df.describe().transpose()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first operation is to normalize the features of the dataset. When a network is fit on unscaled data that has a range of values large inputs can slow down the learning and convergence of your network and in some cases prevent the network from effectively learning your problem. To solve this issue, it is highly recommended to normalize the dataset. Normalization is a rescaling of the data from the original range so that all values are within the range of 0 and 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# select values of dataframe\n",
    "values = df.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove output data y to rescale later\n",
    "values_x = np.delete(values, -1, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rescale data between 0 and 1\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "scaled_x = scaler.fit_transform(values_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# select values of output data y\n",
    "values_y = values[:, -1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rescale manually output data y\n",
    "scaled_y = (values_y -values_y.mean())/ values_y.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# reshape scaled date to concatenate with scaled input data\n",
    "scaled_y = scaled_y.reshape((scaled_y.shape[0], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# concatenate input data x and output data y\n",
    "scaled = np.concatenate((scaled_x, scaled_y), axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transformation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The second operation is to transform the time series into a supervised learning series. Time series forecasting problems must be re-framed as supervised learning problems. From a sequence to pairs of input and output sequences. A time series is a sequence of numbers that are ordered by a time index. This can be thought of as a list or column of ordered values. A supervised learning problem is comprised of input patterns (X) and output patterns (y), such that an algorithm can learn how to predict the output patterns from the input patterns. To do this, we used the “series_to_supervised” function of Jason Brownlee and assumed that the discharge of Ansongo and Kandadji both takes one timestep (a day) to get to Niamey."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# function to transform time series to supervised learning series\n",
    "# this function was extracted from \n",
    "# https://machinelearningmastery.com/convert-time-series-supervised-learning-problem-python/\n",
    "from pandas import DataFrame\n",
    "from pandas import concat\n",
    "\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 observations as input (X).\n",
    "        n_out: Number of observations as output (y).\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 = 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 = 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": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# using function above\n",
    "reframed = series_to_supervised(scaled, 1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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-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)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.021827</td>\n",
       "      <td>0.745837</td>\n",
       "      <td>0.772000</td>\n",
       "      <td>0.007353</td>\n",
       "      <td>0.009230</td>\n",
       "      <td>-1.239013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.030442</td>\n",
       "      <td>0.794020</td>\n",
       "      <td>0.783623</td>\n",
       "      <td>0.006828</td>\n",
       "      <td>0.007602</td>\n",
       "      <td>-1.239844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.014168</td>\n",
       "      <td>0.763185</td>\n",
       "      <td>0.743876</td>\n",
       "      <td>0.006828</td>\n",
       "      <td>0.006453</td>\n",
       "      <td>-1.240657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.082520</td>\n",
       "      <td>0.777211</td>\n",
       "      <td>0.812902</td>\n",
       "      <td>0.006828</td>\n",
       "      <td>0.006214</td>\n",
       "      <td>-1.241877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.011296</td>\n",
       "      <td>0.775912</td>\n",
       "      <td>0.746368</td>\n",
       "      <td>0.006828</td>\n",
       "      <td>0.005639</td>\n",
       "      <td>-1.242708</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   var1(t-1)  var2(t-1)  var3(t-1)  var4(t-1)  var5(t-1)   var6(t)\n",
       "1   0.021827   0.745837   0.772000   0.007353   0.009230 -1.239013\n",
       "2   0.030442   0.794020   0.783623   0.006828   0.007602 -1.239844\n",
       "3   0.014168   0.763185   0.743876   0.006828   0.006453 -1.240657\n",
       "4   0.082520   0.777211   0.812902   0.006828   0.006214 -1.241877\n",
       "5   0.011296   0.775912   0.746368   0.006828   0.005639 -1.242708"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# drop columns we don't want to predict\n",
    "reframed.drop(reframed.columns[[5,6,7,8,9,10]], axis=1, inplace=True)\n",
    "reframed.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5quWIq-kJlIn"
   },
   "source": [
    "## Data split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The third operation is to split the data into training, validation and training sets. \n",
    "\n",
    "The model is initially fit in a **training set**, which is a set of examples used to fit the parameters of the model.\n",
    "\n",
    "Successively, the fitted model is used to predict the responses for the observations in a second dataset called the **validation dataset**. The validation dataset provides an unbiased evaluation of a model fit on the training dataset. \n",
    "\n",
    "The **test dataset** is a dataset used to provide an unbiased evaluation of a final model fit on the training dataset. \n",
    "\n",
    "The dataset was also split into inputs and outputs. The figure below illustrates the different divisions of the dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![subdivisions of dataset](./images/split_dataset.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split into train, val and test sets\n",
    "values = reframed.values\n",
    "\n",
    "n_train_years = 7305 # days in 20 years for training\n",
    "n_val_years = 9131 - n_train_years # days in 5 years for validating\n",
    "n_val_sum = n_train_years + n_val_years\n",
    "\n",
    "train = values[:n_train_years, :]\n",
    "val = values[n_train_years:n_val_sum, :]\n",
    "test = values[n_val_sum:, :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split into input and outputs\n",
    "x_train, y_train = train[:, :-1], train[:, -1]\n",
    "x_val, y_val = val[:, :-1], val[:, -1]\n",
    "x_test, y_test = test[:, :-1], test[:, -1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To finish this transformation, the dataset was reshaped in a 3D format with the final shape of each subdivision of the dataset being: (number of samples, time steps, features)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7305, 1, 5) (7305,) (1826, 1, 5) (1826,) (1674, 1, 5) (1674,)\n"
     ]
    }
   ],
   "source": [
    "# reshape input to be 3D [samples, timesteps, features]\n",
    "x_train = x_train.reshape((x_train.shape[0], 1, x_train.shape[1]))\n",
    "x_val = x_val.reshape((x_val.shape[0], 1, x_val.shape[1]))\n",
    "x_test = x_test.reshape((x_test.shape[0], 1, x_test.shape[1]))\n",
    "print(x_train.shape, y_train.shape, x_val.shape, y_val.shape, x_test.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Whbni_sdhEz0"
   },
   "source": [
    "## LSTM Setting Up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A sequential model was used for this work. A Sequential model is appropriate for a plain stack of layers where each layer has exactly one input tensor and one output tensor. An LSTM layer was added to the model and the number of its internal units was set using the value obtained after hyperparameter optimization. A dense layer of 1 unit was added to the LSTM units to sum up the result of the LSTM units and to set it up as output data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "S21VRZ_8hv9S"
   },
   "outputs": [],
   "source": [
    "# set LSTM neural network\n",
    "model = Sequential()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that the model is defined we can compile it and specify its optimizer, learning rate and loss function. The process of minimizing any mathematical expression is called optimization. Optimizers are algorithms or methods used to change the attributes of the neural network such as weights, biases and learning rate to reduce the losses. The optimizer used here is the Adaptive Moment Estimation (Adam) optimizer. Adam optimization is a stochastic gradient descent method that is based on the adaptive estimation of first-order and second-order moments. According to Kingma the method is “computationally efficient, has little memory requirement, invariant to diagonal rescaling of gradients, and is well suited for problems that are large in terms of data/parameters”. \n",
    "\n",
    "The learning rate obtained from the hyperparameter optimization process is also set here. The last step of compiling our LSTM model would be to set its loss function. Since the present work would use a large number of samples, we would not set a regularization term because the chances of overfitting the data are low. As for the data loss function, the mean square error would be chosen because larger mistakes are attributed to bigger penalties due to the squaring inside the function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build model\n",
    "model.add(LSTM(units=30, input_shape=(x_train.shape[1], x_train.shape[2])))\n",
    "model.add(Dense(units=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "xjJSLd8GJlIs"
   },
   "outputs": [],
   "source": [
    "# set compiler\n",
    "optimizer = Adam(lr=0.01561601307656165)\n",
    "model.compile(loss='mse', optimizer=optimizer, metrics=['mae'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "hvmivlFWJlIs",
    "outputId": "bf059ad5-23d3-4a62-d001-1627bb16a5a0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm (LSTM)                  (None, 30)                4320      \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1)                 31        \n",
      "=================================================================\n",
      "Total params: 4,351\n",
      "Trainable params: 4,351\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![lstm model](../images/lstm_model.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "N0pcQkRyoDhh"
   },
   "source": [
    "## LSTM Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the training phase of our LSTM, the number of epochs and the batch size obtained after hyperparameter optimization are set in the model to fit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "58/58 - 8s - loss: 0.3072 - mae: 0.4206 - val_loss: 0.1328 - val_mae: 0.2688\n",
      "Epoch 2/100\n",
      "58/58 - 0s - loss: 0.0836 - mae: 0.2164 - val_loss: 0.0987 - val_mae: 0.2250\n",
      "Epoch 3/100\n",
      "58/58 - 0s - loss: 0.0652 - mae: 0.1894 - val_loss: 0.0802 - val_mae: 0.2019\n",
      "Epoch 4/100\n",
      "58/58 - 0s - loss: 0.0545 - mae: 0.1710 - val_loss: 0.0702 - val_mae: 0.1917\n",
      "Epoch 5/100\n",
      "58/58 - 0s - loss: 0.0464 - mae: 0.1539 - val_loss: 0.0648 - val_mae: 0.1880\n",
      "Epoch 6/100\n",
      "58/58 - 0s - loss: 0.0409 - mae: 0.1407 - val_loss: 0.0616 - val_mae: 0.1846\n",
      "Epoch 7/100\n",
      "58/58 - 0s - loss: 0.0375 - mae: 0.1324 - val_loss: 0.0595 - val_mae: 0.1806\n",
      "Epoch 8/100\n",
      "58/58 - 0s - loss: 0.0354 - mae: 0.1275 - val_loss: 0.0581 - val_mae: 0.1768\n",
      "Epoch 9/100\n",
      "58/58 - 0s - loss: 0.0341 - mae: 0.1244 - val_loss: 0.0571 - val_mae: 0.1738\n",
      "Epoch 10/100\n",
      "58/58 - 0s - loss: 0.0332 - mae: 0.1225 - val_loss: 0.0564 - val_mae: 0.1713\n",
      "Epoch 11/100\n",
      "58/58 - 0s - loss: 0.0326 - mae: 0.1211 - val_loss: 0.0559 - val_mae: 0.1694\n",
      "Epoch 12/100\n",
      "58/58 - 0s - loss: 0.0322 - mae: 0.1201 - val_loss: 0.0555 - val_mae: 0.1678\n",
      "Epoch 13/100\n",
      "58/58 - 0s - loss: 0.0318 - mae: 0.1194 - val_loss: 0.0551 - val_mae: 0.1665\n",
      "Epoch 14/100\n",
      "58/58 - 0s - loss: 0.0316 - mae: 0.1188 - val_loss: 0.0548 - val_mae: 0.1653\n",
      "Epoch 15/100\n",
      "58/58 - 0s - loss: 0.0314 - mae: 0.1183 - val_loss: 0.0546 - val_mae: 0.1642\n",
      "Epoch 16/100\n",
      "58/58 - 0s - loss: 0.0312 - mae: 0.1179 - val_loss: 0.0543 - val_mae: 0.1633\n",
      "Epoch 17/100\n",
      "58/58 - 0s - loss: 0.0310 - mae: 0.1175 - val_loss: 0.0541 - val_mae: 0.1624\n",
      "Epoch 18/100\n",
      "58/58 - 0s - loss: 0.0309 - mae: 0.1172 - val_loss: 0.0539 - val_mae: 0.1616\n",
      "Epoch 19/100\n",
      "58/58 - 0s - loss: 0.0307 - mae: 0.1169 - val_loss: 0.0538 - val_mae: 0.1608\n",
      "Epoch 20/100\n",
      "58/58 - 0s - loss: 0.0306 - mae: 0.1166 - val_loss: 0.0536 - val_mae: 0.1601\n",
      "Epoch 21/100\n",
      "58/58 - 0s - loss: 0.0305 - mae: 0.1163 - val_loss: 0.0534 - val_mae: 0.1594\n",
      "Epoch 22/100\n",
      "58/58 - 0s - loss: 0.0304 - mae: 0.1161 - val_loss: 0.0533 - val_mae: 0.1588\n",
      "Epoch 23/100\n",
      "58/58 - 0s - loss: 0.0302 - mae: 0.1158 - val_loss: 0.0532 - val_mae: 0.1582\n",
      "Epoch 24/100\n",
      "58/58 - 0s - loss: 0.0301 - mae: 0.1155 - val_loss: 0.0530 - val_mae: 0.1576\n",
      "Epoch 25/100\n",
      "58/58 - 0s - loss: 0.0300 - mae: 0.1153 - val_loss: 0.0529 - val_mae: 0.1570\n",
      "Epoch 26/100\n",
      "58/58 - 0s - loss: 0.0299 - mae: 0.1150 - val_loss: 0.0528 - val_mae: 0.1565\n",
      "Epoch 27/100\n",
      "58/58 - 0s - loss: 0.0298 - mae: 0.1148 - val_loss: 0.0527 - val_mae: 0.1560\n",
      "Epoch 28/100\n",
      "58/58 - 0s - loss: 0.0297 - mae: 0.1145 - val_loss: 0.0526 - val_mae: 0.1555\n",
      "Epoch 29/100\n",
      "58/58 - 0s - loss: 0.0296 - mae: 0.1143 - val_loss: 0.0526 - val_mae: 0.1551\n",
      "Epoch 30/100\n",
      "58/58 - 0s - loss: 0.0295 - mae: 0.1141 - val_loss: 0.0525 - val_mae: 0.1546\n",
      "Epoch 31/100\n",
      "58/58 - 0s - loss: 0.0294 - mae: 0.1138 - val_loss: 0.0524 - val_mae: 0.1542\n",
      "Epoch 32/100\n",
      "58/58 - 0s - loss: 0.0293 - mae: 0.1136 - val_loss: 0.0524 - val_mae: 0.1538\n",
      "Epoch 33/100\n",
      "58/58 - 0s - loss: 0.0292 - mae: 0.1134 - val_loss: 0.0523 - val_mae: 0.1534\n",
      "Epoch 34/100\n",
      "58/58 - 0s - loss: 0.0291 - mae: 0.1132 - val_loss: 0.0523 - val_mae: 0.1530\n",
      "Epoch 35/100\n",
      "58/58 - 0s - loss: 0.0290 - mae: 0.1130 - val_loss: 0.0522 - val_mae: 0.1527\n",
      "Epoch 36/100\n",
      "58/58 - 0s - loss: 0.0289 - mae: 0.1128 - val_loss: 0.0522 - val_mae: 0.1523\n",
      "Epoch 37/100\n",
      "58/58 - 0s - loss: 0.0288 - mae: 0.1126 - val_loss: 0.0521 - val_mae: 0.1519\n",
      "Epoch 38/100\n",
      "58/58 - 0s - loss: 0.0288 - mae: 0.1124 - val_loss: 0.0521 - val_mae: 0.1516\n",
      "Epoch 39/100\n",
      "58/58 - 0s - loss: 0.0287 - mae: 0.1123 - val_loss: 0.0521 - val_mae: 0.1512\n",
      "Epoch 40/100\n",
      "58/58 - 0s - loss: 0.0286 - mae: 0.1121 - val_loss: 0.0520 - val_mae: 0.1509\n",
      "Epoch 41/100\n",
      "58/58 - 0s - loss: 0.0285 - mae: 0.1119 - val_loss: 0.0520 - val_mae: 0.1506\n",
      "Epoch 42/100\n",
      "58/58 - 0s - loss: 0.0284 - mae: 0.1117 - val_loss: 0.0520 - val_mae: 0.1503\n",
      "Epoch 43/100\n",
      "58/58 - 0s - loss: 0.0284 - mae: 0.1116 - val_loss: 0.0519 - val_mae: 0.1499\n",
      "Epoch 44/100\n",
      "58/58 - 0s - loss: 0.0283 - mae: 0.1114 - val_loss: 0.0519 - val_mae: 0.1496\n",
      "Epoch 45/100\n",
      "58/58 - 0s - loss: 0.0282 - mae: 0.1113 - val_loss: 0.0519 - val_mae: 0.1493\n",
      "Epoch 46/100\n",
      "58/58 - 0s - loss: 0.0281 - mae: 0.1111 - val_loss: 0.0519 - val_mae: 0.1490\n",
      "Epoch 47/100\n",
      "58/58 - 0s - loss: 0.0280 - mae: 0.1109 - val_loss: 0.0519 - val_mae: 0.1487\n",
      "Epoch 48/100\n",
      "58/58 - 0s - loss: 0.0280 - mae: 0.1108 - val_loss: 0.0519 - val_mae: 0.1483\n",
      "Epoch 49/100\n",
      "58/58 - 0s - loss: 0.0279 - mae: 0.1106 - val_loss: 0.0519 - val_mae: 0.1480\n",
      "Epoch 50/100\n",
      "58/58 - 0s - loss: 0.0278 - mae: 0.1104 - val_loss: 0.0519 - val_mae: 0.1477\n",
      "Epoch 51/100\n",
      "58/58 - 0s - loss: 0.0277 - mae: 0.1102 - val_loss: 0.0519 - val_mae: 0.1474\n",
      "Epoch 52/100\n",
      "58/58 - 0s - loss: 0.0276 - mae: 0.1101 - val_loss: 0.0519 - val_mae: 0.1471\n",
      "Epoch 53/100\n",
      "58/58 - 0s - loss: 0.0275 - mae: 0.1099 - val_loss: 0.0519 - val_mae: 0.1468\n",
      "Epoch 54/100\n",
      "58/58 - 0s - loss: 0.0275 - mae: 0.1097 - val_loss: 0.0519 - val_mae: 0.1465\n",
      "Epoch 55/100\n",
      "58/58 - 0s - loss: 0.0274 - mae: 0.1095 - val_loss: 0.0519 - val_mae: 0.1462\n",
      "Epoch 56/100\n",
      "58/58 - 0s - loss: 0.0273 - mae: 0.1093 - val_loss: 0.0520 - val_mae: 0.1460\n",
      "Epoch 57/100\n",
      "58/58 - 0s - loss: 0.0272 - mae: 0.1092 - val_loss: 0.0520 - val_mae: 0.1458\n",
      "Epoch 58/100\n",
      "58/58 - 0s - loss: 0.0272 - mae: 0.1090 - val_loss: 0.0520 - val_mae: 0.1455\n",
      "Epoch 59/100\n",
      "58/58 - 0s - loss: 0.0271 - mae: 0.1088 - val_loss: 0.0521 - val_mae: 0.1454\n",
      "Epoch 60/100\n",
      "58/58 - 0s - loss: 0.0270 - mae: 0.1086 - val_loss: 0.0521 - val_mae: 0.1452\n",
      "Epoch 61/100\n",
      "58/58 - 0s - loss: 0.0269 - mae: 0.1085 - val_loss: 0.0522 - val_mae: 0.1450\n",
      "Epoch 62/100\n",
      "58/58 - 0s - loss: 0.0269 - mae: 0.1083 - val_loss: 0.0522 - val_mae: 0.1449\n",
      "Epoch 63/100\n",
      "58/58 - 0s - loss: 0.0268 - mae: 0.1082 - val_loss: 0.0523 - val_mae: 0.1448\n",
      "Epoch 64/100\n",
      "58/58 - 0s - loss: 0.0268 - mae: 0.1080 - val_loss: 0.0523 - val_mae: 0.1446\n",
      "Epoch 65/100\n",
      "58/58 - 0s - loss: 0.0267 - mae: 0.1079 - val_loss: 0.0524 - val_mae: 0.1446\n",
      "Epoch 66/100\n",
      "58/58 - 0s - loss: 0.0267 - mae: 0.1078 - val_loss: 0.0525 - val_mae: 0.1445\n",
      "Epoch 67/100\n",
      "58/58 - 0s - loss: 0.0266 - mae: 0.1076 - val_loss: 0.0525 - val_mae: 0.1444\n",
      "Epoch 68/100\n",
      "58/58 - 0s - loss: 0.0266 - mae: 0.1075 - val_loss: 0.0526 - val_mae: 0.1444\n",
      "Epoch 69/100\n",
      "58/58 - 0s - loss: 0.0265 - mae: 0.1074 - val_loss: 0.0527 - val_mae: 0.1443\n",
      "Epoch 70/100\n",
      "58/58 - 0s - loss: 0.0265 - mae: 0.1073 - val_loss: 0.0527 - val_mae: 0.1443\n",
      "Epoch 71/100\n",
      "58/58 - 0s - loss: 0.0264 - mae: 0.1072 - val_loss: 0.0528 - val_mae: 0.1443\n",
      "Epoch 72/100\n",
      "58/58 - 0s - loss: 0.0264 - mae: 0.1072 - val_loss: 0.0528 - val_mae: 0.1443\n",
      "Epoch 73/100\n",
      "58/58 - 0s - loss: 0.0264 - mae: 0.1071 - val_loss: 0.0529 - val_mae: 0.1443\n",
      "Epoch 74/100\n",
      "58/58 - 0s - loss: 0.0264 - mae: 0.1070 - val_loss: 0.0530 - val_mae: 0.1444\n",
      "Epoch 75/100\n",
      "58/58 - 0s - loss: 0.0263 - mae: 0.1070 - val_loss: 0.0530 - val_mae: 0.1444\n",
      "Epoch 76/100\n",
      "58/58 - 0s - loss: 0.0263 - mae: 0.1069 - val_loss: 0.0531 - val_mae: 0.1444\n",
      "Epoch 77/100\n",
      "58/58 - 0s - loss: 0.0263 - mae: 0.1068 - val_loss: 0.0531 - val_mae: 0.1445\n",
      "Epoch 78/100\n",
      "58/58 - 0s - loss: 0.0263 - mae: 0.1068 - val_loss: 0.0532 - val_mae: 0.1445\n",
      "Epoch 79/100\n",
      "58/58 - 0s - loss: 0.0262 - mae: 0.1068 - val_loss: 0.0532 - val_mae: 0.1446\n",
      "Epoch 80/100\n",
      "58/58 - 0s - loss: 0.0262 - mae: 0.1067 - val_loss: 0.0533 - val_mae: 0.1446\n",
      "Epoch 81/100\n",
      "58/58 - 0s - loss: 0.0262 - mae: 0.1067 - val_loss: 0.0534 - val_mae: 0.1447\n",
      "Epoch 82/100\n",
      "58/58 - 0s - loss: 0.0262 - mae: 0.1066 - val_loss: 0.0534 - val_mae: 0.1447\n",
      "Epoch 83/100\n",
      "58/58 - 0s - loss: 0.0262 - mae: 0.1066 - val_loss: 0.0535 - val_mae: 0.1448\n",
      "Epoch 84/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1066 - val_loss: 0.0535 - val_mae: 0.1448\n",
      "Epoch 85/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1065 - val_loss: 0.0536 - val_mae: 0.1449\n",
      "Epoch 86/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1065 - val_loss: 0.0536 - val_mae: 0.1450\n",
      "Epoch 87/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1065 - val_loss: 0.0537 - val_mae: 0.1451\n",
      "Epoch 88/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1065 - val_loss: 0.0537 - val_mae: 0.1452\n",
      "Epoch 89/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1065 - val_loss: 0.0538 - val_mae: 0.1453\n",
      "Epoch 90/100\n",
      "58/58 - 0s - loss: 0.0261 - mae: 0.1064 - val_loss: 0.0538 - val_mae: 0.1454\n",
      "Epoch 91/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0539 - val_mae: 0.1455\n",
      "Epoch 92/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0539 - val_mae: 0.1456\n",
      "Epoch 93/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0540 - val_mae: 0.1457\n",
      "Epoch 94/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0540 - val_mae: 0.1458\n",
      "Epoch 95/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0541 - val_mae: 0.1459\n",
      "Epoch 96/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0541 - val_mae: 0.1461\n",
      "Epoch 97/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0542 - val_mae: 0.1462\n",
      "Epoch 98/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0543 - val_mae: 0.1463\n",
      "Epoch 99/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0543 - val_mae: 0.1465\n",
      "Epoch 100/100\n",
      "58/58 - 0s - loss: 0.0260 - mae: 0.1064 - val_loss: 0.0544 - val_mae: 0.1467\n",
      "Wall time: 30.9 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "history = model.fit(x_train, y_train, epochs=100, batch_size=128, validation_data=(x_val, y_val), verbose=2, shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JpEmavvGwyJ9"
   },
   "source": [
    "## LSTM Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The evaluation of the model is done using the test dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ISD1CdNFmasu",
    "outputId": "8a556c8e-20d6-471a-c75f-3a9ca9b863b0",
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x238e885ea58>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plot history\n",
    "plt.figure(figsize=(8,6))\n",
    "plt.plot(history.history['loss'], label='Training')\n",
    "plt.plot(history.history['val_loss'], label='Validation')\n",
    "#plt.title('LSTM Model Loss')\n",
    "plt.xlabel('Epoch', fontsize=16)\n",
    "plt.ylabel('Loss', fontsize=16)\n",
    "plt.legend()\n",
    "#plt.savefig('../images/evaluation_lstm.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The trend of the loss curve shows that the model converged reasonably quickly and both train and test performance remained equivalent. The performance and convergence behavior of the model suggest that mean squared error is a good match for a neural network learning this problem. The validation curve being above the the training curve might suggest that the model is overfitting but the very small difference between the losses means that the model does not have significant overfitting issues. Moreover, the training/validation plot implies that the model presents low bias and low variance which attests that the model has a right balance and is neither overfitting nor underfitting."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LSTM Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "pUqpSIEAK35C"
   },
   "outputs": [],
   "source": [
    "# make a prediction\n",
    "yhat = model.predict(x_test)\n",
    "x_test = x_test.reshape((x_test.shape[0], x_test.shape[2]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "BLfMzm2gK308"
   },
   "outputs": [],
   "source": [
    "# rescaling predictions\n",
    "inv_yhat = concatenate((yhat, x_test), axis=1)\n",
    "inv_yhat = yhat * values_y.std() + values_y.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get initial values of y_test\n",
    "inv_y = df['Niamey']\n",
    "inv_y = inv_y[n_val_sum+1:].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next step would be to evaluate our LSTM model over the test set of data. The figure below presents the observed discharge vs the predicted discharge."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x238e87d79b0>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot predicitons\n",
    "plt.figure(figsize=(8,6))\n",
    "plt.plot(inv_y, label='observations')\n",
    "plt.plot(inv_yhat, label='predictions')\n",
    "#plt.title('Predicted vs Observed')\n",
    "plt.xlabel('Days', fontsize=16)\n",
    "plt.ylabel('Discharge ($m^3/s$)', fontsize=16)\n",
    "plt.legend()\n",
    "#plt.savefig('../images/prediction_vs_observation_lstm.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When presented with the LSTM model with a new set of data, we notice that the Guinean flood is well simulated while the red flood is not well simulated. The peaks of the red flood are underestimated by the deep learning model. This may be explained by the fact that maybe the right set of climate variables were not included in the model because the red flood originates from the climatic conditions within the Ansongo-Niamey basin."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2: 0.951\n"
     ]
    }
   ],
   "source": [
    "# Coefficient of Determination (R2)\n",
    "inv_yhat = inv_yhat = inv_yhat.reshape(inv_yhat.shape[0])\n",
    "correlation_matrix = np.corrcoef(inv_y, inv_yhat)\n",
    "correlation_xy = correlation_matrix[0,1]\n",
    "R2 = correlation_xy**2\n",
    "print('R2: %.3f' % R2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NSE: 0.933\n"
     ]
    }
   ],
   "source": [
    "# Nash-Sutcliffe Efficiency (NSE) \n",
    "nse = 1 - ( sum((inv_y - inv_yhat) ** 2 ) / sum( (inv_y - mean(inv_y)) ** 2) ) \n",
    "print('NSE: %.3f' % nse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "UpDh39lWK3th"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 153.730\n"
     ]
    }
   ],
   "source": [
    "# Root Meam Square Error (RMSE)\n",
    "rmse = sqrt(1/len(inv_y)* sum((inv_y - inv_yhat) ** 2 ))\n",
    "print('RMSE: %.3f' % rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The R2, NSE and RMSE calculated were respectively 0.951, 0.933 and 153.730. This confirms the fact that the LSTM model is overall good at simulating discharge in general and particularly at the outlet of the Ansongo-Niamey basin."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert lstm data from list csv\n",
    "lstm_loss = pd.DataFrame(list(zip(history.history['loss'], history.history['val_loss'])), \n",
    "                         columns = ['lstm_loss', 'lstm_val_loss'])\n",
    "lstm_prediction = pd.DataFrame(inv_yhat, columns = ['lstm_prediction'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#save lstm data to csv\n",
    "lstm_loss.to_csv('./model_data/lstm_loss.csv', index=False)\n",
    "lstm_prediction.to_csv('./model_data/lstm_prediction.csv', index=False)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "model_test3.ipynb",
   "provenance": []
  },
  "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": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}