{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Unnamed: 0  internetActivity\n",
      "0           2         50.432001\n",
      "1           2         46.158441\n",
      "2           2         35.017853\n",
      "3           2         30.943659\n",
      "4           2         32.049897\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x152004e50>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#https://github.com/ivanarielcaceres/timeseries-lstm-keras/blob/master/timeseries-prediction.ipynb\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from keras.models import Sequential\n",
    "from keras.layers import GRU, Dense\n",
    "from keras.layers import LSTM\n",
    "from keras  import callbacks\n",
    "from keras import optimizers\n",
    "import pandas as pd \n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "df = pd.read_csv('g2.csv')\n",
    "df=df.drop(['gridID','smsIn','smsOut','callIn','callOut'],axis=1)\n",
    "print(df.head())\n",
    "df.plot()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total rows: 168\n"
     ]
    },
    {
     "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>internetActivity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.043200e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.615844e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.501785e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.094366e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.204990e+07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   internetActivity\n",
       "0      5.043200e+07\n",
       "1      4.615844e+07\n",
       "2      3.501785e+07\n",
       "3      3.094366e+07\n",
       "4      3.204990e+07"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "columns_to_keep = ['internetActivity']\n",
    "df = df[columns_to_keep]\n",
    "df['internetActivity'] = df['internetActivity'].apply(lambda x: x*1000)\n",
    "print('Total rows: {}'.format(len(df)))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>internetActivity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.680000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.201730e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.099645e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.887199e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.648076e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.888482e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.668335e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.372602e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       internetActivity\n",
       "count      1.680000e+02\n",
       "mean       6.201730e+07\n",
       "std        2.099645e+07\n",
       "min        2.887199e+07\n",
       "25%        4.648076e+07\n",
       "50%        5.888482e+07\n",
       "75%        7.668335e+07\n",
       "max        1.372602e+08"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1124ccd50>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "internetActivity    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#null값\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min internetActivity    2.887199e+07\n",
      "dtype: float64\n",
      "Max internetActivity    1.372602e+08\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print('Min', np.min(df))\n",
    "print('Max', np.max(df))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min 0.0\n",
      "Max 1.0\n"
     ]
    }
   ],
   "source": [
    "dataset = df.astype('float64')\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "scaled = scaler.fit_transform(dataset)\n",
    "print('Min', np.min(scaled))\n",
    "print('Max', np.max(scaled))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.19891473]\n",
      " [0.15948645]\n",
      " [0.05670233]\n",
      " [0.01911343]\n",
      " [0.02931969]\n",
      " [0.01740058]\n",
      " [0.10112466]\n",
      " [0.32882381]\n",
      " [0.23963485]\n",
      " [0.19864216]\n",
      " [0.20776207]\n",
      " [0.27312013]\n",
      " [0.24562279]\n",
      " [0.2107183 ]\n",
      " [0.23933122]\n",
      " [0.37494352]\n",
      " [0.38074023]\n",
      " [0.44719646]\n",
      " [0.47908806]\n",
      " [0.47430953]\n",
      " [0.57145109]\n",
      " [0.57540974]\n",
      " [0.44117377]\n",
      " [0.35526852]]\n"
     ]
    }
   ],
   "source": [
    "print(scaled[:24])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train: 117\n",
      "test: 51\n"
     ]
    }
   ],
   "source": [
    "#Create RNN\n",
    "train_size = int(len(scaled) * 0.70)\n",
    "test_size = len(scaled - train_size)\n",
    "train, test = scaled[0:train_size, :], scaled[train_size: len(scaled), :]\n",
    "print('train: {}\\ntest: {}'.format(len(train), len(test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_dataset(dataset, look_back=1):\n",
    "    print(len(dataset), look_back)\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(dataset)-look_back-1):\n",
    "        a = dataset[i:(i+look_back), 0]\n",
    "        print(i)\n",
    "        print('X {} to {}'.format(i, i+look_back))\n",
    "        print(a)\n",
    "        print('Y {}'.format(i + look_back))\n",
    "        print(dataset[i + look_back, 0])\n",
    "        dataset[i + look_back, 0]\n",
    "        dataX.append(a)\n",
    "        dataY.append(dataset[i + look_back, 0])\n",
    "    return np.array(dataX), np.array(dataY)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "117 1\n",
      "0\n",
      "X 0 to 1\n",
      "[0.19891473]\n",
      "Y 1\n",
      "0.15948645465844052\n",
      "1\n",
      "X 1 to 2\n",
      "[0.15948645]\n",
      "Y 2\n",
      "0.056702327854409706\n",
      "2\n",
      "X 2 to 3\n",
      "[0.05670233]\n",
      "Y 3\n",
      "0.01911343435814472\n",
      "3\n",
      "X 3 to 4\n",
      "[0.01911343]\n",
      "Y 4\n",
      "0.029319686768286057\n",
      "4\n",
      "X 4 to 5\n",
      "[0.02931969]\n",
      "Y 5\n",
      "0.017400576158421333\n",
      "5\n",
      "X 5 to 6\n",
      "[0.01740058]\n",
      "Y 6\n",
      "0.10112466078073373\n",
      "6\n",
      "X 6 to 7\n",
      "[0.10112466]\n",
      "Y 7\n",
      "0.32882380528660454\n",
      "7\n",
      "X 7 to 8\n",
      "[0.32882381]\n",
      "Y 8\n",
      "0.23963485459260186\n",
      "8\n",
      "X 8 to 9\n",
      "[0.23963485]\n",
      "Y 9\n",
      "0.19864215905662552\n",
      "9\n",
      "X 9 to 10\n",
      "[0.19864216]\n",
      "Y 10\n",
      "0.20776206547621795\n",
      "10\n",
      "X 10 to 11\n",
      "[0.20776207]\n",
      "Y 11\n",
      "0.27312013013917896\n",
      "11\n",
      "X 11 to 12\n",
      "[0.27312013]\n",
      "Y 12\n",
      "0.24562279204990328\n",
      "12\n",
      "X 12 to 13\n",
      "[0.24562279]\n",
      "Y 13\n",
      "0.2107183024592938\n",
      "13\n",
      "X 13 to 14\n",
      "[0.2107183]\n",
      "Y 14\n",
      "0.23933122053714145\n",
      "14\n",
      "X 14 to 15\n",
      "[0.23933122]\n",
      "Y 15\n",
      "0.3749435223315562\n",
      "15\n",
      "X 15 to 16\n",
      "[0.37494352]\n",
      "Y 16\n",
      "0.38074023343871904\n",
      "16\n",
      "X 16 to 17\n",
      "[0.38074023]\n",
      "Y 17\n",
      "0.4471964584555595\n",
      "17\n",
      "X 17 to 18\n",
      "[0.44719646]\n",
      "Y 18\n",
      "0.47908805791639847\n",
      "18\n",
      "X 18 to 19\n",
      "[0.47908806]\n",
      "Y 19\n",
      "0.47430952639892426\n",
      "19\n",
      "X 19 to 20\n",
      "[0.47430953]\n",
      "Y 20\n",
      "0.5714510915871254\n",
      "20\n",
      "X 20 to 21\n",
      "[0.57145109]\n",
      "Y 21\n",
      "0.5754097434424896\n",
      "21\n",
      "X 21 to 22\n",
      "[0.57540974]\n",
      "Y 22\n",
      "0.4411737729352006\n",
      "22\n",
      "X 22 to 23\n",
      "[0.44117377]\n",
      "Y 23\n",
      "0.3552685196313384\n",
      "23\n",
      "X 23 to 24\n",
      "[0.35526852]\n",
      "Y 24\n",
      "0.25472464416011625\n",
      "24\n",
      "X 24 to 25\n",
      "[0.25472464]\n",
      "Y 25\n",
      "0.19353350969498456\n",
      "25\n",
      "X 25 to 26\n",
      "[0.19353351]\n",
      "Y 26\n",
      "0.16166784110143306\n",
      "26\n",
      "X 26 to 27\n",
      "[0.16166784]\n",
      "Y 27\n",
      "0.09120462278106467\n",
      "27\n",
      "X 27 to 28\n",
      "[0.09120462]\n",
      "Y 28\n",
      "0.048233521626078524\n",
      "28\n",
      "X 28 to 29\n",
      "[0.04823352]\n",
      "Y 29\n",
      "0.05240980815836527\n",
      "29\n",
      "X 29 to 30\n",
      "[0.05240981]\n",
      "Y 30\n",
      "0.15415562663744303\n",
      "30\n",
      "X 30 to 31\n",
      "[0.15415563]\n",
      "Y 31\n",
      "0.28819913945711323\n",
      "31\n",
      "X 31 to 32\n",
      "[0.28819914]\n",
      "Y 32\n",
      "0.30022400392355136\n",
      "32\n",
      "X 32 to 33\n",
      "[0.300224]\n",
      "Y 33\n",
      "0.2504568766168323\n",
      "33\n",
      "X 33 to 34\n",
      "[0.25045688]\n",
      "Y 34\n",
      "0.282202286952184\n",
      "34\n",
      "X 34 to 35\n",
      "[0.28220229]\n",
      "Y 35\n",
      "0.24119197583246738\n",
      "35\n",
      "X 35 to 36\n",
      "[0.24119198]\n",
      "Y 36\n",
      "0.26968471738245464\n",
      "36\n",
      "X 36 to 37\n",
      "[0.26968472]\n",
      "Y 37\n",
      "0.2769012453016391\n",
      "37\n",
      "X 37 to 38\n",
      "[0.27690125]\n",
      "Y 38\n",
      "0.22657803649543412\n",
      "38\n",
      "X 38 to 39\n",
      "[0.22657804]\n",
      "Y 39\n",
      "0.2516209184096259\n",
      "39\n",
      "X 39 to 40\n",
      "[0.25162092]\n",
      "Y 40\n",
      "0.30280674739983054\n",
      "40\n",
      "X 40 to 41\n",
      "[0.30280675]\n",
      "Y 41\n",
      "0.3492525287081462\n",
      "41\n",
      "X 41 to 42\n",
      "[0.34925253]\n",
      "Y 42\n",
      "0.4565543419010856\n",
      "42\n",
      "X 42 to 43\n",
      "[0.45655434]\n",
      "Y 43\n",
      "0.5051955405660398\n",
      "43\n",
      "X 43 to 44\n",
      "[0.50519554]\n",
      "Y 44\n",
      "0.5142115153237719\n",
      "44\n",
      "X 44 to 45\n",
      "[0.51421152]\n",
      "Y 45\n",
      "0.7469253300781242\n",
      "45\n",
      "X 45 to 46\n",
      "[0.74692533]\n",
      "Y 46\n",
      "1.0\n",
      "46\n",
      "X 46 to 47\n",
      "[1.]\n",
      "Y 47\n",
      "0.39509191080181655\n",
      "47\n",
      "X 47 to 48\n",
      "[0.39509191]\n",
      "Y 48\n",
      "0.25472464416011625\n",
      "48\n",
      "X 48 to 49\n",
      "[0.25472464]\n",
      "Y 49\n",
      "0.19353350969498456\n",
      "49\n",
      "X 49 to 50\n",
      "[0.19353351]\n",
      "Y 50\n",
      "0.16166784110143306\n",
      "50\n",
      "X 50 to 51\n",
      "[0.16166784]\n",
      "Y 51\n",
      "0.09120462278106467\n",
      "51\n",
      "X 51 to 52\n",
      "[0.09120462]\n",
      "Y 52\n",
      "0.048233521626078524\n",
      "52\n",
      "X 52 to 53\n",
      "[0.04823352]\n",
      "Y 53\n",
      "0.05240980815836527\n",
      "53\n",
      "X 53 to 54\n",
      "[0.05240981]\n",
      "Y 54\n",
      "0.15415562663744303\n",
      "54\n",
      "X 54 to 55\n",
      "[0.15415563]\n",
      "Y 55\n",
      "0.28819913945711323\n",
      "55\n",
      "X 55 to 56\n",
      "[0.28819914]\n",
      "Y 56\n",
      "0.30022400392355136\n",
      "56\n",
      "X 56 to 57\n",
      "[0.300224]\n",
      "Y 57\n",
      "0.2504568766168323\n",
      "57\n",
      "X 57 to 58\n",
      "[0.25045688]\n",
      "Y 58\n",
      "0.282202286952184\n",
      "58\n",
      "X 58 to 59\n",
      "[0.28220229]\n",
      "Y 59\n",
      "0.24119197583246738\n",
      "59\n",
      "X 59 to 60\n",
      "[0.24119198]\n",
      "Y 60\n",
      "0.26968471738245464\n",
      "60\n",
      "X 60 to 61\n",
      "[0.26968472]\n",
      "Y 61\n",
      "0.2769012453016391\n",
      "61\n",
      "X 61 to 62\n",
      "[0.27690125]\n",
      "Y 62\n",
      "0.22657803649543412\n",
      "62\n",
      "X 62 to 63\n",
      "[0.22657804]\n",
      "Y 63\n",
      "0.2516209184096259\n",
      "63\n",
      "X 63 to 64\n",
      "[0.25162092]\n",
      "Y 64\n",
      "0.30280674739983054\n",
      "64\n",
      "X 64 to 65\n",
      "[0.30280675]\n",
      "Y 65\n",
      "0.3492525287081462\n",
      "65\n",
      "X 65 to 66\n",
      "[0.34925253]\n",
      "Y 66\n",
      "0.4565543419010856\n",
      "66\n",
      "X 66 to 67\n",
      "[0.45655434]\n",
      "Y 67\n",
      "0.5051955405660398\n",
      "67\n",
      "X 67 to 68\n",
      "[0.50519554]\n",
      "Y 68\n",
      "0.5142115153237719\n",
      "68\n",
      "X 68 to 69\n",
      "[0.51421152]\n",
      "Y 69\n",
      "0.7469253300781242\n",
      "69\n",
      "X 69 to 70\n",
      "[0.74692533]\n",
      "Y 70\n",
      "1.0\n",
      "70\n",
      "X 70 to 71\n",
      "[1.]\n",
      "Y 71\n",
      "0.39509191080181655\n",
      "71\n",
      "X 71 to 72\n",
      "[0.39509191]\n",
      "Y 72\n",
      "0.25472464416011625\n",
      "72\n",
      "X 72 to 73\n",
      "[0.25472464]\n",
      "Y 73\n",
      "0.19353350969498456\n",
      "73\n",
      "X 73 to 74\n",
      "[0.19353351]\n",
      "Y 74\n",
      "0.16166784110143306\n",
      "74\n",
      "X 74 to 75\n",
      "[0.16166784]\n",
      "Y 75\n",
      "0.09120462278106467\n",
      "75\n",
      "X 75 to 76\n",
      "[0.09120462]\n",
      "Y 76\n",
      "0.048233521626078524\n",
      "76\n",
      "X 76 to 77\n",
      "[0.04823352]\n",
      "Y 77\n",
      "0.05240980815836527\n",
      "77\n",
      "X 77 to 78\n",
      "[0.05240981]\n",
      "Y 78\n",
      "0.15415562663744303\n",
      "78\n",
      "X 78 to 79\n",
      "[0.15415563]\n",
      "Y 79\n",
      "0.28819913945711323\n",
      "79\n",
      "X 79 to 80\n",
      "[0.28819914]\n",
      "Y 80\n",
      "0.30022400392355136\n",
      "80\n",
      "X 80 to 81\n",
      "[0.300224]\n",
      "Y 81\n",
      "0.2504568766168323\n",
      "81\n",
      "X 81 to 82\n",
      "[0.25045688]\n",
      "Y 82\n",
      "0.282202286952184\n",
      "82\n",
      "X 82 to 83\n",
      "[0.28220229]\n",
      "Y 83\n",
      "0.24119197583246738\n",
      "83\n",
      "X 83 to 84\n",
      "[0.24119198]\n",
      "Y 84\n",
      "0.26968471738245464\n",
      "84\n",
      "X 84 to 85\n",
      "[0.26968472]\n",
      "Y 85\n",
      "0.2769012453016391\n",
      "85\n",
      "X 85 to 86\n",
      "[0.27690125]\n",
      "Y 86\n",
      "0.22657803649543412\n",
      "86\n",
      "X 86 to 87\n",
      "[0.22657804]\n",
      "Y 87\n",
      "0.2516209184096259\n",
      "87\n",
      "X 87 to 88\n",
      "[0.25162092]\n",
      "Y 88\n",
      "0.30280674739983054\n",
      "88\n",
      "X 88 to 89\n",
      "[0.30280675]\n",
      "Y 89\n",
      "0.3492525287081462\n",
      "89\n",
      "X 89 to 90\n",
      "[0.34925253]\n",
      "Y 90\n",
      "0.4565543419010856\n",
      "90\n",
      "X 90 to 91\n",
      "[0.45655434]\n",
      "Y 91\n",
      "0.5051955405660398\n",
      "91\n",
      "X 91 to 92\n",
      "[0.50519554]\n",
      "Y 92\n",
      "0.5142115153237719\n",
      "92\n",
      "X 92 to 93\n",
      "[0.51421152]\n",
      "Y 93\n",
      "0.7469253300781242\n",
      "93\n",
      "X 93 to 94\n",
      "[0.74692533]\n",
      "Y 94\n",
      "1.0\n",
      "94\n",
      "X 94 to 95\n",
      "[1.]\n",
      "Y 95\n",
      "0.39509191080181655\n",
      "95\n",
      "X 95 to 96\n",
      "[0.39509191]\n",
      "Y 96\n",
      "0.19808404737648883\n",
      "96\n",
      "X 96 to 97\n",
      "[0.19808405]\n",
      "Y 97\n",
      "0.1594099038653644\n",
      "97\n",
      "X 97 to 98\n",
      "[0.1594099]\n",
      "Y 98\n",
      "0.0570305082146037\n",
      "98\n",
      "X 98 to 99\n",
      "[0.05703051]\n",
      "Y 99\n",
      "0.018840196967654266\n",
      "99\n",
      "X 99 to 100\n",
      "[0.0188402]\n",
      "Y 100\n",
      "0.028661271457859683\n",
      "100\n",
      "X 100 to 101\n",
      "[0.02866127]\n",
      "Y 101\n",
      "0.01649244735923605\n",
      "101\n",
      "X 101 to 102\n",
      "[0.01649245]\n",
      "Y 102\n",
      "0.09928190127933029\n",
      "102\n",
      "X 102 to 103\n",
      "[0.0992819]\n",
      "Y 103\n",
      "0.3257120709526608\n",
      "103\n",
      "X 103 to 104\n",
      "[0.32571207]\n",
      "Y 104\n",
      "0.23616959225950324\n",
      "104\n",
      "X 104 to 105\n",
      "[0.23616959]\n",
      "Y 105\n",
      "0.19554109012268245\n",
      "105\n",
      "X 105 to 106\n",
      "[0.19554109]\n",
      "Y 106\n",
      "0.204874304542287\n",
      "106\n",
      "X 106 to 107\n",
      "[0.2048743]\n",
      "Y 107\n",
      "0.27086282823991653\n",
      "107\n",
      "X 107 to 108\n",
      "[0.27086283]\n",
      "Y 108\n",
      "0.2429778882761599\n",
      "108\n",
      "X 108 to 109\n",
      "[0.24297789]\n",
      "Y 109\n",
      "0.20869346934359517\n",
      "109\n",
      "X 109 to 110\n",
      "[0.20869347]\n",
      "Y 110\n",
      "0.23726795042793386\n",
      "110\n",
      "X 110 to 111\n",
      "[0.23726795]\n",
      "Y 111\n",
      "0.3714057029590602\n",
      "111\n",
      "X 111 to 112\n",
      "[0.3714057]\n",
      "Y 112\n",
      "0.3763104449640066\n",
      "112\n",
      "X 112 to 113\n",
      "[0.37631044]\n",
      "Y 113\n",
      "0.4435399345661743\n",
      "113\n",
      "X 113 to 114\n",
      "[0.44353993]\n",
      "Y 114\n",
      "0.4742108766884144\n",
      "114\n",
      "X 114 to 115\n",
      "[0.47421088]\n",
      "Y 115\n",
      "0.47059441206538005\n",
      "51 1\n",
      "0\n",
      "X 0 to 1\n",
      "[0.57231922]\n",
      "Y 1\n",
      "0.4379282593956218\n",
      "1\n",
      "X 1 to 2\n",
      "[0.43792826]\n",
      "Y 2\n",
      "0.35290705694245966\n",
      "2\n",
      "X 2 to 3\n",
      "[0.35290706]\n",
      "Y 3\n",
      "0.23716393546112596\n",
      "3\n",
      "X 3 to 4\n",
      "[0.23716394]\n",
      "Y 4\n",
      "0.15340649449414645\n",
      "4\n",
      "X 4 to 5\n",
      "[0.15340649]\n",
      "Y 5\n",
      "0.09424374138413522\n",
      "5\n",
      "X 5 to 6\n",
      "[0.09424374]\n",
      "Y 6\n",
      "0.08198187924354283\n",
      "6\n",
      "X 6 to 7\n",
      "[0.08198188]\n",
      "Y 7\n",
      "0.026599655234762265\n",
      "7\n",
      "X 7 to 8\n",
      "[0.02659966]\n",
      "Y 8\n",
      "0.0\n",
      "8\n",
      "X 8 to 9\n",
      "[0.]\n",
      "Y 9\n",
      "0.05002197941902198\n",
      "9\n",
      "X 9 to 10\n",
      "[0.05002198]\n",
      "Y 10\n",
      "0.0822388227924582\n",
      "10\n",
      "X 10 to 11\n",
      "[0.08223882]\n",
      "Y 11\n",
      "0.15691003051616598\n",
      "11\n",
      "X 11 to 12\n",
      "[0.15691003]\n",
      "Y 12\n",
      "0.4037341028435943\n",
      "12\n",
      "X 12 to 13\n",
      "[0.4037341]\n",
      "Y 13\n",
      "0.3647101699631971\n",
      "13\n",
      "X 13 to 14\n",
      "[0.36471017]\n",
      "Y 14\n",
      "0.4660280439458598\n",
      "14\n",
      "X 14 to 15\n",
      "[0.46602804]\n",
      "Y 15\n",
      "0.48372009996390947\n",
      "15\n",
      "X 15 to 16\n",
      "[0.4837201]\n",
      "Y 16\n",
      "0.46031564353849347\n",
      "16\n",
      "X 16 to 17\n",
      "[0.46031564]\n",
      "Y 17\n",
      "0.5213564830726727\n",
      "17\n",
      "X 17 to 18\n",
      "[0.52135648]\n",
      "Y 18\n",
      "0.5976413831337681\n",
      "18\n",
      "X 18 to 19\n",
      "[0.59764138]\n",
      "Y 19\n",
      "0.47757721819136395\n",
      "19\n",
      "X 19 to 20\n",
      "[0.47757722]\n",
      "Y 20\n",
      "0.5352838443319592\n",
      "20\n",
      "X 20 to 21\n",
      "[0.53528384]\n",
      "Y 21\n",
      "0.5182579780534364\n",
      "21\n",
      "X 21 to 22\n",
      "[0.51825798]\n",
      "Y 22\n",
      "0.4633101214701571\n",
      "22\n",
      "X 22 to 23\n",
      "[0.46331012]\n",
      "Y 23\n",
      "0.4431588476344949\n",
      "23\n",
      "X 23 to 24\n",
      "[0.44315885]\n",
      "Y 24\n",
      "0.4741509943645896\n",
      "24\n",
      "X 24 to 25\n",
      "[0.47415099]\n",
      "Y 25\n",
      "0.3638723863219478\n",
      "25\n",
      "X 25 to 26\n",
      "[0.36387239]\n",
      "Y 26\n",
      "0.29627226900410003\n",
      "26\n",
      "X 26 to 27\n",
      "[0.29627227]\n",
      "Y 27\n",
      "0.1627242789441734\n",
      "27\n",
      "X 27 to 28\n",
      "[0.16272428]\n",
      "Y 28\n",
      "0.13568943421743695\n",
      "28\n",
      "X 28 to 29\n",
      "[0.13568943]\n",
      "Y 29\n",
      "0.0962225321796934\n",
      "29\n",
      "X 29 to 30\n",
      "[0.09622253]\n",
      "Y 30\n",
      "0.050775501746612994\n",
      "30\n",
      "X 30 to 31\n",
      "[0.0507755]\n",
      "Y 31\n",
      "0.018774412392277173\n",
      "31\n",
      "X 31 to 32\n",
      "[0.01877441]\n",
      "Y 32\n",
      "0.019429049742061943\n",
      "32\n",
      "X 32 to 33\n",
      "[0.01942905]\n",
      "Y 33\n",
      "0.039221248251786356\n",
      "33\n",
      "X 33 to 34\n",
      "[0.03922125]\n",
      "Y 34\n",
      "0.15658729000693616\n",
      "34\n",
      "X 34 to 35\n",
      "[0.15658729]\n",
      "Y 35\n",
      "0.20582698250027487\n",
      "35\n",
      "X 35 to 36\n",
      "[0.20582698]\n",
      "Y 36\n",
      "0.3247530404707604\n",
      "36\n",
      "X 36 to 37\n",
      "[0.32475304]\n",
      "Y 37\n",
      "0.32624946356258805\n",
      "37\n",
      "X 37 to 38\n",
      "[0.32624946]\n",
      "Y 38\n",
      "0.3500839637299334\n",
      "38\n",
      "X 38 to 39\n",
      "[0.35008396]\n",
      "Y 39\n",
      "0.4404317552426001\n",
      "39\n",
      "X 39 to 40\n",
      "[0.44043176]\n",
      "Y 40\n",
      "0.4012271757846905\n",
      "40\n",
      "X 40 to 41\n",
      "[0.40122718]\n",
      "Y 41\n",
      "0.3798319400154672\n",
      "41\n",
      "X 41 to 42\n",
      "[0.37983194]\n",
      "Y 42\n",
      "0.3671242036949293\n",
      "42\n",
      "X 42 to 43\n",
      "[0.3671242]\n",
      "Y 43\n",
      "0.4410916296130268\n",
      "43\n",
      "X 43 to 44\n",
      "[0.44109163]\n",
      "Y 44\n",
      "0.43961996053801117\n",
      "44\n",
      "X 44 to 45\n",
      "[0.43961996]\n",
      "Y 45\n",
      "0.5474096005511901\n",
      "45\n",
      "X 45 to 46\n",
      "[0.5474096]\n",
      "Y 46\n",
      "0.5412130586818085\n",
      "46\n",
      "X 46 to 47\n",
      "[0.54121306]\n",
      "Y 47\n",
      "0.5521488307426416\n",
      "47\n",
      "X 47 to 48\n",
      "[0.55214883]\n",
      "Y 48\n",
      "0.7696334419151003\n",
      "48\n",
      "X 48 to 49\n",
      "[0.76963344]\n",
      "Y 49\n",
      "0.4562007042978879\n"
     ]
    }
   ],
   "source": [
    "look_back = 1\n",
    "X_train, y_train = create_dataset(train, look_back)\n",
    "X_test, y_test = create_dataset(test, look_back)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(115, 1, 1)\n",
      "(49, 1, 1)\n"
     ]
    }
   ],
   "source": [
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))\n",
    "X_test = np.reshape(X_test, (X_test.shape[0], X_test.shape[1], 1))\n",
    "print(X_train.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      " - 0s - loss: 0.0332\n",
      "Epoch 2/100\n",
      " - 0s - loss: 0.0283\n",
      "Epoch 3/100\n",
      " - 0s - loss: 0.0275\n",
      "Epoch 4/100\n",
      " - 0s - loss: 0.0249\n",
      "Epoch 5/100\n",
      " - 0s - loss: 0.0249\n",
      "Epoch 6/100\n",
      " - 0s - loss: 0.0230\n",
      "Epoch 7/100\n",
      " - 0s - loss: 0.0237\n",
      "Epoch 8/100\n",
      " - 0s - loss: 0.0221\n",
      "Epoch 9/100\n",
      " - 0s - loss: 0.0219\n",
      "Epoch 10/100\n",
      " - 0s - loss: 0.0193\n",
      "Epoch 11/100\n",
      " - 0s - loss: 0.0169\n",
      "Epoch 12/100\n",
      " - 0s - loss: 0.0182\n",
      "Epoch 13/100\n",
      " - 0s - loss: 0.0188\n",
      "Epoch 14/100\n",
      " - 0s - loss: 0.0178\n",
      "Epoch 15/100\n",
      " - 0s - loss: 0.0177\n",
      "Epoch 16/100\n",
      " - 0s - loss: 0.0161\n",
      "Epoch 17/100\n",
      " - 0s - loss: 0.0159\n",
      "Epoch 18/100\n",
      " - 0s - loss: 0.0158\n",
      "Epoch 19/100\n",
      " - 0s - loss: 0.0158\n",
      "Epoch 20/100\n",
      " - 0s - loss: 0.0154\n",
      "Epoch 21/100\n",
      " - 0s - loss: 0.0158\n",
      "Epoch 22/100\n",
      " - 0s - loss: 0.0158\n",
      "Epoch 23/100\n",
      " - 0s - loss: 0.0147\n",
      "Epoch 24/100\n",
      " - 0s - loss: 0.0152\n",
      "Epoch 25/100\n",
      " - 0s - loss: 0.0146\n",
      "Epoch 26/100\n",
      " - 0s - loss: 0.0152\n",
      "Epoch 27/100\n",
      " - 0s - loss: 0.0147\n",
      "Epoch 28/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 29/100\n",
      " - 0s - loss: 0.0157\n",
      "Epoch 30/100\n",
      " - 0s - loss: 0.0144\n",
      "Epoch 31/100\n",
      " - 0s - loss: 0.0147\n",
      "Epoch 32/100\n",
      " - 0s - loss: 0.0142\n",
      "Epoch 33/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 34/100\n",
      " - 0s - loss: 0.0136\n",
      "Epoch 35/100\n",
      " - 0s - loss: 0.0146\n",
      "Epoch 36/100\n",
      " - 0s - loss: 0.0145\n",
      "Epoch 37/100\n",
      " - 0s - loss: 0.0147\n",
      "Epoch 38/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 39/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 40/100\n",
      " - 0s - loss: 0.0143\n",
      "Epoch 41/100\n",
      " - 0s - loss: 0.0139\n",
      "Epoch 42/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 43/100\n",
      " - 0s - loss: 0.0144\n",
      "Epoch 44/100\n",
      " - 0s - loss: 0.0143\n",
      "Epoch 45/100\n",
      " - 0s - loss: 0.0150\n",
      "Epoch 46/100\n",
      " - 0s - loss: 0.0150\n",
      "Epoch 47/100\n",
      " - 0s - loss: 0.0148\n",
      "Epoch 48/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 49/100\n",
      " - 0s - loss: 0.0133\n",
      "Epoch 50/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 51/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 52/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 53/100\n",
      " - 0s - loss: 0.0142\n",
      "Epoch 54/100\n",
      " - 0s - loss: 0.0154\n",
      "Epoch 55/100\n",
      " - 0s - loss: 0.0136\n",
      "Epoch 56/100\n",
      " - 0s - loss: 0.0136\n",
      "Epoch 57/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 58/100\n",
      " - 0s - loss: 0.0143\n",
      "Epoch 59/100\n",
      " - 0s - loss: 0.0142\n",
      "Epoch 60/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 61/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 62/100\n",
      " - 0s - loss: 0.0134\n",
      "Epoch 63/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 64/100\n",
      " - 0s - loss: 0.0148\n",
      "Epoch 65/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 66/100\n",
      " - 0s - loss: 0.0130\n",
      "Epoch 67/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 68/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 69/100\n",
      " - 0s - loss: 0.0136\n",
      "Epoch 70/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 71/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 72/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 73/100\n",
      " - 0s - loss: 0.0134\n",
      "Epoch 74/100\n",
      " - 0s - loss: 0.0139\n",
      "Epoch 75/100\n",
      " - 0s - loss: 0.0142\n",
      "Epoch 76/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 77/100\n",
      " - 0s - loss: 0.0135\n",
      "Epoch 78/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 79/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 80/100\n",
      " - 0s - loss: 0.0133\n",
      "Epoch 81/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 82/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 83/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 84/100\n",
      " - 0s - loss: 0.0134\n",
      "Epoch 85/100\n",
      " - 0s - loss: 0.0136\n",
      "Epoch 86/100\n",
      " - 0s - loss: 0.0135\n",
      "Epoch 87/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 88/100\n",
      " - 0s - loss: 0.0142\n",
      "Epoch 89/100\n",
      " - 0s - loss: 0.0132\n",
      "Epoch 90/100\n",
      " - 0s - loss: 0.0128\n",
      "Epoch 91/100\n",
      " - 0s - loss: 0.0141\n",
      "Epoch 92/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 93/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 94/100\n",
      " - 0s - loss: 0.0137\n",
      "Epoch 95/100\n",
      " - 0s - loss: 0.0139\n",
      "Epoch 96/100\n",
      " - 0s - loss: 0.0140\n",
      "Epoch 97/100\n",
      " - 0s - loss: 0.0131\n",
      "Epoch 98/100\n",
      " - 0s - loss: 0.0133\n",
      "Epoch 99/100\n",
      " - 0s - loss: 0.0138\n",
      "Epoch 100/100\n",
      " - 0s - loss: 0.0140\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x1541c3590>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "batch_size = 1\n",
    "model = Sequential()\n",
    "model.add(LSTM(4, batch_input_shape=(batch_size, look_back, 1), stateful=True))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')\n",
    "model.fit(X_train, y_train, epochs=100, batch_size=batch_size, verbose=2, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Score: 12555007.92 RMSE\n",
      "Test Score: 9689664.50 RMSE\n"
     ]
    }
   ],
   "source": [
    "import math\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "trainPredict = model.predict(X_train, batch_size=batch_size)\n",
    "model.reset_states()\n",
    "\n",
    "testPredict = model.predict(X_test, batch_size=batch_size)\n",
    "# invert predictions\n",
    "trainPredict = scaler.inverse_transform(trainPredict)\n",
    "y_train = scaler.inverse_transform([y_train])\n",
    "testPredict = scaler.inverse_transform(testPredict)\n",
    "y_test = scaler.inverse_transform([y_test])\n",
    "# calculate root mean squared error\n",
    "trainScore = math.sqrt(mean_squared_error(y_train[0], trainPredict[:,0]))\n",
    "print('Train Score: %.2f RMSE' % (trainScore))\n",
    "testScore = math.sqrt(mean_squared_error(y_test[0], testPredict[:,0]))\n",
    "print('Test Score: %.2f RMSE' % (testScore))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "trainPredictPlot = np.empty_like(scaled)\n",
    "trainPredictPlot[:, :] = np.nan\n",
    "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict\n",
    "# shift test predictions for plotting\n",
    "testPredictPlot = np.empty_like(scaled)\n",
    "testPredictPlot[:, :] = np.nan\n",
    "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(scaled)-1, :] = testPredict\n",
    "# plot baseline and predictions\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.plot(scaler.inverse_transform(scaled))\n",
    "plt.plot(trainPredictPlot)\n",
    "plt.plot(testPredictPlot)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}