Stock_price_sudeep.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_csv = pd.read_csv('/home/sudeepgarg865660/iitr-deep-learning-spl-tf2/processing_sequences_using_rnn_cnns/BAJFINANCE.NS.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>Date</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2002-07-01</td>\n",
       "      <td>5.239737</td>\n",
       "      <td>5.540927</td>\n",
       "      <td>5.239737</td>\n",
       "      <td>5.412679</td>\n",
       "      <td>4.001257</td>\n",
       "      <td>21923.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2002-07-02</td>\n",
       "      <td>5.523439</td>\n",
       "      <td>5.537041</td>\n",
       "      <td>5.421423</td>\n",
       "      <td>5.443769</td>\n",
       "      <td>4.024240</td>\n",
       "      <td>61044.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2002-07-03</td>\n",
       "      <td>5.505950</td>\n",
       "      <td>5.545785</td>\n",
       "      <td>5.328151</td>\n",
       "      <td>5.466116</td>\n",
       "      <td>4.040760</td>\n",
       "      <td>34160.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2002-07-04</td>\n",
       "      <td>5.501093</td>\n",
       "      <td>5.572018</td>\n",
       "      <td>5.417537</td>\n",
       "      <td>5.528297</td>\n",
       "      <td>4.086726</td>\n",
       "      <td>27892.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2002-07-05</td>\n",
       "      <td>5.563274</td>\n",
       "      <td>5.656546</td>\n",
       "      <td>5.474860</td>\n",
       "      <td>5.590478</td>\n",
       "      <td>4.132692</td>\n",
       "      <td>58976.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date      Open      High       Low     Close  Adj Close   Volume\n",
       "0  2002-07-01  5.239737  5.540927  5.239737  5.412679   4.001257  21923.0\n",
       "1  2002-07-02  5.523439  5.537041  5.421423  5.443769   4.024240  61044.0\n",
       "2  2002-07-03  5.505950  5.545785  5.328151  5.466116   4.040760  34160.0\n",
       "3  2002-07-04  5.501093  5.572018  5.417537  5.528297   4.086726  27892.0\n",
       "4  2002-07-05  5.563274  5.656546  5.474860  5.590478   4.132692  58976.0"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_csv.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4.783000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>881.251588</td>\n",
       "      <td>894.893722</td>\n",
       "      <td>866.828719</td>\n",
       "      <td>881.173791</td>\n",
       "      <td>875.709215</td>\n",
       "      <td>9.259254e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1529.944729</td>\n",
       "      <td>1552.382819</td>\n",
       "      <td>1507.610521</td>\n",
       "      <td>1530.507254</td>\n",
       "      <td>1527.947657</td>\n",
       "      <td>2.042785e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3.673548</td>\n",
       "      <td>4.081612</td>\n",
       "      <td>3.673548</td>\n",
       "      <td>4.064124</td>\n",
       "      <td>3.004354</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>27.824630</td>\n",
       "      <td>28.569833</td>\n",
       "      <td>27.282487</td>\n",
       "      <td>27.908670</td>\n",
       "      <td>25.080693</td>\n",
       "      <td>7.395600e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>76.560600</td>\n",
       "      <td>77.969398</td>\n",
       "      <td>75.316978</td>\n",
       "      <td>76.405151</td>\n",
       "      <td>72.249939</td>\n",
       "      <td>2.433500e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1047.750000</td>\n",
       "      <td>1065.525024</td>\n",
       "      <td>1025.000000</td>\n",
       "      <td>1047.719971</td>\n",
       "      <td>1037.340332</td>\n",
       "      <td>8.984685e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7949.850098</td>\n",
       "      <td>7999.000000</td>\n",
       "      <td>7804.649902</td>\n",
       "      <td>7899.049805</td>\n",
       "      <td>7899.049805</td>\n",
       "      <td>2.596005e+07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Open         High          Low        Close    Adj Close  \\\n",
       "count  4783.000000  4783.000000  4783.000000  4783.000000  4783.000000   \n",
       "mean    881.251588   894.893722   866.828719   881.173791   875.709215   \n",
       "std    1529.944729  1552.382819  1507.610521  1530.507254  1527.947657   \n",
       "min       3.673548     4.081612     3.673548     4.064124     3.004354   \n",
       "25%      27.824630    28.569833    27.282487    27.908670    25.080693   \n",
       "50%      76.560600    77.969398    75.316978    76.405151    72.249939   \n",
       "75%    1047.750000  1065.525024  1025.000000  1047.719971  1037.340332   \n",
       "max    7949.850098  7999.000000  7804.649902  7899.049805  7899.049805   \n",
       "\n",
       "             Volume  \n",
       "count  4.783000e+03  \n",
       "mean   9.259254e+05  \n",
       "std    2.042785e+06  \n",
       "min    0.000000e+00  \n",
       "25%    7.395600e+04  \n",
       "50%    2.433500e+05  \n",
       "75%    8.984685e+05  \n",
       "max    2.596005e+07  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_csv.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 4800 entries, 0 to 4799\n",
      "Data columns (total 7 columns):\n",
      " #   Column     Non-Null Count  Dtype  \n",
      "---  ------     --------------  -----  \n",
      " 0   Date       4800 non-null   object \n",
      " 1   Open       4783 non-null   float64\n",
      " 2   High       4783 non-null   float64\n",
      " 3   Low        4783 non-null   float64\n",
      " 4   Close      4783 non-null   float64\n",
      " 5   Adj Close  4783 non-null   float64\n",
      " 6   Volume     4783 non-null   float64\n",
      "dtypes: float64(6), object(1)\n",
      "memory usage: 262.6+ KB\n"
     ]
    }
   ],
   "source": [
    "data_csv.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_csv = data_csv.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "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>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>4.783000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>881.251588</td>\n",
       "      <td>894.893722</td>\n",
       "      <td>866.828719</td>\n",
       "      <td>881.173791</td>\n",
       "      <td>875.709215</td>\n",
       "      <td>9.259254e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1529.944729</td>\n",
       "      <td>1552.382819</td>\n",
       "      <td>1507.610521</td>\n",
       "      <td>1530.507254</td>\n",
       "      <td>1527.947657</td>\n",
       "      <td>2.042785e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3.673548</td>\n",
       "      <td>4.081612</td>\n",
       "      <td>3.673548</td>\n",
       "      <td>4.064124</td>\n",
       "      <td>3.004354</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>27.824630</td>\n",
       "      <td>28.569833</td>\n",
       "      <td>27.282487</td>\n",
       "      <td>27.908670</td>\n",
       "      <td>25.080693</td>\n",
       "      <td>7.395600e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>76.560600</td>\n",
       "      <td>77.969398</td>\n",
       "      <td>75.316978</td>\n",
       "      <td>76.405151</td>\n",
       "      <td>72.249939</td>\n",
       "      <td>2.433500e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1047.750000</td>\n",
       "      <td>1065.525024</td>\n",
       "      <td>1025.000000</td>\n",
       "      <td>1047.719971</td>\n",
       "      <td>1037.340332</td>\n",
       "      <td>8.984685e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7949.850098</td>\n",
       "      <td>7999.000000</td>\n",
       "      <td>7804.649902</td>\n",
       "      <td>7899.049805</td>\n",
       "      <td>7899.049805</td>\n",
       "      <td>2.596005e+07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Open         High          Low        Close    Adj Close  \\\n",
       "count  4783.000000  4783.000000  4783.000000  4783.000000  4783.000000   \n",
       "mean    881.251588   894.893722   866.828719   881.173791   875.709215   \n",
       "std    1529.944729  1552.382819  1507.610521  1530.507254  1527.947657   \n",
       "min       3.673548     4.081612     3.673548     4.064124     3.004354   \n",
       "25%      27.824630    28.569833    27.282487    27.908670    25.080693   \n",
       "50%      76.560600    77.969398    75.316978    76.405151    72.249939   \n",
       "75%    1047.750000  1065.525024  1025.000000  1047.719971  1037.340332   \n",
       "max    7949.850098  7999.000000  7804.649902  7899.049805  7899.049805   \n",
       "\n",
       "             Volume  \n",
       "count  4.783000e+03  \n",
       "mean   9.259254e+05  \n",
       "std    2.042785e+06  \n",
       "min    0.000000e+00  \n",
       "25%    7.395600e+04  \n",
       "50%    2.433500e+05  \n",
       "75%    8.984685e+05  \n",
       "max    2.596005e+07  "
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_csv.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = data_csv.reset_index()['Close']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4783,)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f83dc4e5c18>]"
      ]
     },
     "execution_count": 68,
     "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": [
    "plt.plot(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "scaler = MinMaxScaler(feature_range=(0,1))\n",
    "df1 = scaler.fit_transform(np.array(df1).reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1.70811583e-04]\n",
      " [1.74749525e-04]\n",
      " [1.77580056e-04]\n",
      " ...\n",
      " [9.79240000e-01]\n",
      " [9.80854949e-01]\n",
      " [9.78866382e-01]]\n"
     ]
    }
   ],
   "source": [
    "print(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4783, 1)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 4783 entries, 0 to 4799\n",
      "Data columns (total 7 columns):\n",
      " #   Column     Non-Null Count  Dtype  \n",
      "---  ------     --------------  -----  \n",
      " 0   Date       4783 non-null   object \n",
      " 1   Open       4783 non-null   float64\n",
      " 2   High       4783 non-null   float64\n",
      " 3   Low        4783 non-null   float64\n",
      " 4   Close      4783 non-null   float64\n",
      " 5   Adj Close  4783 non-null   float64\n",
      " 6   Volume     4783 non-null   float64\n",
      "dtypes: float64(6), object(1)\n",
      "memory usage: 298.9+ KB\n"
     ]
    }
   ],
   "source": [
    "data_csv.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_size = int(len(df1) * .7)\n",
    "test_size  = len(df1) - training_size\n",
    "train_data = df1[0:training_size,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3348, 1)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1435"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1435, 1)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data = df1[training_size:len(df1),:]\n",
    "test_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3348, 1435)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_size,test_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.00017081],\n",
       "       [0.00017475],\n",
       "       [0.00017758],\n",
       "       ...,\n",
       "       [0.06962785],\n",
       "       [0.06946129],\n",
       "       [0.06836375]])"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy\n",
    "# convert an array of values into a dataset matrix\n",
    "def create_dataset(dataset, time_step=1):\n",
    "\tdataX, dataY = [], []\n",
    "\tfor i in range(len(dataset)-time_step-1):\n",
    "\t\ta = dataset[i:(i+time_step), 0]   ###i=0, 0,1,2,3-----99   100 \n",
    "\t\tdataX.append(a)\n",
    "\t\tdataY.append(dataset[i + time_step, 0])\n",
    "\treturn numpy.array(dataX), numpy.array(dataY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "time_step=50\n",
    "X_train,y_train = create_dataset(train_data,time_step)\n",
    "X_test,y_test = create_dataset(test_data,time_step)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3297, 50)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3297,)\n"
     ]
    }
   ],
   "source": [
    "print(y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1384, 50)\n",
      "(1384,)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(None, None)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(X_test.shape), print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train =X_train.reshape(X_train.shape[0],X_train.shape[1] , 1)\n",
    "X_test = X_test.reshape(X_test.shape[0],X_test.shape[1] , 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3297, 50, 1)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1384, 50, 1)"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "##from tensorflow import keras\n",
    "##model=Sequential()\n",
    "##model.add(LSTM(50,return_sequences=True,input_shape=[None, 1]))\n",
    "##model.add(LSTM(50,return_sequences=True))\n",
    "#model.add(LSTM(50))\n",
    "##keras.layers.TimeDistributed(keras.layers.Dense(10))\n",
    "##model.compile(loss='mean_squared_error',optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "model = keras.models.Sequential([\n",
    "    keras.layers.LSTM(50, return_sequences=True, input_shape=[None, 1]),\n",
    "    keras.layers.LSTM(50, return_sequences=True),\n",
    "    keras.layers.LSTM(50),\n",
    "    keras.layers.Dense(1)])\n",
    "model.compile(loss='mean_squared_error',optimizer='adam')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'model2' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-31-378732ef5219>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mmodel2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msummary\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'model2' is not defined"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='mean_squared_error',optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm_16 (LSTM)               (None, None, 50)          10400     \n",
      "_________________________________________________________________\n",
      "lstm_17 (LSTM)               (None, None, 50)          20200     \n",
      "_________________________________________________________________\n",
      "lstm_18 (LSTM)               (None, 50)                20200     \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 51        \n",
      "=================================================================\n",
      "Total params: 50,851\n",
      "Trainable params: 50,851\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm_10 (LSTM)               (None, None, 50)          10400     \n",
      "_________________________________________________________________\n",
      "lstm_11 (LSTM)               (None, None, 50)          20200     \n",
      "_________________________________________________________________\n",
      "lstm_12 (LSTM)               (None, 50)                20200     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 51        \n",
      "=================================================================\n",
      "Total params: 50,851\n",
      "Trainable params: 50,851\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3297, 50, 1)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "52/52 [==============================] - 13s 155ms/step - loss: 4.0800e-05 - val_loss: 0.0048\n",
      "Epoch 2/100\n",
      "52/52 [==============================] - 7s 134ms/step - loss: 1.8105e-06 - val_loss: 0.0056\n",
      "Epoch 3/100\n",
      "52/52 [==============================] - 7s 143ms/step - loss: 1.7643e-06 - val_loss: 0.0056\n",
      "Epoch 4/100\n",
      "52/52 [==============================] - 7s 134ms/step - loss: 1.8649e-06 - val_loss: 0.0052\n",
      "Epoch 5/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.9104e-06 - val_loss: 0.0050\n",
      "Epoch 6/100\n",
      "52/52 [==============================] - 7s 132ms/step - loss: 1.8516e-06 - val_loss: 0.0037\n",
      "Epoch 7/100\n",
      "52/52 [==============================] - 7s 135ms/step - loss: 2.1845e-06 - val_loss: 0.0050\n",
      "Epoch 8/100\n",
      "52/52 [==============================] - 9s 165ms/step - loss: 1.9499e-06 - val_loss: 0.0056\n",
      "Epoch 9/100\n",
      "52/52 [==============================] - 7s 135ms/step - loss: 1.9929e-06 - val_loss: 0.0050\n",
      "Epoch 10/100\n",
      "52/52 [==============================] - 7s 130ms/step - loss: 1.9562e-06 - val_loss: 0.0041\n",
      "Epoch 11/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.8151e-06 - val_loss: 0.0056\n",
      "Epoch 12/100\n",
      "52/52 [==============================] - 8s 154ms/step - loss: 1.9832e-06 - val_loss: 0.0051\n",
      "Epoch 13/100\n",
      "52/52 [==============================] - 7s 132ms/step - loss: 1.6628e-06 - val_loss: 0.0052\n",
      "Epoch 14/100\n",
      "52/52 [==============================] - 7s 135ms/step - loss: 1.8014e-06 - val_loss: 0.0050\n",
      "Epoch 15/100\n",
      "52/52 [==============================] - 7s 132ms/step - loss: 2.2154e-06 - val_loss: 0.0051\n",
      "Epoch 16/100\n",
      "52/52 [==============================] - 8s 149ms/step - loss: 1.8999e-06 - val_loss: 0.0048\n",
      "Epoch 17/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 2.1656e-06 - val_loss: 0.0070\n",
      "Epoch 18/100\n",
      "52/52 [==============================] - 7s 143ms/step - loss: 1.8684e-06 - val_loss: 0.0053\n",
      "Epoch 19/100\n",
      "52/52 [==============================] - 7s 139ms/step - loss: 1.5809e-06 - val_loss: 0.0050\n",
      "Epoch 20/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.8481e-06 - val_loss: 0.0061\n",
      "Epoch 21/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.5995e-06 - val_loss: 0.0038\n",
      "Epoch 22/100\n",
      "52/52 [==============================] - 8s 145ms/step - loss: 1.6794e-06 - val_loss: 0.0060\n",
      "Epoch 23/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 1.8942e-06 - val_loss: 0.0057\n",
      "Epoch 24/100\n",
      "52/52 [==============================] - 8s 149ms/step - loss: 1.5956e-06 - val_loss: 0.0047\n",
      "Epoch 25/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 1.5560e-06 - val_loss: 0.0045\n",
      "Epoch 26/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.5012e-06 - val_loss: 0.0050\n",
      "Epoch 27/100\n",
      "52/52 [==============================] - 7s 139ms/step - loss: 1.4390e-06 - val_loss: 0.0041\n",
      "Epoch 28/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.4289e-06 - val_loss: 0.0043\n",
      "Epoch 29/100\n",
      "52/52 [==============================] - 7s 133ms/step - loss: 1.4271e-06 - val_loss: 0.0042\n",
      "Epoch 30/100\n",
      "52/52 [==============================] - 7s 135ms/step - loss: 1.9271e-06 - val_loss: 0.0023\n",
      "Epoch 31/100\n",
      "52/52 [==============================] - 7s 134ms/step - loss: 1.9453e-06 - val_loss: 0.0040\n",
      "Epoch 32/100\n",
      "52/52 [==============================] - 7s 133ms/step - loss: 2.0517e-06 - val_loss: 0.0045\n",
      "Epoch 33/100\n",
      "52/52 [==============================] - 8s 145ms/step - loss: 1.4220e-06 - val_loss: 0.0035\n",
      "Epoch 34/100\n",
      "52/52 [==============================] - 7s 134ms/step - loss: 1.5506e-06 - val_loss: 0.0043\n",
      "Epoch 35/100\n",
      "52/52 [==============================] - 7s 131ms/step - loss: 1.6746e-06 - val_loss: 0.0039\n",
      "Epoch 36/100\n",
      "52/52 [==============================] - 7s 132ms/step - loss: 1.6817e-06 - val_loss: 0.0037\n",
      "Epoch 37/100\n",
      "52/52 [==============================] - 7s 133ms/step - loss: 2.2426e-06 - val_loss: 0.0034\n",
      "Epoch 38/100\n",
      "52/52 [==============================] - 7s 132ms/step - loss: 1.6269e-06 - val_loss: 0.0041\n",
      "Epoch 39/100\n",
      "52/52 [==============================] - 7s 135ms/step - loss: 1.9296e-06 - val_loss: 0.0059\n",
      "Epoch 40/100\n",
      "52/52 [==============================] - 7s 133ms/step - loss: 1.3635e-06 - val_loss: 0.0032\n",
      "Epoch 41/100\n",
      "52/52 [==============================] - 8s 153ms/step - loss: 1.9279e-06 - val_loss: 0.0038\n",
      "Epoch 42/100\n",
      "52/52 [==============================] - 10s 201ms/step - loss: 1.2442e-06 - val_loss: 0.0041\n",
      "Epoch 43/100\n",
      "52/52 [==============================] - 8s 158ms/step - loss: 1.3877e-06 - val_loss: 0.0033\n",
      "Epoch 44/100\n",
      "52/52 [==============================] - 8s 150ms/step - loss: 1.2958e-06 - val_loss: 0.0057\n",
      "Epoch 45/100\n",
      "52/52 [==============================] - 7s 140ms/step - loss: 1.2500e-06 - val_loss: 0.0037\n",
      "Epoch 46/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.3912e-06 - val_loss: 0.0045\n",
      "Epoch 47/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.3788e-06 - val_loss: 0.0042\n",
      "Epoch 48/100\n",
      "52/52 [==============================] - 7s 139ms/step - loss: 1.3838e-06 - val_loss: 0.0039\n",
      "Epoch 49/100\n",
      "52/52 [==============================] - 8s 146ms/step - loss: 1.2773e-06 - val_loss: 0.0040\n",
      "Epoch 50/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.5606e-06 - val_loss: 0.0046\n",
      "Epoch 51/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.7349e-06 - val_loss: 0.0036\n",
      "Epoch 52/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.7144e-06 - val_loss: 0.0038\n",
      "Epoch 53/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.1318e-06 - val_loss: 0.0042\n",
      "Epoch 54/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.2197e-06 - val_loss: 0.0040\n",
      "Epoch 55/100\n",
      "52/52 [==============================] - 7s 140ms/step - loss: 1.6598e-06 - val_loss: 0.0031\n",
      "Epoch 56/100\n",
      "52/52 [==============================] - 8s 145ms/step - loss: 1.7719e-06 - val_loss: 0.0049\n",
      "Epoch 57/100\n",
      "52/52 [==============================] - 8s 147ms/step - loss: 1.6601e-06 - val_loss: 0.0043\n",
      "Epoch 58/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.6478e-06 - val_loss: 0.0045\n",
      "Epoch 59/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1826e-06 - val_loss: 0.0032\n",
      "Epoch 60/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.2215e-06 - val_loss: 0.0039\n",
      "Epoch 61/100\n",
      "52/52 [==============================] - 7s 136ms/step - loss: 1.1168e-06 - val_loss: 0.0044\n",
      "Epoch 62/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.2556e-06 - val_loss: 0.0043\n",
      "Epoch 63/100\n",
      "52/52 [==============================] - 7s 140ms/step - loss: 1.0596e-06 - val_loss: 0.0044\n",
      "Epoch 64/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 1.2702e-06 - val_loss: 0.0037\n",
      "Epoch 65/100\n",
      "52/52 [==============================] - 7s 143ms/step - loss: 1.6000e-06 - val_loss: 0.0044\n",
      "Epoch 66/100\n",
      "52/52 [==============================] - 7s 144ms/step - loss: 1.2050e-06 - val_loss: 0.0044\n",
      "Epoch 67/100\n",
      "52/52 [==============================] - 7s 139ms/step - loss: 1.1883e-06 - val_loss: 0.0031\n",
      "Epoch 68/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.2553e-06 - val_loss: 0.0042\n",
      "Epoch 69/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.2446e-06 - val_loss: 0.0041\n",
      "Epoch 70/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1991e-06 - val_loss: 0.0033\n",
      "Epoch 71/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.2711e-06 - val_loss: 0.0040\n",
      "Epoch 72/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.0895e-06 - val_loss: 0.0041\n",
      "Epoch 73/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.0983e-06 - val_loss: 0.0037\n",
      "Epoch 74/100\n",
      "52/52 [==============================] - 8s 146ms/step - loss: 1.0473e-06 - val_loss: 0.0034\n",
      "Epoch 75/100\n",
      "52/52 [==============================] - 8s 161ms/step - loss: 9.8937e-07 - val_loss: 0.0033\n",
      "Epoch 76/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1985e-06 - val_loss: 0.0029\n",
      "Epoch 77/100\n",
      "52/52 [==============================] - 7s 139ms/step - loss: 1.5179e-06 - val_loss: 0.0038\n",
      "Epoch 78/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1397e-06 - val_loss: 0.0039\n",
      "Epoch 79/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 9.8324e-07 - val_loss: 0.0049\n",
      "Epoch 80/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.3282e-06 - val_loss: 0.0032\n",
      "Epoch 81/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 1.3334e-06 - val_loss: 0.0042\n",
      "Epoch 82/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 9.4557e-07 - val_loss: 0.0047\n",
      "Epoch 83/100\n",
      "52/52 [==============================] - 7s 143ms/step - loss: 1.0377e-06 - val_loss: 0.0049\n",
      "Epoch 84/100\n",
      "52/52 [==============================] - 7s 140ms/step - loss: 1.1275e-06 - val_loss: 0.0046\n",
      "Epoch 85/100\n",
      "52/52 [==============================] - 8s 147ms/step - loss: 1.0249e-06 - val_loss: 0.0042\n",
      "Epoch 86/100\n",
      "52/52 [==============================] - 8s 149ms/step - loss: 1.2373e-06 - val_loss: 0.0037\n",
      "Epoch 87/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.2531e-06 - val_loss: 0.0047\n",
      "Epoch 88/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.0256e-06 - val_loss: 0.0035\n",
      "Epoch 89/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 9.8447e-07 - val_loss: 0.0036\n",
      "Epoch 90/100\n",
      "52/52 [==============================] - 7s 142ms/step - loss: 1.0365e-06 - val_loss: 0.0041\n",
      "Epoch 91/100\n",
      "52/52 [==============================] - 8s 146ms/step - loss: 1.0316e-06 - val_loss: 0.0033\n",
      "Epoch 92/100\n",
      "52/52 [==============================] - 8s 147ms/step - loss: 9.3588e-07 - val_loss: 0.0030\n",
      "Epoch 93/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.0482e-06 - val_loss: 0.0047\n",
      "Epoch 94/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1718e-06 - val_loss: 0.0037\n",
      "Epoch 95/100\n",
      "52/52 [==============================] - 7s 137ms/step - loss: 1.1820e-06 - val_loss: 0.0036\n",
      "Epoch 96/100\n",
      "52/52 [==============================] - 7s 140ms/step - loss: 1.3977e-06 - val_loss: 0.0038\n",
      "Epoch 97/100\n",
      "52/52 [==============================] - 7s 138ms/step - loss: 1.0349e-06 - val_loss: 0.0035\n",
      "Epoch 98/100\n",
      "52/52 [==============================] - 7s 143ms/step - loss: 9.8727e-07 - val_loss: 0.0036\n",
      "Epoch 99/100\n",
      "52/52 [==============================] - 8s 151ms/step - loss: 1.0445e-06 - val_loss: 0.0037\n",
      "Epoch 100/100\n",
      "52/52 [==============================] - 7s 141ms/step - loss: 9.5710e-07 - val_loss: 0.0034\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train,y_train,validation_data=(X_test,y_test),epochs=100,batch_size=64,verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save(\"models/sudeep_forecasting_several_steps_11oct_final_without_time_distribution_layer.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model = keras.models.load_model(\"models/sudeep_forecasting_several_steps_11oct_main.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.6.0'"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_predict=model.predict(X_train)\n",
    "test_predict=model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1384, 1)"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_predict.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3297, 1)"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_predict.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_predict=scaler.inverse_transform(train_predict)\n",
    "test_predict=scaler.inverse_transform(test_predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "153.2174412127819"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import math\n",
    "from sklearn.metrics import mean_squared_error\n",
    "math.sqrt(mean_squared_error(y_train,train_predict))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2865.4964264268397"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "math.sqrt(mean_squared_error(y_test,test_predict))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "### Plotting \n",
    "# shift train predictions for plotting\n",
    "look_back=50\n",
    "trainPredictPlot = numpy.empty_like(df1)\n",
    "trainPredictPlot[:, :] = np.nan\n",
    "trainPredictPlot[look_back:len(train_predict)+look_back, :] = train_predict\n",
    "# shift test predictions for plotting\n",
    "testPredictPlot = numpy.empty_like(df1)\n",
    "testPredictPlot[:, :] = numpy.nan\n",
    "testPredictPlot[len(train_predict)+(look_back*2)+1:len(df1)-1, :] = test_predict\n",
    "# plot baseline and predictions\n",
    "plt.plot(scaler.inverse_transform(df1))\n",
    "plt.plot(trainPredictPlot)\n",
    "plt.plot(testPredictPlot)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1435"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 50)"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_input=test_data[1385:].reshape(1,-1)\n",
    "x_input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_input=list(x_input)\n",
    "temp_input=temp_input[0].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.810139021454167,\n",
       " 0.7883530414727298,\n",
       " 0.784857139248813,\n",
       " 0.8015006022402944,\n",
       " 0.8049521601152605,\n",
       " 0.7909052819471479,\n",
       " 0.7878716861981856,\n",
       " 0.7839261561796873,\n",
       " 0.7810066063627102,\n",
       " 0.7807152450996321,\n",
       " 0.7847304765745022,\n",
       " 0.7792902900136369,\n",
       " 0.8072320882528528,\n",
       " 0.8114056481972739,\n",
       " 0.8283784465549053,\n",
       " 0.8431979794492551,\n",
       " 0.8547369613652377,\n",
       " 0.8836033725543627,\n",
       " 0.8791511918386216,\n",
       " 0.8775045770725828,\n",
       " 0.8818934264511685,\n",
       " 0.9070866199586209,\n",
       " 0.9525585200361557,\n",
       " 0.9519378853449756,\n",
       " 0.949258982069584,\n",
       " 0.9524191786814719,\n",
       " 0.9502596049103591,\n",
       " 0.9440467636232563,\n",
       " 0.9447941228001325,\n",
       " 0.9406712156391561,\n",
       " 0.9425584772001059,\n",
       " 0.9390119223194068,\n",
       " 0.9385812815130298,\n",
       " 0.9382456378390487,\n",
       " 0.9404368774054019,\n",
       " 0.9421658723081858,\n",
       " 0.9892083772608937,\n",
       " 0.9866497664139336,\n",
       " 1.0000000000000002,\n",
       " 0.9866181007453559,\n",
       " 0.9868777963398565,\n",
       " 0.954939753613975,\n",
       " 0.9514058772869863,\n",
       " 0.9707472924801118,\n",
       " 0.9523368603561121,\n",
       " 0.974160814440621,\n",
       " 0.9778530807952205,\n",
       " 0.9792399999667587,\n",
       " 0.9808549490642199,\n",
       " 0.9788663821897059]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.7598462]\n",
      "51\n",
      "1 day input [0.78835304 0.78485714 0.8015006  0.80495216 0.79090528 0.78787169\n",
      " 0.78392616 0.78100661 0.78071525 0.78473048 0.77929029 0.80723209\n",
      " 0.81140565 0.82837845 0.84319798 0.85473696 0.88360337 0.87915119\n",
      " 0.87750458 0.88189343 0.90708662 0.95255852 0.95193789 0.94925898\n",
      " 0.95241918 0.9502596  0.94404676 0.94479412 0.94067122 0.94255848\n",
      " 0.93901192 0.93858128 0.93824564 0.94043688 0.94216587 0.98920838\n",
      " 0.98664977 1.         0.9866181  0.9868778  0.95493975 0.95140588\n",
      " 0.97074729 0.95233686 0.97416081 0.97785308 0.97924    0.98085495\n",
      " 0.97886638 0.75984621]\n",
      "1 day output [[0.74568826]]\n",
      "2 day input [0.78485714 0.8015006  0.80495216 0.79090528 0.78787169 0.78392616\n",
      " 0.78100661 0.78071525 0.78473048 0.77929029 0.80723209 0.81140565\n",
      " 0.82837845 0.84319798 0.85473696 0.88360337 0.87915119 0.87750458\n",
      " 0.88189343 0.90708662 0.95255852 0.95193789 0.94925898 0.95241918\n",
      " 0.9502596  0.94404676 0.94479412 0.94067122 0.94255848 0.93901192\n",
      " 0.93858128 0.93824564 0.94043688 0.94216587 0.98920838 0.98664977\n",
      " 1.         0.9866181  0.9868778  0.95493975 0.95140588 0.97074729\n",
      " 0.95233686 0.97416081 0.97785308 0.97924    0.98085495 0.97886638\n",
      " 0.75984621 0.74568826]\n",
      "2 day output [[0.7192522]]\n",
      "3 day input [0.8015006  0.80495216 0.79090528 0.78787169 0.78392616 0.78100661\n",
      " 0.78071525 0.78473048 0.77929029 0.80723209 0.81140565 0.82837845\n",
      " 0.84319798 0.85473696 0.88360337 0.87915119 0.87750458 0.88189343\n",
      " 0.90708662 0.95255852 0.95193789 0.94925898 0.95241918 0.9502596\n",
      " 0.94404676 0.94479412 0.94067122 0.94255848 0.93901192 0.93858128\n",
      " 0.93824564 0.94043688 0.94216587 0.98920838 0.98664977 1.\n",
      " 0.9866181  0.9868778  0.95493975 0.95140588 0.97074729 0.95233686\n",
      " 0.97416081 0.97785308 0.97924    0.98085495 0.97886638 0.75984621\n",
      " 0.74568826 0.71925223]\n",
      "3 day output [[0.68686694]]\n",
      "4 day input [0.80495216 0.79090528 0.78787169 0.78392616 0.78100661 0.78071525\n",
      " 0.78473048 0.77929029 0.80723209 0.81140565 0.82837845 0.84319798\n",
      " 0.85473696 0.88360337 0.87915119 0.87750458 0.88189343 0.90708662\n",
      " 0.95255852 0.95193789 0.94925898 0.95241918 0.9502596  0.94404676\n",
      " 0.94479412 0.94067122 0.94255848 0.93901192 0.93858128 0.93824564\n",
      " 0.94043688 0.94216587 0.98920838 0.98664977 1.         0.9866181\n",
      " 0.9868778  0.95493975 0.95140588 0.97074729 0.95233686 0.97416081\n",
      " 0.97785308 0.97924    0.98085495 0.97886638 0.75984621 0.74568826\n",
      " 0.71925223 0.68686694]\n",
      "4 day output [[0.6531802]]\n",
      "5 day input [0.79090528 0.78787169 0.78392616 0.78100661 0.78071525 0.78473048\n",
      " 0.77929029 0.80723209 0.81140565 0.82837845 0.84319798 0.85473696\n",
      " 0.88360337 0.87915119 0.87750458 0.88189343 0.90708662 0.95255852\n",
      " 0.95193789 0.94925898 0.95241918 0.9502596  0.94404676 0.94479412\n",
      " 0.94067122 0.94255848 0.93901192 0.93858128 0.93824564 0.94043688\n",
      " 0.94216587 0.98920838 0.98664977 1.         0.9866181  0.9868778\n",
      " 0.95493975 0.95140588 0.97074729 0.95233686 0.97416081 0.97785308\n",
      " 0.97924    0.98085495 0.97886638 0.75984621 0.74568826 0.71925223\n",
      " 0.68686694 0.65318018]\n",
      "5 day output [[0.6207937]]\n",
      "6 day input [0.78787169 0.78392616 0.78100661 0.78071525 0.78473048 0.77929029\n",
      " 0.80723209 0.81140565 0.82837845 0.84319798 0.85473696 0.88360337\n",
      " 0.87915119 0.87750458 0.88189343 0.90708662 0.95255852 0.95193789\n",
      " 0.94925898 0.95241918 0.9502596  0.94404676 0.94479412 0.94067122\n",
      " 0.94255848 0.93901192 0.93858128 0.93824564 0.94043688 0.94216587\n",
      " 0.98920838 0.98664977 1.         0.9866181  0.9868778  0.95493975\n",
      " 0.95140588 0.97074729 0.95233686 0.97416081 0.97785308 0.97924\n",
      " 0.98085495 0.97886638 0.75984621 0.74568826 0.71925223 0.68686694\n",
      " 0.65318018 0.6207937 ]\n",
      "6 day output [[0.59089595]]\n",
      "7 day input [0.78392616 0.78100661 0.78071525 0.78473048 0.77929029 0.80723209\n",
      " 0.81140565 0.82837845 0.84319798 0.85473696 0.88360337 0.87915119\n",
      " 0.87750458 0.88189343 0.90708662 0.95255852 0.95193789 0.94925898\n",
      " 0.95241918 0.9502596  0.94404676 0.94479412 0.94067122 0.94255848\n",
      " 0.93901192 0.93858128 0.93824564 0.94043688 0.94216587 0.98920838\n",
      " 0.98664977 1.         0.9866181  0.9868778  0.95493975 0.95140588\n",
      " 0.97074729 0.95233686 0.97416081 0.97785308 0.97924    0.98085495\n",
      " 0.97886638 0.75984621 0.74568826 0.71925223 0.68686694 0.65318018\n",
      " 0.6207937  0.59089595]\n",
      "7 day output [[0.5638928]]\n",
      "8 day input [0.78100661 0.78071525 0.78473048 0.77929029 0.80723209 0.81140565\n",
      " 0.82837845 0.84319798 0.85473696 0.88360337 0.87915119 0.87750458\n",
      " 0.88189343 0.90708662 0.95255852 0.95193789 0.94925898 0.95241918\n",
      " 0.9502596  0.94404676 0.94479412 0.94067122 0.94255848 0.93901192\n",
      " 0.93858128 0.93824564 0.94043688 0.94216587 0.98920838 0.98664977\n",
      " 1.         0.9866181  0.9868778  0.95493975 0.95140588 0.97074729\n",
      " 0.95233686 0.97416081 0.97785308 0.97924    0.98085495 0.97886638\n",
      " 0.75984621 0.74568826 0.71925223 0.68686694 0.65318018 0.6207937\n",
      " 0.59089595 0.56389278]\n",
      "8 day output [[0.5398139]]\n",
      "9 day input [0.78071525 0.78473048 0.77929029 0.80723209 0.81140565 0.82837845\n",
      " 0.84319798 0.85473696 0.88360337 0.87915119 0.87750458 0.88189343\n",
      " 0.90708662 0.95255852 0.95193789 0.94925898 0.95241918 0.9502596\n",
      " 0.94404676 0.94479412 0.94067122 0.94255848 0.93901192 0.93858128\n",
      " 0.93824564 0.94043688 0.94216587 0.98920838 0.98664977 1.\n",
      " 0.9866181  0.9868778  0.95493975 0.95140588 0.97074729 0.95233686\n",
      " 0.97416081 0.97785308 0.97924    0.98085495 0.97886638 0.75984621\n",
      " 0.74568826 0.71925223 0.68686694 0.65318018 0.6207937  0.59089595\n",
      " 0.56389278 0.53981388]\n",
      "9 day output [[0.518524]]\n",
      "[[0.7598462104797363], [0.745688259601593], [0.7192522287368774], [0.6868669390678406], [0.6531801819801331], [0.6207937002182007], [0.59089595079422], [0.5638927817344666], [0.5398138761520386], [0.5185239911079407]]\n"
     ]
    }
   ],
   "source": [
    "# demonstrate prediction for next 10 days\n",
    "from numpy import array\n",
    "\n",
    "lst_output=[]\n",
    "n_steps=50\n",
    "i=0\n",
    "while(i<10):\n",
    "    \n",
    "    if(len(temp_input)>50):\n",
    "        #print(temp_input)\n",
    "        x_input=np.array(temp_input[1:])\n",
    "        print(\"{} day input {}\".format(i,x_input))\n",
    "        x_input=x_input.reshape(1,-1)\n",
    "        x_input = x_input.reshape((1, n_steps, 1))\n",
    "        #print(x_input)\n",
    "        yhat = model.predict(x_input, verbose=0)\n",
    "        print(\"{} day output {}\".format(i,yhat))\n",
    "        temp_input.extend(yhat[0].tolist())\n",
    "        temp_input=temp_input[1:]\n",
    "        #print(temp_input)\n",
    "        lst_output.extend(yhat.tolist())\n",
    "        i=i+1\n",
    "    else:\n",
    "        x_input = x_input.reshape((1, n_steps,1))\n",
    "        yhat = model.predict(x_input, verbose=0)\n",
    "        print(yhat[0])\n",
    "        temp_input.extend(yhat[0].tolist())\n",
    "        print(len(temp_input))\n",
    "        lst_output.extend(yhat.tolist())\n",
    "        i=i+1\n",
    "    \n",
    "\n",
    "print(lst_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "day_new=np.arange(1,51)\n",
    "day_pred=np.arange(51,61)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4783"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f83bcfe7390>]"
      ]
     },
     "execution_count": 110,
     "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": [
    "plt.plot(day_new,scaler.inverse_transform(df1[4733:]))\n",
    "plt.plot(day_pred,scaler.inverse_transform(lst_output))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f83bcf89eb8>]"
      ]
     },
     "execution_count": 111,
     "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": [
    "df3=df1.tolist()\n",
    "df3.extend(lst_output)\n",
    "plt.plot(df3[4000:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "df3=scaler.inverse_transform(df3).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f83bc242278>]"
      ]
     },
     "execution_count": 113,
     "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": [
    "plt.plot(df3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tensorflow2",
   "language": "python",
   "name": "tensorflow2"
  },
  "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"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}