Stock_prediction_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_epochs = 100\n",
    "batch_size = 32\n",
    "split = 0.4 # best at 0.7/0.6 . Big difference for 0.3,0.4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Baseline reading with only stock prices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./datasets/AAPL.csv')"
   ]
  },
  {
   "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>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>2008-02-01</td>\n",
       "      <td>19.462856</td>\n",
       "      <td>19.512857</td>\n",
       "      <td>18.882856</td>\n",
       "      <td>19.107143</td>\n",
       "      <td>12.740701</td>\n",
       "      <td>252686000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008-02-04</td>\n",
       "      <td>19.172857</td>\n",
       "      <td>19.414286</td>\n",
       "      <td>18.774286</td>\n",
       "      <td>18.807142</td>\n",
       "      <td>12.540661</td>\n",
       "      <td>224808500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008-02-05</td>\n",
       "      <td>18.632856</td>\n",
       "      <td>19.142857</td>\n",
       "      <td>18.414286</td>\n",
       "      <td>18.480000</td>\n",
       "      <td>12.322518</td>\n",
       "      <td>285260500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008-02-06</td>\n",
       "      <td>18.690001</td>\n",
       "      <td>18.845715</td>\n",
       "      <td>17.395714</td>\n",
       "      <td>17.428572</td>\n",
       "      <td>11.621424</td>\n",
       "      <td>393318100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008-02-07</td>\n",
       "      <td>17.138571</td>\n",
       "      <td>17.825714</td>\n",
       "      <td>16.752857</td>\n",
       "      <td>17.320000</td>\n",
       "      <td>11.549030</td>\n",
       "      <td>520832900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date       Open       High        Low      Close  Adj Close  \\\n",
       "0  2008-02-01  19.462856  19.512857  18.882856  19.107143  12.740701   \n",
       "1  2008-02-04  19.172857  19.414286  18.774286  18.807142  12.540661   \n",
       "2  2008-02-05  18.632856  19.142857  18.414286  18.480000  12.322518   \n",
       "3  2008-02-06  18.690001  18.845715  17.395714  17.428572  11.621424   \n",
       "4  2008-02-07  17.138571  17.825714  16.752857  17.320000  11.549030   \n",
       "\n",
       "      Volume  \n",
       "0  252686000  \n",
       "1  224808500  \n",
       "2  285260500  \n",
       "3  393318100  \n",
       "4  520832900  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## We are only going to consider the opening price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008-02-01</td>\n",
       "      <td>19.462856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008-02-04</td>\n",
       "      <td>19.172857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008-02-05</td>\n",
       "      <td>18.632856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008-02-06</td>\n",
       "      <td>18.690001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008-02-07</td>\n",
       "      <td>17.138571</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date       Open\n",
       "0  2008-02-01  19.462856\n",
       "1  2008-02-04  19.172857\n",
       "2  2008-02-05  18.632856\n",
       "3  2008-02-06  18.690001\n",
       "4  2008-02-07  17.138571"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.drop(df.columns[2:],axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Graph of the entire dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fb51bc57da0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb51bc578d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Open'].plot(figsize=(12,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "training = split\n",
    "total = df.shape[0]\n",
    "train_index = int(total*training)\n",
    "\n",
    "training_set = df[:train_index]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "sc = MinMaxScaler(feature_range = (0, 1))\n",
    "training_set_scaled = sc.fit_transform(pd.DataFrame(training_set['Open']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Take time window to be 60 days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = []\n",
    "y_train = []\n",
    "for i in range(60, train_index):\n",
    "    X_train.append(training_set_scaled[i-60:i, 0])\n",
    "    y_train.append(training_set_scaled[i, 0])\n",
    "X_train, y_train = np.array(X_train), np.array(y_train)\n",
    "\n",
    "# Reshaping\n",
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM\n",
    "from keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the LSTM model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/aveek/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From /home/aveek/anaconda3/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True, input_shape = (X_train.shape[1], 1)))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(Dense(units = 1))\n",
    "\n",
    "model.compile(optimizer = 'adam', loss = 'mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/aveek/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/100\n",
      "1048/1048 [==============================] - 6s 6ms/step - loss: 0.0235\n",
      "Epoch 2/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0034\n",
      "Epoch 3/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0034\n",
      "Epoch 4/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0030\n",
      "Epoch 5/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0029\n",
      "Epoch 6/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0024\n",
      "Epoch 7/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0029\n",
      "Epoch 8/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0032\n",
      "Epoch 9/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0026\n",
      "Epoch 10/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0026\n",
      "Epoch 11/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0023\n",
      "Epoch 12/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0024\n",
      "Epoch 13/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0024\n",
      "Epoch 14/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0023\n",
      "Epoch 15/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0021\n",
      "Epoch 16/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0020\n",
      "Epoch 17/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0019\n",
      "Epoch 18/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0018\n",
      "Epoch 19/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0018\n",
      "Epoch 20/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0018\n",
      "Epoch 21/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0018\n",
      "Epoch 22/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0019\n",
      "Epoch 23/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0019\n",
      "Epoch 24/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0019\n",
      "Epoch 25/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0021\n",
      "Epoch 26/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0016\n",
      "Epoch 27/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0017\n",
      "Epoch 28/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0016\n",
      "Epoch 29/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0015\n",
      "Epoch 30/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0016\n",
      "Epoch 31/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0017\n",
      "Epoch 32/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0016\n",
      "Epoch 33/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0017\n",
      "Epoch 34/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0016\n",
      "Epoch 35/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0014\n",
      "Epoch 36/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0014\n",
      "Epoch 37/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 38/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 39/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0015\n",
      "Epoch 40/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 41/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0013\n",
      "Epoch 42/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0015\n",
      "Epoch 43/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 44/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0015\n",
      "Epoch 45/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0015\n",
      "Epoch 46/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0014\n",
      "Epoch 47/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 48/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0014\n",
      "Epoch 49/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 50/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0015\n",
      "Epoch 51/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0014\n",
      "Epoch 52/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0015\n",
      "Epoch 53/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 54/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 55/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 56/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 57/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 58/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 59/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 60/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 61/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 62/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 63/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 64/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 65/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 66/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 67/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0010\n",
      "Epoch 68/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 69/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 70/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.9513e-04\n",
      "Epoch 71/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0013\n",
      "Epoch 72/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0010\n",
      "Epoch 73/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.6183e-04\n",
      "Epoch 74/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 75/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.7174e-04\n",
      "Epoch 76/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.9566e-04\n",
      "Epoch 77/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0010\n",
      "Epoch 78/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0010\n",
      "Epoch 79/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.4432e-04\n",
      "Epoch 80/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.8849e-04\n",
      "Epoch 81/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 82/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0012\n",
      "Epoch 83/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 84/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 8.6156e-04\n",
      "Epoch 85/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 86/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 8.7569e-04\n",
      "Epoch 87/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0010\n",
      "Epoch 88/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.9313e-04\n",
      "Epoch 89/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.0638e-04\n",
      "Epoch 90/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 8.7854e-04\n",
      "Epoch 91/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 9.1137e-04\n",
      "Epoch 92/100\n",
      "1048/1048 [==============================] - 4s 4ms/step - loss: 0.0011\n",
      "Epoch 93/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1048/1048 [==============================] - 4s 3ms/step - loss: 9.7069e-04\n",
      "Epoch 94/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0012\n",
      "Epoch 95/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 9.4330e-04\n",
      "Epoch 96/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0010\n",
      "Epoch 97/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 9.9466e-04\n",
      "Epoch 98/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 8.9992e-04\n",
      "Epoch 99/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0012\n",
      "Epoch 100/100\n",
      "1048/1048 [==============================] - 4s 3ms/step - loss: 0.0011\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fb50b455eb8>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs = num_epochs, batch_size = batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Open</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1108</th>\n",
       "      <td>2012-06-25</td>\n",
       "      <td>82.471428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1109</th>\n",
       "      <td>2012-06-26</td>\n",
       "      <td>81.618568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1110</th>\n",
       "      <td>2012-06-27</td>\n",
       "      <td>82.142860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1111</th>\n",
       "      <td>2012-06-28</td>\n",
       "      <td>81.667145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1112</th>\n",
       "      <td>2012-06-29</td>\n",
       "      <td>82.571426</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Date       Open\n",
       "1108  2012-06-25  82.471428\n",
       "1109  2012-06-26  81.618568\n",
       "1110  2012-06-27  82.142860\n",
       "1111  2012-06-28  81.667145\n",
       "1112  2012-06-29  82.571426"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testing_set = df[train_index:]\n",
    "testing_set = pd.DataFrame(testing_set)\n",
    "testing_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "real_stock_price = pd.DataFrame(testing_set['Open']).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_set=testing_set['Open']\n",
    "test_set=pd.DataFrame(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = df['Open'][len(df) - len(testing_set) - 60:].values\n",
    "# print(inputs)\n",
    "inputs = inputs.reshape(-1,1)\n",
    "# print(inputs)\n",
    "inputs = sc.transform(inputs)\n",
    "# print(inputs)\n",
    "X_test = []\n",
    "for i in range(60, 60+(total-train_index)):\n",
    "    X_test.append(inputs[i-60:i, 0])\n",
    "X_test = np.array(X_test)\n",
    "X_test = np.reshape(X_test, (X_test.shape[0], X_test.shape[1], 1))\n",
    "predicted_stock_price = model.predict(X_test)\n",
    "predicted_stock_price = sc.inverse_transform(predicted_stock_price)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_stock_price=pd.DataFrame(predicted_stock_price)\n",
    "predicted_stock_price_old = predicted_stock_price"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting predicted vs actual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb4e8359dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = [12, 10]\n",
    "plt.plot(real_stock_price, color = 'red', label = 'Real Apple Stock Price')\n",
    "plt.plot(predicted_stock_price_old, color = 'green', label = 'Predicted Apple Stock Price')\n",
    "plt.title('Apple Stock Price Prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Apple Stock Price')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using sentiment from news articles and twitter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "from sklearn.feature_extraction.text import TfidfTransformer, TfidfVectorizer\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get the cleaned datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_news = pd.read_csv('./datasets/news_clean.csv')\n",
    "final_tweets = pd.read_csv('./datasets/tweets_clean.csv')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the stock prices + logistic regression sentiment score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_lstm = pd.read_csv('./datasets/AAPL_regression.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Add the sentiment values as new columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.03701003, 0.82142857],\n",
       "       [0.03568848, 0.14285714],\n",
       "       [0.03322765, 0.92857143],\n",
       "       ...,\n",
       "       [0.69226012, 0.60714286],\n",
       "       [0.70529339, 0.89795918],\n",
       "       [0.70916693, 0.92857143]])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "sc = MinMaxScaler(feature_range = (0, 1))\n",
    "\n",
    "dataset = pd.DataFrame(df['Open'])\n",
    "dataset['Avg'] = df_lstm['Avg']\n",
    "\n",
    "dataset_scaled = sc.fit_transform(dataset)\n",
    "\n",
    "dataset_scaled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split into training and testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "training = split\n",
    "total = dataset_scaled.shape[0]\n",
    "train_index = int(total*training)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Taking time window as 60 days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = []\n",
    "y = []\n",
    "for i in range(60, total):\n",
    "    X.append(dataset_scaled[i-60:i])\n",
    "    y.append(dataset_scaled[i])\n",
    "X, y = np.array(X), np.array(y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = X[:train_index]\n",
    "y_train = y[:train_index]\n",
    "X_test = X[train_index:]\n",
    "# X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True, input_shape = (X_train.shape[1], 2)))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(LSTM(units = 50))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "model.add(Dense(units = 2))\n",
    "\n",
    "model.compile(optimizer = 'adam', loss = 'mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "1108/1108 [==============================] - 6s 6ms/step - loss: 0.0634\n",
      "Epoch 2/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0229\n",
      "Epoch 3/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0189\n",
      "Epoch 4/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0183\n",
      "Epoch 5/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0185\n",
      "Epoch 6/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0171\n",
      "Epoch 7/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0170\n",
      "Epoch 8/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0165\n",
      "Epoch 9/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0171\n",
      "Epoch 10/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0161\n",
      "Epoch 11/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0174\n",
      "Epoch 12/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0169\n",
      "Epoch 13/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0157\n",
      "Epoch 14/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0152\n",
      "Epoch 15/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0153\n",
      "Epoch 16/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0159\n",
      "Epoch 17/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0162\n",
      "Epoch 18/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0150\n",
      "Epoch 19/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0159\n",
      "Epoch 20/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0152\n",
      "Epoch 21/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0154\n",
      "Epoch 22/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0155\n",
      "Epoch 23/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0158\n",
      "Epoch 24/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0150\n",
      "Epoch 25/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0158\n",
      "Epoch 26/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0155\n",
      "Epoch 27/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0151\n",
      "Epoch 28/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0152\n",
      "Epoch 29/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0153\n",
      "Epoch 30/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0149\n",
      "Epoch 31/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0144\n",
      "Epoch 32/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0150\n",
      "Epoch 33/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0147\n",
      "Epoch 34/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0152\n",
      "Epoch 35/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0150\n",
      "Epoch 36/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0146\n",
      "Epoch 37/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0152\n",
      "Epoch 38/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0141\n",
      "Epoch 39/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0152\n",
      "Epoch 40/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0144\n",
      "Epoch 41/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0151\n",
      "Epoch 42/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0142\n",
      "Epoch 43/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0149\n",
      "Epoch 44/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0146\n",
      "Epoch 45/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0145\n",
      "Epoch 46/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 47/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0148\n",
      "Epoch 48/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0144\n",
      "Epoch 49/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0147\n",
      "Epoch 50/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0146\n",
      "Epoch 51/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0143\n",
      "Epoch 52/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0144\n",
      "Epoch 53/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0147\n",
      "Epoch 54/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0143\n",
      "Epoch 55/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0143\n",
      "Epoch 56/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 57/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 58/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 59/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 60/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 61/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0139\n",
      "Epoch 62/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 63/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 64/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0145\n",
      "Epoch 65/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 66/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0143\n",
      "Epoch 67/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0149\n",
      "Epoch 68/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 69/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 70/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0136\n",
      "Epoch 71/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0139\n",
      "Epoch 72/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0146\n",
      "Epoch 73/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 74/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 75/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 76/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 77/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 78/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0137\n",
      "Epoch 79/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0148\n",
      "Epoch 80/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0139\n",
      "Epoch 81/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 82/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0138\n",
      "Epoch 83/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0136\n",
      "Epoch 84/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 85/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 86/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0138\n",
      "Epoch 87/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0143\n",
      "Epoch 88/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 89/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0141\n",
      "Epoch 90/100\n",
      "1108/1108 [==============================] - 5s 4ms/step - loss: 0.0137\n",
      "Epoch 91/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0139\n",
      "Epoch 92/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0139\n",
      "Epoch 93/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0138\n",
      "Epoch 94/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0140\n",
      "Epoch 95/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0135\n",
      "Epoch 96/100\n",
      "1108/1108 [==============================] - 4s 4ms/step - loss: 0.0142\n",
      "Epoch 97/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0136\n",
      "Epoch 98/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0136\n",
      "Epoch 99/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0136\n",
      "Epoch 100/100\n",
      "1108/1108 [==============================] - 4s 3ms/step - loss: 0.0134\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fb4e802ef98>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs = num_epochs, batch_size = batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1108</th>\n",
       "      <td>2012-06-25</td>\n",
       "      <td>82.471428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1109</th>\n",
       "      <td>2012-06-26</td>\n",
       "      <td>81.618568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1110</th>\n",
       "      <td>2012-06-27</td>\n",
       "      <td>82.142860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1111</th>\n",
       "      <td>2012-06-28</td>\n",
       "      <td>81.667145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1112</th>\n",
       "      <td>2012-06-29</td>\n",
       "      <td>82.571426</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Date       Open\n",
       "1108  2012-06-25  82.471428\n",
       "1109  2012-06-26  81.618568\n",
       "1110  2012-06-27  82.142860\n",
       "1111  2012-06-28  81.667145\n",
       "1112  2012-06-29  82.571426"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testing_set = df[train_index:]\n",
    "testing_set = pd.DataFrame(testing_set)\n",
    "testing_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "real_stock_price = pd.DataFrame(testing_set['Open']).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_stock_price = model.predict(X_test)\n",
    "predicted_stock_price = sc.inverse_transform(predicted_stock_price)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted=pd.DataFrame(predicted_stock_price)\n",
    "real_price = real_stock_price[60:].reset_index(drop=True)\n",
    "predicted_old = predicted_stock_price_old[60:].reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting predicted with sentimet vs predicted without sentiment vs actual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb49ed4c668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = [12, 10]\n",
    "plt.plot(real_price, color = 'red', label = 'Real Stock Price')\n",
    "plt.plot(predicted_old, color = 'green', label = 'Predicted without Sentiment score')\n",
    "plt.plot(predicted[0], color = 'blue', label = 'Predicted with Sentiment score')\n",
    "plt.title('Apple Stock Price Prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Apple Stock Price')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE of prediction with sentiment score:  185.4433103179256\n",
      "MSE of prediction without sentiment score:  791.2428374815804\n"
     ]
    }
   ],
   "source": [
    "print(\"MSE of prediction with sentiment score: \", mean_squared_error(real_price, predicted[0]))\n",
    "print(\"MSE of prediction without sentiment score: \", mean_squared_error(real_price, predicted_old))    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Observation\n",
    "\n",
    "We can see that the Mean Squared error for predicted stock prices are better when we use stock prices along with sentiment scores, which proves the "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}