{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Partie 1 - Préparation des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the libraries\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the training set\n",
    "dataset_train = pd.read_csv('Google_Stock_Price_Train.csv')\n",
    "training_set = dataset_train.iloc[:, 1:2].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature Scaling\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "sc = MinMaxScaler(feature_range = (0, 1))\n",
    "training_set_scaled = sc.fit_transform(training_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a data structure with 60 timesteps and 1 output\n",
    "X_train = []\n",
    "y_train = []\n",
    "for i in range(60, 1258):\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)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshaping\n",
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Partie 2 - RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the Keras libraries and packages\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM\n",
    "from keras.layers import Dropout\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialising the RNN\n",
    "regressor = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding the first LSTM layer and some Dropout regularisation\n",
    "regressor.add(LSTM(units = 50, return_sequences = True, input_shape = (X_train.shape[1], 1)))\n",
    "regressor.add(Dropout(0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding a second LSTM layer and some Dropout regularisation\n",
    "regressor.add(LSTM(units = 50, return_sequences = True))\n",
    "regressor.add(Dropout(0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding a third LSTM layer and some Dropout regularisation\n",
    "regressor.add(LSTM(units = 50, return_sequences = True))\n",
    "regressor.add(Dropout(0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding a fourth LSTM layer and some Dropout regularisation\n",
    "regressor.add(LSTM(units = 50))\n",
    "regressor.add(Dropout(0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding the output layer\n",
    "regressor.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compiling the RNN\n",
    "regressor.compile(optimizer = 'adam', loss = 'mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fitting the RNN to the Training set\n",
    "regressor.fit(X_train, y_train, epochs = 100, batch_size = 32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Partie 3 - Prédictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Getting the real stock price of 2017\n",
    "dataset_test = pd.read_csv('Google_Stock_Price_Test.csv')\n",
    "real_stock_price = dataset_test.iloc[:, 1:2].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Getting the predicted stock price of 2017\n",
    "dataset_total = pd.concat((dataset_train['Open'], dataset_test['Open']), axis = 0)\n",
    "inputs = dataset_total[len(dataset_total) - len(dataset_test) - 60:].values\n",
    "inputs = inputs.reshape(-1,1)\n",
    "inputs = sc.transform(inputs)\n",
    "X_test = []\n",
    "for i in range(60, 80):\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 = regressor.predict(X_test)\n",
    "predicted_stock_price = sc.inverse_transform(predicted_stock_price)\n",
    "\n",
    "# Visualising the results\n",
    "plt.plot(real_stock_price, color = 'red', label = 'Real Google Stock Price')\n",
    "plt.plot(predicted_stock_price, color = 'blue', label = 'Predicted Google Stock Price')\n",
    "plt.title('Google Stock Price Prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Google Stock Price')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}