regressionintro.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import math, datetime\n",
    "import numpy as np\n",
    "from sklearn import preprocessing, cross_validation, svm\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "import pickle\n",
    "\n",
    "style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 Open       High        Low      Close  Adj Close    Volume\n",
      "Date                                                                       \n",
      "2013-09-16  18.430000  18.719999  18.170000  18.450001  17.225540  33961100\n",
      "2013-09-17  18.590000  18.600000  18.299999  18.530001  17.300230  18948400\n",
      "2013-09-18  18.410000  19.000000  18.160000  18.900000  17.645674  30957700\n",
      "2013-09-19  18.959999  19.309999  18.760000  19.160000  17.888418  24470200\n",
      "2013-09-20  19.090000  19.180000  18.780001  18.940001  17.683020  18955300\n"
     ]
    }
   ],
   "source": [
    "#Reading data\n",
    "df = pd.read_csv('PETR4.SA.csv', header=0, index_col='Date', parse_dates=True)\n",
    "print(df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# High - Low percentage\n",
    "df['HL_PCT'] = (df['High'] - df['Low']) / df['Low'] * 100.0\n",
    "# Percentage change\n",
    "df['PCT_change'] = (df['Close'] - df['Open']) / df['Open'] * 100.0\n",
    "\n",
    "df = df[['Close', 'HL_PCT', 'PCT_change', 'Volume']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13\n"
     ]
    }
   ],
   "source": [
    "# Select wich col will be used to forecast the prices\n",
    "forecast_col = 'Close'\n",
    "# Fill the NaN values with -99999, the algorithm will ignore these values\n",
    "df.fillna(-99999, inplace=True)\n",
    "\n",
    "# The number of predictions that will be done\n",
    "forecast_out = int(math.ceil(0.01*len(df)))\n",
    "print(forecast_out)\n",
    "\n",
    "# Put all the values up the forecast_out number\n",
    "df['label'] = df[forecast_col].shift(-forecast_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array(df.drop(['label'], 1))\n",
    "# Processing the entire data ignoring the 'label' col\n",
    "X = preprocessing.scale(X)\n",
    "\n",
    "# Array of predictions \n",
    "X_lately = X[-forecast_out:]\n",
    "X = X[:-forecast_out]\n",
    "\n",
    "df.dropna(inplace=True)\n",
    "y = np.array(df['label'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8207421271348417\n"
     ]
    }
   ],
   "source": [
    "# Setting up the machine learning algorithm\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.2)\n",
    "\n",
    "\n",
    "# This is an example to use with big data, we don't need to train data every time we run the program\n",
    "# with Pickle we can save the data trained before.\n",
    "##clf = LinearRegression(n_jobs=-1)\n",
    "##clf.fit(X_train, y_train)\n",
    "##with open('linearregression.pickle', 'wb') as f:\n",
    "##    pickle.dump(clf, f)\n",
    "\n",
    "# And now we read the file created by pickle\n",
    "pickle_in = open('linearregression.pickle', 'rb')\n",
    "clf = pickle.load(pickle_in)\n",
    "\n",
    "# This is an interesting value, this determines how much our algorithm can be \"trusted\" in percentage\n",
    "accuracy = clf.score(X_test, y_test)\n",
    "print(accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[18.05832822 18.78399509 18.19473054 18.78409045 18.84048094 18.28818431\n",
      " 18.40067368 18.5988051  18.8047416  18.11847462 18.54127022 18.40731917\n",
      " 18.43331584] 0.8207421271348417 13\n"
     ]
    }
   ],
   "source": [
    "# Predicting some values\n",
    "forecast_set = clf.predict(X_lately)\n",
    "print(forecast_set, accuracy, forecast_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Plotting a graph with real prices and forecasts generated by our algorithm \n",
    "df['Forecast'] = np.nan\n",
    "\n",
    "last_date = df.iloc[-1].name\n",
    "last_unix = last_date.timestamp()\n",
    "one_day = 86400\n",
    "next_unix = last_unix + one_day\n",
    "\n",
    "for i in forecast_set:\n",
    "    next_date = datetime.datetime.fromtimestamp(next_unix)\n",
    "    next_unix += one_day\n",
    "    df.loc[next_date] =  [np.nan for _ in range(len(df.columns)-1)] + [i]\n",
    "\n",
    "df['Close'].plot()\n",
    "df['Forecast'].plot()\n",
    "plt.legend(loc=4)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[NbConvertApp] Converting notebook regressionintro.ipynb to script\n",
      "[NbConvertApp] Writing 2794 bytes to regressionintro.py\n"
     ]
    }
   ],
   "source": [
    "!jupyter nbconvert --to script regressionintro.ipynb\n"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}