Part_2_Training_the_binary_classifiers.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Downloading and combining all data, applying various algorithms for predictions\n",
    "This notebook is used per cashtag to dowload its combined data of tweets and stock data, then applying each of the six binary classifiers to it. The result is an overview of the accuracy of the classifiers on each cashtag, using an 80/20 Train/Test split and 10-fold cross validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import os\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn import tree\n",
    "import pydotplus\n",
    "from sklearn.externals.six import StringIO  \n",
    "from IPython.display import Image\n",
    "from sklearn.ensemble import RandomForestClassifier \n",
    "from sklearn.model_selection import cross_val_score\n",
    "import xlsxwriter\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### List all files in the folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# import glob\n",
    "\n",
    "# glob.glob(\"C:/Users/jangn/OneDrive/CODE/Stock_markets/Cashtags/*.xlsx\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### The code snippet below will initialize a blank DataFrame then append all of the individual files into the all_data DataFrame. The resulting dataframe contains all the combined data created with the first notebook (Part 1) for each cashtag;  c. 6800 rows = 74 days * 92 companies. \n",
    "https://pbpython.com/excel-file-combine.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# all_data = pd.DataFrame()\n",
    "# for f in glob.glob(\"C:/Users/jangn/OneDrive/CODE/Stock_markets/Cashtags/*.xlsx\"):\n",
    "#     df = pd.read_excel(f)\n",
    "#     all_data = all_data.append(df,ignore_index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Reading only one excel at a time\n",
    "The binary classifiers are trained on each cashtag's twitter and stock data individually."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "xls = pd.ExcelFile('C:/Users/jangn/OneDrive/CODE/Stock_markets/Cashtags/$AAL_minusones.xlsx') # Update cashtag; separate for each run.\n",
    "all_data = pd.read_excel(xls, header = 0,encoding='latin-1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Remove the last row for each cashtag, since its 'Predicted_change' is  NaN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "all_data = all_data[np.isfinite(all_data['Predicted_change_stock'])]\n",
    "all_data = all_data[np.isfinite(all_data['Buy/Sell'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "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>Volume_stock</th>\n",
       "      <th>Adj_Close_stock</th>\n",
       "      <th>HiLo_vola_stock</th>\n",
       "      <th>Pct_change_stock</th>\n",
       "      <th>Pct_change_stock_scaled</th>\n",
       "      <th>Favs</th>\n",
       "      <th>RTs</th>\n",
       "      <th>Followers</th>\n",
       "      <th>Following</th>\n",
       "      <th>Is a RT</th>\n",
       "      <th>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>Compound_multiplied</th>\n",
       "      <th>Compound_multiplied_scaled</th>\n",
       "      <th>Predicted_change_stock</th>\n",
       "      <th>Buy/Sell</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Volume_stock, Adj_Close_stock, HiLo_vola_stock, Pct_change_stock, Pct_change_stock_scaled, Favs, RTs, Followers, Following, Is a RT, compound, neg, neu, pos, Compound_multiplied, Compound_multiplied_scaled, Predicted_change_stock, Buy/Sell]\n",
       "Index: []"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nan_rows = all_data[all_data['Predicted_change_stock'].isnull()]\n",
    "nan_rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "74"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "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>Volume_stock</th>\n",
       "      <th>Adj_Close_stock</th>\n",
       "      <th>HiLo_vola_stock</th>\n",
       "      <th>Pct_change_stock</th>\n",
       "      <th>Pct_change_stock_scaled</th>\n",
       "      <th>Favs</th>\n",
       "      <th>RTs</th>\n",
       "      <th>Followers</th>\n",
       "      <th>Following</th>\n",
       "      <th>Is a RT</th>\n",
       "      <th>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>Compound_multiplied</th>\n",
       "      <th>Compound_multiplied_scaled</th>\n",
       "      <th>Predicted_change_stock</th>\n",
       "      <th>Buy/Sell</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>7.400000e+01</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>74.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>9.731227e+06</td>\n",
       "      <td>33.997063</td>\n",
       "      <td>2.866570</td>\n",
       "      <td>-0.447087</td>\n",
       "      <td>0.022178</td>\n",
       "      <td>1.873590</td>\n",
       "      <td>2.362778</td>\n",
       "      <td>4158.740790</td>\n",
       "      <td>1047.781398</td>\n",
       "      <td>0.136907</td>\n",
       "      <td>0.218461</td>\n",
       "      <td>0.055512</td>\n",
       "      <td>0.796866</td>\n",
       "      <td>0.147624</td>\n",
       "      <td>442.440350</td>\n",
       "      <td>0.022435</td>\n",
       "      <td>-0.488144</td>\n",
       "      <td>-0.243243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.143877e+06</td>\n",
       "      <td>3.336924</td>\n",
       "      <td>0.999271</td>\n",
       "      <td>1.512450</td>\n",
       "      <td>0.860596</td>\n",
       "      <td>1.489856</td>\n",
       "      <td>1.779835</td>\n",
       "      <td>6194.575281</td>\n",
       "      <td>1178.949550</td>\n",
       "      <td>0.143181</td>\n",
       "      <td>0.130938</td>\n",
       "      <td>0.027647</td>\n",
       "      <td>0.036900</td>\n",
       "      <td>0.040814</td>\n",
       "      <td>3204.139740</td>\n",
       "      <td>0.091246</td>\n",
       "      <td>1.581633</td>\n",
       "      <td>0.976586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.018200e+06</td>\n",
       "      <td>30.006098</td>\n",
       "      <td>1.373308</td>\n",
       "      <td>-5.089362</td>\n",
       "      <td>-2.619312</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.042245</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.695500</td>\n",
       "      <td>0.063667</td>\n",
       "      <td>-24255.853656</td>\n",
       "      <td>-0.680912</td>\n",
       "      <td>-5.089362</td>\n",
       "      <td>-1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>7.338450e+06</td>\n",
       "      <td>31.092009</td>\n",
       "      <td>2.060415</td>\n",
       "      <td>-1.211461</td>\n",
       "      <td>-0.412757</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1367.875962</td>\n",
       "      <td>395.832358</td>\n",
       "      <td>0.034483</td>\n",
       "      <td>0.121496</td>\n",
       "      <td>0.038277</td>\n",
       "      <td>0.780541</td>\n",
       "      <td>0.126429</td>\n",
       "      <td>124.810639</td>\n",
       "      <td>0.013390</td>\n",
       "      <td>-1.211461</td>\n",
       "      <td>-1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>9.032517e+06</td>\n",
       "      <td>32.221743</td>\n",
       "      <td>2.770293</td>\n",
       "      <td>-0.418915</td>\n",
       "      <td>0.038208</td>\n",
       "      <td>1.450000</td>\n",
       "      <td>2.305556</td>\n",
       "      <td>2583.105392</td>\n",
       "      <td>723.038462</td>\n",
       "      <td>0.099408</td>\n",
       "      <td>0.214772</td>\n",
       "      <td>0.056847</td>\n",
       "      <td>0.797208</td>\n",
       "      <td>0.137407</td>\n",
       "      <td>407.177512</td>\n",
       "      <td>0.021431</td>\n",
       "      <td>-0.418915</td>\n",
       "      <td>-1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.160547e+07</td>\n",
       "      <td>37.443207</td>\n",
       "      <td>3.503908</td>\n",
       "      <td>0.482330</td>\n",
       "      <td>0.551023</td>\n",
       "      <td>1.968597</td>\n",
       "      <td>2.501522</td>\n",
       "      <td>4460.559091</td>\n",
       "      <td>1371.403125</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.269424</td>\n",
       "      <td>0.073731</td>\n",
       "      <td>0.816801</td>\n",
       "      <td>0.161935</td>\n",
       "      <td>1124.456466</td>\n",
       "      <td>0.041857</td>\n",
       "      <td>0.482330</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.093950e+07</td>\n",
       "      <td>40.145294</td>\n",
       "      <td>5.669799</td>\n",
       "      <td>3.565896</td>\n",
       "      <td>2.305596</td>\n",
       "      <td>10.666667</td>\n",
       "      <td>12.428571</td>\n",
       "      <td>37904.563380</td>\n",
       "      <td>8271.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.851300</td>\n",
       "      <td>0.139414</td>\n",
       "      <td>0.865750</td>\n",
       "      <td>0.304500</td>\n",
       "      <td>7057.999731</td>\n",
       "      <td>0.210830</td>\n",
       "      <td>3.565896</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Volume_stock  Adj_Close_stock  HiLo_vola_stock  Pct_change_stock  \\\n",
       "count  7.400000e+01        74.000000        74.000000         74.000000   \n",
       "mean   9.731227e+06        33.997063         2.866570         -0.447087   \n",
       "std    3.143877e+06         3.336924         0.999271          1.512450   \n",
       "min    5.018200e+06        30.006098         1.373308         -5.089362   \n",
       "25%    7.338450e+06        31.092009         2.060415         -1.211461   \n",
       "50%    9.032517e+06        32.221743         2.770293         -0.418915   \n",
       "75%    1.160547e+07        37.443207         3.503908          0.482330   \n",
       "max    2.093950e+07        40.145294         5.669799          3.565896   \n",
       "\n",
       "       Pct_change_stock_scaled       Favs        RTs     Followers  \\\n",
       "count                74.000000  74.000000  74.000000     74.000000   \n",
       "mean                  0.022178   1.873590   2.362778   4158.740790   \n",
       "std                   0.860596   1.489856   1.779835   6194.575281   \n",
       "min                  -2.619312   1.000000   1.000000     28.000000   \n",
       "25%                  -0.412757   1.000000   1.000000   1367.875962   \n",
       "50%                   0.038208   1.450000   2.305556   2583.105392   \n",
       "75%                   0.551023   1.968597   2.501522   4460.559091   \n",
       "max                   2.305596  10.666667  12.428571  37904.563380   \n",
       "\n",
       "         Following    Is a RT   compound        neg        neu        pos  \\\n",
       "count    74.000000  74.000000  74.000000  74.000000  74.000000  74.000000   \n",
       "mean   1047.781398   0.136907   0.218461   0.055512   0.796866   0.147624   \n",
       "std    1178.949550   0.143181   0.130938   0.027647   0.036900   0.040814   \n",
       "min      11.000000   0.000000  -0.042245   0.000000   0.695500   0.063667   \n",
       "25%     395.832358   0.034483   0.121496   0.038277   0.780541   0.126429   \n",
       "50%     723.038462   0.099408   0.214772   0.056847   0.797208   0.137407   \n",
       "75%    1371.403125   0.200000   0.269424   0.073731   0.816801   0.161935   \n",
       "max    8271.600000   0.600000   0.851300   0.139414   0.865750   0.304500   \n",
       "\n",
       "       Compound_multiplied  Compound_multiplied_scaled  \\\n",
       "count            74.000000                   74.000000   \n",
       "mean            442.440350                    0.022435   \n",
       "std            3204.139740                    0.091246   \n",
       "min          -24255.853656                   -0.680912   \n",
       "25%             124.810639                    0.013390   \n",
       "50%             407.177512                    0.021431   \n",
       "75%            1124.456466                    0.041857   \n",
       "max            7057.999731                    0.210830   \n",
       "\n",
       "       Predicted_change_stock   Buy/Sell  \n",
       "count               74.000000  74.000000  \n",
       "mean                -0.488144  -0.243243  \n",
       "std                  1.581633   0.976586  \n",
       "min                 -5.089362  -1.000000  \n",
       "25%                 -1.211461  -1.000000  \n",
       "50%                 -0.418915  -1.000000  \n",
       "75%                  0.482330   1.000000  \n",
       "max                  3.565896   1.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 74 entries, 2016-04-01 to 2016-06-13\n",
      "Data columns (total 18 columns):\n",
      "Volume_stock                  74 non-null float64\n",
      "Adj_Close_stock               74 non-null float64\n",
      "HiLo_vola_stock               74 non-null float64\n",
      "Pct_change_stock              74 non-null float64\n",
      "Pct_change_stock_scaled       74 non-null float64\n",
      "Favs                          74 non-null float64\n",
      "RTs                           74 non-null float64\n",
      "Followers                     74 non-null float64\n",
      "Following                     74 non-null float64\n",
      "Is a RT                       74 non-null float64\n",
      "compound                      74 non-null float64\n",
      "neg                           74 non-null float64\n",
      "neu                           74 non-null float64\n",
      "pos                           74 non-null float64\n",
      "Compound_multiplied           74 non-null float64\n",
      "Compound_multiplied_scaled    74 non-null float64\n",
      "Predicted_change_stock        74 non-null float64\n",
      "Buy/Sell                      74 non-null float64\n",
      "dtypes: float64(18)\n",
      "memory usage: 11.0 KB\n"
     ]
    }
   ],
   "source": [
    "all_data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### How are the 'Compound_multiplied_scaled' -features distributed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jangn\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\Default_Python3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6510: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7EAAAD8CAYAAAChDSN2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_size = plt.rcParams[\"figure.figsize\"]\n",
    "fig_size[0] =16.0\n",
    "fig_size[1] = 4.0\n",
    "x = all_data['Compound_multiplied_scaled'] \n",
    "plt.hist(x, normed=True, bins=250)\n",
    "plt.ylabel('Compound_multiplied_scaled'); \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Train - Test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "x = np.array(all_data[['Compound_multiplied_scaled']]) #\n",
    "y = np.array(all_data['Buy/Sell'])\n",
    "\n",
    "# from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.2, random_state = 42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "### KNN - K-Nearest-Neighbors\n",
    "Accuracy from running only 20% Testing data against 80% Training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.59999999999999998"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn.neighbors import KNeighborsClassifier\n",
    "# from sklearn.model_selection import cross_val_score\n",
    "\n",
    "neigh = KNeighborsClassifier(n_neighbors=5)\n",
    "neigh.fit(x_train, y_train) \n",
    "neigh.score( x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy after cross validation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.575238095238\n"
     ]
    }
   ],
   "source": [
    "neigh_cv = cross_val_score(neigh, x_train, y_train, cv=10) \n",
    "print(neigh_cv.mean()) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Logistic Regression\n",
    "Accuracy from running only 20% Testing data against 80% Training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.59999999999999998"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logreg = LogisticRegression(random_state=42)\n",
    "logreg.fit(x_train, y_train)\n",
    "logreg.score( x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy after cross validation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.627619047619\n"
     ]
    }
   ],
   "source": [
    "logreg_cv = cross_val_score(logreg, x_train, y_train, cv=10)\n",
    "print(logreg_cv.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Support Vector Machines (SVM) with linear kernel\n",
    "Accuracy from running only 20% Testing data against 80% Training data.\n",
    "#### kernel = 'linear'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.59999999999999998"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn.svm import SVC\n",
    "\n",
    "svm_linear = SVC( kernel = 'linear')\n",
    "svm_linear.fit(x_train, y_train)\n",
    "svm_linear.score(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy for 'linear' after cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.627619047619\n"
     ]
    }
   ],
   "source": [
    "svm_linear_cv = cross_val_score(svm_linear, x_train, y_train, cv=10)\n",
    "print(svm_linear_cv.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### kernel = 'rbf' - NOT USED"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# svm_rbf = SVC( kernel = 'rbf')\n",
    "# svm_rbf.fit(x_train, y_train) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy for 'rbf' after cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# svm_rbf_cv = cross_val_score(svm_rbf, x_train, y_train, cv=10) \n",
    "# print(svm_rbf_cv.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Naive Bayes\n",
    "\n",
    "Please note! A MinMaxScaler is needed to get the features in the range MultinomialNB requires. No 20/80 testing was done, only cross-validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# from sklearn.naive_bayes import MultinomialNB\n",
    "\n",
    "# from sklearn.preprocessing import MinMaxScaler\n",
    "scaler = MinMaxScaler() \n",
    "X_minmax = scaler.fit_transform(x_train)\n",
    "\n",
    "mnb = MultinomialNB()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy after cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.627619047619\n"
     ]
    }
   ],
   "source": [
    "mnb_cv = cross_val_score(mnb, X_minmax, y_train, cv=10) # uscaled data accuracy same;  6588046192259676\n",
    "print(mnb_cv.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Decision Tree\n",
    "Accuracy from running only 20% Testing data against 80% Training data. Probably overfitting, a typical problem with Decistion Trees."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.73333333333333328"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn import tree\n",
    "dtc = tree.DecisionTreeClassifier(random_state=42)\n",
    "dtc.fit(x_train, y_train) \n",
    "dtc.score(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Decision Tree as an image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = ['Buy/Sell']\n",
    "\n",
    "# import pydotplus\n",
    "# from sklearn.externals.six import StringIO  \n",
    "# from IPython.display import Image\n",
    "dot_data = StringIO()  \n",
    "tree.export_graphviz(dtc, out_file=dot_data,  \n",
    "                         feature_names=features)  \n",
    "graph = pydotplus.graph_from_dot_data(dot_data.getvalue())  \n",
    "Image(graph.create_png()) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy after cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.709523809524\n"
     ]
    }
   ],
   "source": [
    "dtc_cv = cross_val_score(dtc, x_train, y_train, cv=10) \n",
    "print(dtc_cv.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Random Forest\n",
    "Accuracy from running only 20% Testing data against 80% Training data. Apparenlty same overfitting problem as with the Decision Tree above. Probably due to the limited amount of data!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.73333333333333328"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn.ensemble import RandomForestClassifier \n",
    "\n",
    "forest_reg = RandomForestClassifier(random_state=42)\n",
    "forest_reg.fit(x_train, y_train)\n",
    "forest_reg.score(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Accuracy after cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.712857142857\n"
     ]
    }
   ],
   "source": [
    "forest_reg_cv = cross_val_score(forest_reg, x_train, y_train, cv=10) \n",
    "print(forest_reg_cv.mean())  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Results Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN: \t\t\t 0.575238095238\n",
      "Logistic Regression: \t 0.627619047619\n",
      "SVM linear: \t\t 0.627619047619\n",
      "Naive Bayes: \t\t 0.627619047619\n",
      "Decision Tree: \t\t 0.709523809524\n",
      "Random Forest: \t\t 0.712857142857\n"
     ]
    }
   ],
   "source": [
    "print(\"KNN: \\t\\t\\t\", neigh_cv.mean())\n",
    "print(\"Logistic Regression: \\t\", logreg_cv.mean())\n",
    "print(\"SVM linear: \\t\\t\", svm_linear_cv.mean())\n",
    "print(\"Naive Bayes: \\t\\t\", mnb_cv.mean())\n",
    "print(\"Decision Tree: \\t\\t\", dtc_cv.mean())\n",
    "print(\"Random Forest: \\t\\t\", forest_reg_cv.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "results = []\n",
    "cv = [neigh_cv.mean(), logreg_cv.mean(), svm_linear_cv.mean(),mnb_cv.mean(),dtc_cv.mean(), forest_reg_cv.mean()]\n",
    "results.append(cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "results = {'0': ['KNN', 'LogReg', 'SVM linear', 'Naive Bayes', 'Decision Tree', 'Random Forest'],\n",
    "          '1':[neigh_cv.mean(), logreg_cv.mean(), svm_linear_cv.mean(), mnb_cv.mean(),dtc_cv.mean(),forest_reg_cv.mean()]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "summary = pd.DataFrame.from_dict(results)\n",
    "summary = summary.transpose()\n",
    "summary = summary.rename(index = {'0':'Model', '1':'AAL'}) # Update cashtag!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Model</th>\n",
       "      <td>KNN</td>\n",
       "      <td>LogReg</td>\n",
       "      <td>SVM linear</td>\n",
       "      <td>Naive Bayes</td>\n",
       "      <td>Decision Tree</td>\n",
       "      <td>Random Forest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAL</th>\n",
       "      <td>0.575238</td>\n",
       "      <td>0.627619</td>\n",
       "      <td>0.627619</td>\n",
       "      <td>0.627619</td>\n",
       "      <td>0.709524</td>\n",
       "      <td>0.712857</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              0         1           2            3              4  \\\n",
       "Model       KNN    LogReg  SVM linear  Naive Bayes  Decision Tree   \n",
       "AAL    0.575238  0.627619    0.627619     0.627619       0.709524   \n",
       "\n",
       "                   5  \n",
       "Model  Random Forest  \n",
       "AAL         0.712857  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Save the result from each classifier for this cashtag  in an excel\n",
    "The results are used for creating an overall summary for each cashtag and classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# import xlsxwriter\n",
    "# import os\n",
    "\n",
    "writer_df = pd.ExcelWriter('summary.xlsx', engine='xlsxwriter')\n",
    "summary.to_excel(writer_df)\n",
    "writer_df.save()\n",
    "\n",
    "os.rename('summary.xlsx', 'AAL_summary.xlsx') # Update cashtag!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}