FinalProject.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import figure\n",
    "import numpy as np\n",
    "import datetime\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DATA PREPARATION"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### USDX Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_usdx = pd.read_csv(\"./data/USDX-original.csv\", delimiter=',')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_usdx.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def usdx_analysis(usdx, threshold=0.2):\n",
    "    \"\"\"\n",
    "    USDX check by threshold.\n",
    "    :param usdx:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    if float(str(usdx).strip(\"%\")) >= threshold:\n",
    "        return 1\n",
    "    elif float(str(usdx).strip(\"%\")) <= -threshold:\n",
    "        return -1\n",
    "    else:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "for index, row in df_usdx.iterrows():\n",
    "    df_usdx.at[index,'Result'] = usdx_analysis(row['Change %'], 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Price</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Vol.</th>\n",
       "      <th>Change %</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Nov 04, 2019</td>\n",
       "      <td>97.070</td>\n",
       "      <td>97.060</td>\n",
       "      <td>97.155</td>\n",
       "      <td>96.977</td>\n",
       "      <td>-</td>\n",
       "      <td>0.03%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Nov 01, 2019</td>\n",
       "      <td>97.041</td>\n",
       "      <td>97.110</td>\n",
       "      <td>97.265</td>\n",
       "      <td>96.960</td>\n",
       "      <td>14.89K</td>\n",
       "      <td>-0.12%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Oct 31, 2019</td>\n",
       "      <td>97.154</td>\n",
       "      <td>97.255</td>\n",
       "      <td>97.255</td>\n",
       "      <td>97.015</td>\n",
       "      <td>17.25K</td>\n",
       "      <td>-0.29%</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Oct 30, 2019</td>\n",
       "      <td>97.439</td>\n",
       "      <td>97.460</td>\n",
       "      <td>97.800</td>\n",
       "      <td>97.230</td>\n",
       "      <td>16.51K</td>\n",
       "      <td>-0.03%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Oct 29, 2019</td>\n",
       "      <td>97.469</td>\n",
       "      <td>97.525</td>\n",
       "      <td>97.700</td>\n",
       "      <td>97.380</td>\n",
       "      <td>10.43K</td>\n",
       "      <td>-0.06%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date   Price    Open    High     Low    Vol. Change %  Result\n",
       "0  Nov 04, 2019  97.070  97.060  97.155  96.977       -    0.03%     0.0\n",
       "1  Nov 01, 2019  97.041  97.110  97.265  96.960  14.89K   -0.12%     0.0\n",
       "2  Oct 31, 2019  97.154  97.255  97.255  97.015  17.25K   -0.29%    -1.0\n",
       "3  Oct 30, 2019  97.439  97.460  97.800  97.230  16.51K   -0.03%     0.0\n",
       "4  Oct 29, 2019  97.469  97.525  97.700  97.380  10.43K   -0.06%     0.0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0.0    328\n",
       "-1.0    199\n",
       " 1.0    193\n",
       "Name: Result, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx['Result'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_usdx.drop([\"Price\", \"Open\", \"High\", \"Low\", \"Vol.\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>Change %</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Nov 04, 2019</td>\n",
       "      <td>0.03%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Nov 01, 2019</td>\n",
       "      <td>-0.12%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Oct 31, 2019</td>\n",
       "      <td>-0.29%</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Oct 30, 2019</td>\n",
       "      <td>-0.03%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Oct 29, 2019</td>\n",
       "      <td>-0.06%</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>715</td>\n",
       "      <td>Jan 26, 2017</td>\n",
       "      <td>0.36%</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>716</td>\n",
       "      <td>Jan 25, 2017</td>\n",
       "      <td>-0.32%</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>717</td>\n",
       "      <td>Jan 24, 2017</td>\n",
       "      <td>0.21%</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>718</td>\n",
       "      <td>Jan 23, 2017</td>\n",
       "      <td>-0.56%</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>719</td>\n",
       "      <td>Jan 20, 2017</td>\n",
       "      <td>-0.45%</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>720 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Date Change %  Result\n",
       "0    Nov 04, 2019    0.03%     0.0\n",
       "1    Nov 01, 2019   -0.12%     0.0\n",
       "2    Oct 31, 2019   -0.29%    -1.0\n",
       "3    Oct 30, 2019   -0.03%     0.0\n",
       "4    Oct 29, 2019   -0.06%     0.0\n",
       "..            ...      ...     ...\n",
       "715  Jan 26, 2017    0.36%     1.0\n",
       "716  Jan 25, 2017   -0.32%    -1.0\n",
       "717  Jan 24, 2017    0.21%     1.0\n",
       "718  Jan 23, 2017   -0.56%    -1.0\n",
       "719  Jan 20, 2017   -0.45%    -1.0\n",
       "\n",
       "[720 rows x 3 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_usdx.drop([\"Change %\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Nov 04, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Nov 01, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Oct 31, 2019</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Oct 30, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Oct 29, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>715</td>\n",
       "      <td>Jan 26, 2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>716</td>\n",
       "      <td>Jan 25, 2017</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>717</td>\n",
       "      <td>Jan 24, 2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>718</td>\n",
       "      <td>Jan 23, 2017</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>719</td>\n",
       "      <td>Jan 20, 2017</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>720 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Date  Result\n",
       "0    Nov 04, 2019     0.0\n",
       "1    Nov 01, 2019     0.0\n",
       "2    Oct 31, 2019    -1.0\n",
       "3    Oct 30, 2019     0.0\n",
       "4    Oct 29, 2019     0.0\n",
       "..            ...     ...\n",
       "715  Jan 26, 2017     1.0\n",
       "716  Jan 25, 2017    -1.0\n",
       "717  Jan 24, 2017     1.0\n",
       "718  Jan 23, 2017    -1.0\n",
       "719  Jan 20, 2017    -1.0\n",
       "\n",
       "[720 rows x 2 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tweet Data"
   ]
  },
  {
   "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>source</th>\n",
       "      <th>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>is_retweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>count</td>\n",
       "      <td>11859</td>\n",
       "      <td>11859</td>\n",
       "      <td>11857</td>\n",
       "      <td>11857</td>\n",
       "      <td>11857</td>\n",
       "      <td>11799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unique</td>\n",
       "      <td>8</td>\n",
       "      <td>11749</td>\n",
       "      <td>11545</td>\n",
       "      <td>9879</td>\n",
       "      <td>8832</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>top</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>MAKE AMERICA GREAT AGAIN!</td>\n",
       "      <td>07-15-2018 13:33:15</td>\n",
       "      <td>16906</td>\n",
       "      <td>0</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>freq</td>\n",
       "      <td>11287</td>\n",
       "      <td>21</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2683</td>\n",
       "      <td>9111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    source                       text           created_at  \\\n",
       "count                11859                      11859                11857   \n",
       "unique                   8                      11749                11545   \n",
       "top     Twitter for iPhone  MAKE AMERICA GREAT AGAIN!  07-15-2018 13:33:15   \n",
       "freq                 11287                         21                    5   \n",
       "\n",
       "       retweet_count favorite_count is_retweet  \n",
       "count          11857          11857      11799  \n",
       "unique          9879           8832          9  \n",
       "top            16906              0      FALSE  \n",
       "freq               5           2683       9111  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tweet = pd.read_csv(\"./data/trump_tweets.csv\", delimiter=',')\n",
    "df_tweet.head()\n",
    "df = df_tweet.drop(columns=['Unnamed: 6', 'Unnamed: 7', 'Unnamed: 8', 'Unnamed: 9', 'Unnamed: 10', 'Unnamed: 11', 'Unnamed: 12', 'Unnamed: 13'])\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>source</th>\n",
       "      <th>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>is_retweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Twitter Media Studio</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>10-31-2019 22:44:26</td>\n",
       "      <td>30681</td>\n",
       "      <td>103267</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>10-31-2019 15:31:46</td>\n",
       "      <td>37849</td>\n",
       "      <td>181124</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>10-31-2019 14:57:30</td>\n",
       "      <td>24450</td>\n",
       "      <td>105620</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>10-31-2019 14:37:45</td>\n",
       "      <td>11956</td>\n",
       "      <td>47363</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>10-31-2019 14:37:39</td>\n",
       "      <td>14896</td>\n",
       "      <td>59982</td>\n",
       "      <td>FALSE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  source                                               text  \\\n",
       "0   Twitter Media Studio  While the Do Nothing Democrats FAIL the Americ...   \n",
       "15    Twitter for iPhone       The Greatest Witch Hunt In American History!   \n",
       "16    Twitter for iPhone  The Impeachment Hoax is hurting our Stock Mark...   \n",
       "17    Twitter for iPhone  ....our manufacturers. We should have lower in...   \n",
       "18    Twitter for iPhone  People are VERY disappointed in Jay Powell and...   \n",
       "\n",
       "             created_at retweet_count favorite_count is_retweet  \n",
       "0   10-31-2019 22:44:26         30681         103267      FALSE  \n",
       "15  10-31-2019 15:31:46         37849         181124      FALSE  \n",
       "16  10-31-2019 14:57:30         24450         105620      FALSE  \n",
       "17  10-31-2019 14:37:45         11956          47363      FALSE  \n",
       "18  10-31-2019 14:37:39         14896          59982      FALSE  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_nonretweet = df[df['is_retweet'] == 'FALSE']\n",
    "df_nonretweet.describe()\n",
    "df_nonretweet.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "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>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>10-31-2019 22:44:26</td>\n",
       "      <td>30681</td>\n",
       "      <td>103267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>10-31-2019 15:31:46</td>\n",
       "      <td>37849</td>\n",
       "      <td>181124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>10-31-2019 14:57:30</td>\n",
       "      <td>24450</td>\n",
       "      <td>105620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>10-31-2019 14:37:45</td>\n",
       "      <td>11956</td>\n",
       "      <td>47363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>10-31-2019 14:37:39</td>\n",
       "      <td>14896</td>\n",
       "      <td>59982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11854</td>\n",
       "      <td>power from Washington D.C. and giving it back ...</td>\n",
       "      <td>01-20-2017 17:51:58</td>\n",
       "      <td>18362</td>\n",
       "      <td>91143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11855</td>\n",
       "      <td>Today we are not merely transferring power fro...</td>\n",
       "      <td>01-20-2017 17:51:25</td>\n",
       "      <td>20125</td>\n",
       "      <td>109640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11856</td>\n",
       "      <td>It all begins today! I will see you at 11:00 A...</td>\n",
       "      <td>01-20-2017 12:31:53</td>\n",
       "      <td>70523</td>\n",
       "      <td>268372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11857</td>\n",
       "      <td>Thank you for a wonderful evening in Washingto...</td>\n",
       "      <td>01-20-2017 4:24:33</td>\n",
       "      <td>20770</td>\n",
       "      <td>112829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11858</td>\n",
       "      <td>Thank you for joining us at the Lincoln Memori...</td>\n",
       "      <td>01-20-2017 0:40:51</td>\n",
       "      <td>34204</td>\n",
       "      <td>165943</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9111 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text           created_at  \\\n",
       "0      While the Do Nothing Democrats FAIL the Americ...  10-31-2019 22:44:26   \n",
       "15          The Greatest Witch Hunt In American History!  10-31-2019 15:31:46   \n",
       "16     The Impeachment Hoax is hurting our Stock Mark...  10-31-2019 14:57:30   \n",
       "17     ....our manufacturers. We should have lower in...  10-31-2019 14:37:45   \n",
       "18     People are VERY disappointed in Jay Powell and...  10-31-2019 14:37:39   \n",
       "...                                                  ...                  ...   \n",
       "11854  power from Washington D.C. and giving it back ...  01-20-2017 17:51:58   \n",
       "11855  Today we are not merely transferring power fro...  01-20-2017 17:51:25   \n",
       "11856  It all begins today! I will see you at 11:00 A...  01-20-2017 12:31:53   \n",
       "11857  Thank you for a wonderful evening in Washingto...   01-20-2017 4:24:33   \n",
       "11858  Thank you for joining us at the Lincoln Memori...   01-20-2017 0:40:51   \n",
       "\n",
       "      retweet_count favorite_count  \n",
       "0             30681         103267  \n",
       "15            37849         181124  \n",
       "16            24450         105620  \n",
       "17            11956          47363  \n",
       "18            14896          59982  \n",
       "...             ...            ...  \n",
       "11854         18362          91143  \n",
       "11855         20125         109640  \n",
       "11856         70523         268372  \n",
       "11857         20770         112829  \n",
       "11858         34204         165943  \n",
       "\n",
       "[9111 rows x 4 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler = df_nonretweet.drop(columns=['source', 'is_retweet'])\n",
    "df_simpler2 = df_simpler\n",
    "df_simpler2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>fav_rt_ratio</th>\n",
       "      <th>rt_fav_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>10-31-2019 22:44:26</td>\n",
       "      <td>30681</td>\n",
       "      <td>103267</td>\n",
       "      <td>3.365829</td>\n",
       "      <td>0.297104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>10-31-2019 15:31:46</td>\n",
       "      <td>37849</td>\n",
       "      <td>181124</td>\n",
       "      <td>4.785437</td>\n",
       "      <td>0.208967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>10-31-2019 14:57:30</td>\n",
       "      <td>24450</td>\n",
       "      <td>105620</td>\n",
       "      <td>4.319836</td>\n",
       "      <td>0.231490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>10-31-2019 14:37:45</td>\n",
       "      <td>11956</td>\n",
       "      <td>47363</td>\n",
       "      <td>3.961442</td>\n",
       "      <td>0.252433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>10-31-2019 14:37:39</td>\n",
       "      <td>14896</td>\n",
       "      <td>59982</td>\n",
       "      <td>4.026719</td>\n",
       "      <td>0.248341</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text           created_at  \\\n",
       "0   While the Do Nothing Democrats FAIL the Americ...  10-31-2019 22:44:26   \n",
       "15       The Greatest Witch Hunt In American History!  10-31-2019 15:31:46   \n",
       "16  The Impeachment Hoax is hurting our Stock Mark...  10-31-2019 14:57:30   \n",
       "17  ....our manufacturers. We should have lower in...  10-31-2019 14:37:45   \n",
       "18  People are VERY disappointed in Jay Powell and...  10-31-2019 14:37:39   \n",
       "\n",
       "    retweet_count  favorite_count  fav_rt_ratio  rt_fav_ratio  \n",
       "0           30681          103267      3.365829      0.297104  \n",
       "15          37849          181124      4.785437      0.208967  \n",
       "16          24450          105620      4.319836      0.231490  \n",
       "17          11956           47363      3.961442      0.252433  \n",
       "18          14896           59982      4.026719      0.248341  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2['retweet_count'] = pd.to_numeric(df_simpler2['retweet_count'])\n",
    "df_simpler2['favorite_count'] = pd.to_numeric(df_simpler2['favorite_count'])\n",
    "for index, row in df_simpler2.iterrows():\n",
    "    if df_simpler2.at[index,'favorite_count'] == 0:\n",
    "        df_simpler2.at[index,'favorite_count'] = 1\n",
    "    if df_simpler2.at[index,'retweet_count'] == 0:\n",
    "        df_simpler2.at[index,'retweet_count'] = 1\n",
    "        \n",
    "df_simpler2['fav_rt_ratio'] = df_simpler2['favorite_count'].div(df_simpler['retweet_count'])\n",
    "df_simpler2['rt_fav_ratio'] = df_simpler2['retweet_count'].div(df_simpler['favorite_count'])\n",
    "\n",
    "df_simpler2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "text              9111\n",
       "created_at        9111\n",
       "retweet_count     9111\n",
       "favorite_count    9111\n",
       "fav_rt_ratio      9111\n",
       "rt_fav_ratio      9111\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def is_weekend(date):\n",
    "    \"\"\"\n",
    "    If date is weekend return 1 else return 0\n",
    "    :param date:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    dt = datetime.datetime.strptime(date, \"%m-%d-%Y %H:%M:%S\").weekday()\n",
    "    if dt < 5:\n",
    "        return 0\n",
    "    return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>fav_rt_ratio</th>\n",
       "      <th>rt_fav_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>10-31-2019 22:44:26</td>\n",
       "      <td>30681</td>\n",
       "      <td>103267</td>\n",
       "      <td>3.365829</td>\n",
       "      <td>0.297104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>10-31-2019 15:31:46</td>\n",
       "      <td>37849</td>\n",
       "      <td>181124</td>\n",
       "      <td>4.785437</td>\n",
       "      <td>0.208967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>10-31-2019 14:57:30</td>\n",
       "      <td>24450</td>\n",
       "      <td>105620</td>\n",
       "      <td>4.319836</td>\n",
       "      <td>0.231490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>10-31-2019 14:37:45</td>\n",
       "      <td>11956</td>\n",
       "      <td>47363</td>\n",
       "      <td>3.961442</td>\n",
       "      <td>0.252433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>10-31-2019 14:37:39</td>\n",
       "      <td>14896</td>\n",
       "      <td>59982</td>\n",
       "      <td>4.026719</td>\n",
       "      <td>0.248341</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text           created_at  \\\n",
       "0   While the Do Nothing Democrats FAIL the Americ...  10-31-2019 22:44:26   \n",
       "15       The Greatest Witch Hunt In American History!  10-31-2019 15:31:46   \n",
       "16  The Impeachment Hoax is hurting our Stock Mark...  10-31-2019 14:57:30   \n",
       "17  ....our manufacturers. We should have lower in...  10-31-2019 14:37:45   \n",
       "18  People are VERY disappointed in Jay Powell and...  10-31-2019 14:37:39   \n",
       "\n",
       "    retweet_count  favorite_count  fav_rt_ratio  rt_fav_ratio  \n",
       "0           30681          103267      3.365829      0.297104  \n",
       "15          37849          181124      4.785437      0.208967  \n",
       "16          24450          105620      4.319836      0.231490  \n",
       "17          11956           47363      3.961442      0.252433  \n",
       "18          14896           59982      4.026719      0.248341  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for index, row in df_simpler2.iterrows():\n",
    "    if is_weekend(str(row['created_at'])):\n",
    "        df_simpler2.drop(index, inplace=True)\n",
    "\n",
    "df_simpler2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "text              6876\n",
       "created_at        6876\n",
       "retweet_count     6876\n",
       "favorite_count    6876\n",
       "fav_rt_ratio      6876\n",
       "rt_fav_ratio      6876\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_simpler2['created_at'] = df_simpler2['created_at'].str.split(\" \", n = 1, expand = True)[0]\n",
    "df_simpler2['created_at'] = pd.to_datetime(df_simpler2['created_at'], format='%m-%d-%Y')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\John\\Anaconda3\\lib\\site-packages\\pandas\\plotting\\_matplotlib\\converter.py:103: FutureWarning: Using an implicitly registered datetime converter for a matplotlib plotting method. The converter was registered by pandas on import. Future versions of pandas will require you to explicitly register matplotlib converters.\n",
      "\n",
      "To register the converters:\n",
      "\t>>> from pandas.plotting import register_matplotlib_converters\n",
      "\t>>> register_matplotlib_converters()\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1600x1200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure(num=None, figsize=(20, 15), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.plot(df_simpler2['created_at'], df_simpler2['fav_rt_ratio'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    6876.000000\n",
       "mean        4.379187\n",
       "std         0.764742\n",
       "min         0.000242\n",
       "25%         3.910116\n",
       "50%         4.305482\n",
       "75%         4.772913\n",
       "max        21.029834\n",
       "Name: fav_rt_ratio, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2['fav_rt_ratio'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    6876.000000\n",
       "mean        1.406468\n",
       "std        68.667872\n",
       "min         0.047551\n",
       "25%         0.209516\n",
       "50%         0.232262\n",
       "75%         0.255747\n",
       "max      4129.000000\n",
       "Name: rt_fav_ratio, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2['rt_fav_ratio'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rt_fav_threshold(ratio, threshold=1):\n",
    "    \"\"\"\n",
    "    rt-fav ratio check by threshold.\n",
    "    :param ratio, threshold:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    if float(ratio) >= threshold:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "for index, row in df_simpler2.iterrows():\n",
    "    df_simpler2.at[index,'rt_fav'] = rt_fav_threshold(row['rt_fav_ratio'], 0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>fav_rt_ratio</th>\n",
       "      <th>rt_fav_ratio</th>\n",
       "      <th>rt_fav</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>30681</td>\n",
       "      <td>103267</td>\n",
       "      <td>3.365829</td>\n",
       "      <td>0.297104</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>37849</td>\n",
       "      <td>181124</td>\n",
       "      <td>4.785437</td>\n",
       "      <td>0.208967</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>24450</td>\n",
       "      <td>105620</td>\n",
       "      <td>4.319836</td>\n",
       "      <td>0.231490</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>11956</td>\n",
       "      <td>47363</td>\n",
       "      <td>3.961442</td>\n",
       "      <td>0.252433</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>14896</td>\n",
       "      <td>59982</td>\n",
       "      <td>4.026719</td>\n",
       "      <td>0.248341</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text created_at  \\\n",
       "0   While the Do Nothing Democrats FAIL the Americ... 2019-10-31   \n",
       "15       The Greatest Witch Hunt In American History! 2019-10-31   \n",
       "16  The Impeachment Hoax is hurting our Stock Mark... 2019-10-31   \n",
       "17  ....our manufacturers. We should have lower in... 2019-10-31   \n",
       "18  People are VERY disappointed in Jay Powell and... 2019-10-31   \n",
       "\n",
       "    retweet_count  favorite_count  fav_rt_ratio  rt_fav_ratio  rt_fav  \n",
       "0           30681          103267      3.365829      0.297104     1.0  \n",
       "15          37849          181124      4.785437      0.208967     0.0  \n",
       "16          24450          105620      4.319836      0.231490     0.0  \n",
       "17          11956           47363      3.961442      0.252433     1.0  \n",
       "18          14896           59982      4.026719      0.248341     0.0  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    4815\n",
       "1.0    2061\n",
       "Name: rt_fav, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2[\"rt_fav\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>rt_fav</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text created_at  rt_fav\n",
       "0   While the Do Nothing Democrats FAIL the Americ... 2019-10-31     1.0\n",
       "15       The Greatest Witch Hunt In American History! 2019-10-31     0.0\n",
       "16  The Impeachment Hoax is hurting our Stock Mark... 2019-10-31     0.0\n",
       "17  ....our manufacturers. We should have lower in... 2019-10-31     1.0\n",
       "18  People are VERY disappointed in Jay Powell and... 2019-10-31     0.0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_simpler2.drop(['retweet_count','favorite_count','fav_rt_ratio','rt_fav_ratio'], axis=1, inplace=True)\n",
    "df_simpler2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Merge DataSets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Nov 04, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Nov 01, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Oct 31, 2019</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Oct 30, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Oct 29, 2019</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date  Result\n",
       "0  Nov 04, 2019     0.0\n",
       "1  Nov 01, 2019     0.0\n",
       "2  Oct 31, 2019    -1.0\n",
       "3  Oct 30, 2019     0.0\n",
       "4  Oct 29, 2019     0.0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2019-11-01</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2019-10-30</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>2019-10-29</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Date  Result\n",
       "0 2019-11-04     0.0\n",
       "1 2019-11-01     0.0\n",
       "2 2019-10-31    -1.0\n",
       "3 2019-10-30     0.0\n",
       "4 2019-10-29     0.0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_usdx['Date'] = pd.to_datetime(df_usdx['Date'], format='%b %d, %Y')\n",
    "df_usdx.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = ['date']\n",
    "df_simpler2.rename(columns={\"created_at\": \"date\"}, inplace=True)\n",
    "df_usdx.rename(columns={\"Date\":\"date\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_merged = df_simpler2.join(df_usdx.set_index(cols), on=cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>text</th>\n",
       "      <th>date</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11854</td>\n",
       "      <td>power from Washington D.C. and giving it back ...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11855</td>\n",
       "      <td>Today we are not merely transferring power fro...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11856</td>\n",
       "      <td>It all begins today! I will see you at 11:00 A...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11857</td>\n",
       "      <td>Thank you for a wonderful evening in Washingto...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11858</td>\n",
       "      <td>Thank you for joining us at the Lincoln Memori...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6876 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text       date  rt_fav  \\\n",
       "0      While the Do Nothing Democrats FAIL the Americ... 2019-10-31     1.0   \n",
       "15          The Greatest Witch Hunt In American History! 2019-10-31     0.0   \n",
       "16     The Impeachment Hoax is hurting our Stock Mark... 2019-10-31     0.0   \n",
       "17     ....our manufacturers. We should have lower in... 2019-10-31     1.0   \n",
       "18     People are VERY disappointed in Jay Powell and... 2019-10-31     0.0   \n",
       "...                                                  ...        ...     ...   \n",
       "11854  power from Washington D.C. and giving it back ... 2017-01-20     0.0   \n",
       "11855  Today we are not merely transferring power fro... 2017-01-20     0.0   \n",
       "11856  It all begins today! I will see you at 11:00 A... 2017-01-20     1.0   \n",
       "11857  Thank you for a wonderful evening in Washingto... 2017-01-20     0.0   \n",
       "11858  Thank you for joining us at the Lincoln Memori... 2017-01-20     0.0   \n",
       "\n",
       "       Result  \n",
       "0        -1.0  \n",
       "15       -1.0  \n",
       "16       -1.0  \n",
       "17       -1.0  \n",
       "18       -1.0  \n",
       "...       ...  \n",
       "11854    -1.0  \n",
       "11855    -1.0  \n",
       "11856    -1.0  \n",
       "11857    -1.0  \n",
       "11858    -1.0  \n",
       "\n",
       "[6876 rows x 4 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_merged"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_merged2 = df_merged.dropna(axis='rows')\n",
    "df_merged2 = df_merged2.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>text</th>\n",
       "      <th>date</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6838</td>\n",
       "      <td>power from Washington D.C. and giving it back ...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6839</td>\n",
       "      <td>Today we are not merely transferring power fro...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6840</td>\n",
       "      <td>It all begins today! I will see you at 11:00 A...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6841</td>\n",
       "      <td>Thank you for a wonderful evening in Washingto...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6842</td>\n",
       "      <td>Thank you for joining us at the Lincoln Memori...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6843 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   text       date  rt_fav  \\\n",
       "0     While the Do Nothing Democrats FAIL the Americ... 2019-10-31     1.0   \n",
       "1          The Greatest Witch Hunt In American History! 2019-10-31     0.0   \n",
       "2     The Impeachment Hoax is hurting our Stock Mark... 2019-10-31     0.0   \n",
       "3     ....our manufacturers. We should have lower in... 2019-10-31     1.0   \n",
       "4     People are VERY disappointed in Jay Powell and... 2019-10-31     0.0   \n",
       "...                                                 ...        ...     ...   \n",
       "6838  power from Washington D.C. and giving it back ... 2017-01-20     0.0   \n",
       "6839  Today we are not merely transferring power fro... 2017-01-20     0.0   \n",
       "6840  It all begins today! I will see you at 11:00 A... 2017-01-20     1.0   \n",
       "6841  Thank you for a wonderful evening in Washingto... 2017-01-20     0.0   \n",
       "6842  Thank you for joining us at the Lincoln Memori... 2017-01-20     0.0   \n",
       "\n",
       "      Result  \n",
       "0       -1.0  \n",
       "1       -1.0  \n",
       "2       -1.0  \n",
       "3       -1.0  \n",
       "4       -1.0  \n",
       "...      ...  \n",
       "6838    -1.0  \n",
       "6839    -1.0  \n",
       "6840    -1.0  \n",
       "6841    -1.0  \n",
       "6842    -1.0  \n",
       "\n",
       "[6843 rows x 4 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_merged2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>text</th>\n",
       "      <th>date</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>While the Do Nothing Democrats FAIL the Americ...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>The Greatest Witch Hunt In American History!</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>The Impeachment Hoax is hurting our Stock Mark...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>....our manufacturers. We should have lower in...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>People are VERY disappointed in Jay Powell and...</td>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6838</td>\n",
       "      <td>power from Washington D.C. and giving it back ...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6839</td>\n",
       "      <td>Today we are not merely transferring power fro...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6840</td>\n",
       "      <td>It all begins today! I will see you at 11:00 A...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6841</td>\n",
       "      <td>Thank you for a wonderful evening in Washingto...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6842</td>\n",
       "      <td>Thank you for joining us at the Lincoln Memori...</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6843 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   text       date  rt_fav  \\\n",
       "0     While the Do Nothing Democrats FAIL the Americ... 2019-10-31     1.0   \n",
       "1          The Greatest Witch Hunt In American History! 2019-10-31     0.0   \n",
       "2     The Impeachment Hoax is hurting our Stock Mark... 2019-10-31     0.0   \n",
       "3     ....our manufacturers. We should have lower in... 2019-10-31     1.0   \n",
       "4     People are VERY disappointed in Jay Powell and... 2019-10-31     0.0   \n",
       "...                                                 ...        ...     ...   \n",
       "6838  power from Washington D.C. and giving it back ... 2017-01-20     0.0   \n",
       "6839  Today we are not merely transferring power fro... 2017-01-20     0.0   \n",
       "6840  It all begins today! I will see you at 11:00 A... 2017-01-20     1.0   \n",
       "6841  Thank you for a wonderful evening in Washingto... 2017-01-20     0.0   \n",
       "6842  Thank you for joining us at the Lincoln Memori... 2017-01-20     0.0   \n",
       "\n",
       "      Result  \n",
       "0       -1.0  \n",
       "1       -1.0  \n",
       "2       -1.0  \n",
       "3       -1.0  \n",
       "4       -1.0  \n",
       "...      ...  \n",
       "6838    -1.0  \n",
       "6839    -1.0  \n",
       "6840    -1.0  \n",
       "6841    -1.0  \n",
       "6842    -1.0  \n",
       "\n",
       "[6843 rows x 4 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_merged2['Result'] = pd.to_numeric(df_merged2['Result'])\n",
    "df_merged2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "STOP_WORD_LIST = [ \"i\", \"me\", \"my\", \"myself\", \"we\", \"our\", \"ours\", \"ourselves\", \"you\", \"you're\", \"youre\", \"you re\", \"you've\", \"youve\", \"you ve\", \"you'll\", \"youll\", \"you ll\", \"youd\", \"you'd\", \"you d\", \"your\", \"yours\", \"yourself\", \"yourselves\", \"he\", \"him\", \"his\", \"himself\", \"she\", \"she s\", \"she's\", \"her\", \"hers\", \"herself\", \"it\", \"it s\", \"it's\", \"its\", \"itself\", \"they\", \"them\", \"their\", \"theirs\", \"themselves\", \"what\", \"which\", \"who\", \"whom\", \"this\", \"that\", \"that'll\", \"thatll\", \"that ll\", \"these\", \"those\", \"am\", \"is\", \"are\", \"was\", \"were\", \"be\", \"been\", \"being\", \"have\", \"has\", \"had\", \"having\", \"do\", \"does\", \"did\", \"doing\", \"a\", \"an\", \"the\", \"and\", \"but\", \"if\", \"or\", \"because\", \"as\", \"until\", \"while\", \"of\", \"at\", \"by\", \"for\", \"with\", \"about\", \"against\", \"between\", \"into\", \"through\", \"during\", \"before\", \"after\", \"above\", \"below\", \"to\", \"from\", \"up\", \"down\", \"in\", \"out\", \"on\", \"off\", \"over\", \"under\", \"again\", \"further\", \"then\", \"once\", \"here\", \"there\", \"when\", \"where\", \"why\", \"how\", \"all\", \"any\", \"both\", \"each\", \"few\", \"more\", \"most\", \"other\", \"some\", \"such\", \"no\", \"nor\", \"not\", \"only\", \"own\", \"same\", \"so\", \"than\", \"too\", \"very\", \"s\", \"t\", \"can\", \"will\", \"just\", \"don\", \"dont\", \"don't\", \"don t\", \"should\", \"shouldve\", \"should ve\", \"should've\", \"now\", \"d\", \"ll\", \"m\", \"o\", \"re\", \"ve\", \"y\", \"ain\", \"aren\", \"arent\", \"aren t\", \"aren't\", \"couldn\", \"couldnt\", \"couldn t\", \"couldn't\", \"didn\", \"didnt\", \"didn t\", \"didn't\", \"doesn\", \"doesnt\", \"doesn t\", \"doesn't\", \"hadn\", \"hadnt\", \"hadn t\", \"hadn't\", \"hasn\", \"hasnt\", \"hasn t\", \"hasn't\", \"haven\", \"havent\", \"haven t\", \"haven't\", \"isn\", \"isnt\", \"isn t\", \"isn't\", \"ma\", \"mightn\", \"mightnt\", \"mightn t\", \"mightn't\", \"mustn\", \"mustnt\", \"mustn t\", \"mustn't\", \"needn\", \"neednt\", \"needn t\", \"needn't\", \"shan\", \"shant\", \"shan t\", \"shan't\", \"shouldn\", \"shouldnt\", \"shouldn t\", \"shouldn't\", \"wasn\", \"wasnt\", \"wasn t\", \"wasn't\", \"werent\", \"weren t\", \"weren\", \"weren't\", \"won\", \"wont\", \"won t\", \"won't\", \"wouldn\", \"wouldnt\", \"wouldn t\", \"wouldn't\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Further changes on merged data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Generate the vocab\n",
    "vocab = []\n",
    "for line in df_merged2['text']:\n",
    "    words = line.split()\n",
    "    for word in words:\n",
    "        toAppend = word.lower().strip(')').strip('(').strip(',').strip('!').strip(':').strip('?').strip('”').strip('“').strip('’').strip('.').strip('!!').strip(')').strip('!')\n",
    "        if toAppend not in vocab:\n",
    "            if toAppend not in STOP_WORD_LIST:\n",
    "                if 'http' not in toAppend:\n",
    "                    if '...' not in toAppend:\n",
    "                        if len(toAppend) != 1:\n",
    "                            if '&amp' not in toAppend:\n",
    "                                vocab.append(toAppend)   \n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6843 11908\n"
     ]
    }
   ],
   "source": [
    "index_count = len(df_merged2['text'])\n",
    "col_count = len(vocab)\n",
    "print(index_count, col_count)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df_vector = pd.DataFrame(index=np.arange(6843), columns=np.arange(11908))\n",
    "df_vector2 = pd.DataFrame(index=np.arange(6843), columns = vocab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>mountain</th>\n",
       "      <th>american!#inaugurationday</th>\n",
       "      <th>rulers</th>\n",
       "      <th>controls</th>\n",
       "      <th>#inaugurationday</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6838</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6839</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6840</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6841</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6842</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6843 rows × 11908 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     nothing democrats fail american people continue impeachment scam  \\\n",
       "0        NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "1        NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "2        NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "3        NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "4        NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "...      ...       ...  ...      ...    ...      ...         ...  ...   \n",
       "6838     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "6839     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "6840     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "6841     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "6842     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "\n",
       "     administration deliver  ... mountain american!#inaugurationday rulers  \\\n",
       "0               NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "1               NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "2               NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "3               NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "4               NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "...             ...     ...  ...      ...                       ...    ...   \n",
       "6838            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "6839            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "6840            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "6841            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "6842            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "\n",
       "     controls #inaugurationday transferring swearing-in #inauguration  \\\n",
       "0         NaN              NaN          NaN         NaN           NaN   \n",
       "1         NaN              NaN          NaN         NaN           NaN   \n",
       "2         NaN              NaN          NaN         NaN           NaN   \n",
       "3         NaN              NaN          NaN         NaN           NaN   \n",
       "4         NaN              NaN          NaN         NaN           NaN   \n",
       "...       ...              ...          ...         ...           ...   \n",
       "6838      NaN              NaN          NaN         NaN           NaN   \n",
       "6839      NaN              NaN          NaN         NaN           NaN   \n",
       "6840      NaN              NaN          NaN         NaN           NaN   \n",
       "6841      NaN              NaN          NaN         NaN           NaN   \n",
       "6842      NaN              NaN          NaN         NaN           NaN   \n",
       "\n",
       "     tonight-  am…  \n",
       "0         NaN  NaN  \n",
       "1         NaN  NaN  \n",
       "2         NaN  NaN  \n",
       "3         NaN  NaN  \n",
       "4         NaN  NaN  \n",
       "...       ...  ...  \n",
       "6838      NaN  NaN  \n",
       "6839      NaN  NaN  \n",
       "6840      NaN  NaN  \n",
       "6841      NaN  NaN  \n",
       "6842      NaN  NaN  \n",
       "\n",
       "[6843 rows x 11908 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>mountain</th>\n",
       "      <th>american!#inaugurationday</th>\n",
       "      <th>rulers</th>\n",
       "      <th>controls</th>\n",
       "      <th>#inaugurationday</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 11908 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  nothing democrats fail american people continue impeachment scam  \\\n",
       "0     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "1     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "2     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "3     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "4     NaN       NaN  NaN      NaN    NaN      NaN         NaN  NaN   \n",
       "\n",
       "  administration deliver  ... mountain american!#inaugurationday rulers  \\\n",
       "0            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "1            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "2            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "3            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "4            NaN     NaN  ...      NaN                       NaN    NaN   \n",
       "\n",
       "  controls #inaugurationday transferring swearing-in #inauguration tonight-  \\\n",
       "0      NaN              NaN          NaN         NaN           NaN      NaN   \n",
       "1      NaN              NaN          NaN         NaN           NaN      NaN   \n",
       "2      NaN              NaN          NaN         NaN           NaN      NaN   \n",
       "3      NaN              NaN          NaN         NaN           NaN      NaN   \n",
       "4      NaN              NaN          NaN         NaN           NaN      NaN   \n",
       "\n",
       "   am…  \n",
       "0  NaN  \n",
       "1  NaN  \n",
       "2  NaN  \n",
       "3  NaN  \n",
       "4  NaN  \n",
       "\n",
       "[5 rows x 11908 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6842/6843\r"
     ]
    }
   ],
   "source": [
    "#for j in range(len(df_merged2['text'])):\n",
    "#    df_vector2 = df_vector2.append(pd.Series(), ignore_index=True)\n",
    "#    print(str(j) + \"/\" + str(len(df_merged2['text'])), end=\"\\r\", flush=True)\n",
    "i = 0\n",
    "df_vector2['usdx'] = 0\n",
    "df_vector2['rt_fav'] = 0\n",
    "for line in df_merged2['text']:\n",
    "    words = line.split()\n",
    "    for word in words:\n",
    "        toAppend = word.lower().strip(')').strip('(').strip(',').strip('!').strip(':').strip('?').strip('”').strip('“').strip('’').strip('.').strip('!!').strip(')').strip('!')\n",
    "        if toAppend in vocab:\n",
    "            df_vector2.iloc[i, df_vector2.columns.get_loc(toAppend)] = 1\n",
    "    df_vector2.iloc[i, df_vector2.columns.get_loc('usdx')] = df_merged2.iloc[i, df_merged2.columns.get_loc('Result')]\n",
    "    df_vector2.iloc[i, df_vector2.columns.get_loc('rt_fav')] = df_merged2.iloc[i, df_merged2.columns.get_loc('rt_fav')]\n",
    "    print(str(i) + \"/\" + str(len(df_merged2['text'])), end=\"\\r\", flush=True)\n",
    "    i += 1\n",
    "\n",
    "df_vector2 = df_vector2.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>rulers</th>\n",
       "      <th>controls</th>\n",
       "      <th>#inaugurationday</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "      <th>usdx</th>\n",
       "      <th>rt_fav</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6838</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6839</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6840</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6841</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6842</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6843 rows × 11910 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      nothing  democrats  fail  american  people  continue  impeachment  scam  \\\n",
       "0           1          1     1         1       1         1            1     1   \n",
       "1           0          0     0         1       0         0            0     0   \n",
       "2           1          1     0         0       0         0            1     0   \n",
       "3           0          0     0         0       0         0            0     0   \n",
       "4           0          0     0         0       1         0            0     0   \n",
       "...       ...        ...   ...       ...     ...       ...          ...   ...   \n",
       "6838        0          0     0         1       1         0            0     0   \n",
       "6839        0          0     0         0       0         0            0     0   \n",
       "6840        0          0     0         0       0         0            0     0   \n",
       "6841        0          0     0         0       0         0            0     0   \n",
       "6842        0          0     0         0       0         0            0     0   \n",
       "\n",
       "      administration  deliver  ...  rulers  controls  #inaugurationday  \\\n",
       "0                  1        1  ...       0         0                 0   \n",
       "1                  0        0  ...       0         0                 0   \n",
       "2                  0        0  ...       0         0                 0   \n",
       "3                  0        0  ...       0         0                 0   \n",
       "4                  0        0  ...       0         0                 0   \n",
       "...              ...      ...  ...     ...       ...               ...   \n",
       "6838               0        0  ...       0         0                 1   \n",
       "6839               1        0  ...       0         0                 0   \n",
       "6840               0        0  ...       0         0                 0   \n",
       "6841               0        0  ...       0         0                 0   \n",
       "6842               0        0  ...       0         0                 0   \n",
       "\n",
       "      transferring  swearing-in  #inauguration  tonight-  am…  usdx  rt_fav  \n",
       "0                0            0              0         0    0  -1.0     1.0  \n",
       "1                0            0              0         0    0  -1.0     0.0  \n",
       "2                0            0              0         0    0  -1.0     0.0  \n",
       "3                0            0              0         0    0  -1.0     1.0  \n",
       "4                0            0              0         0    0  -1.0     0.0  \n",
       "...            ...          ...            ...       ...  ...   ...     ...  \n",
       "6838             0            0              0         0    0  -1.0     0.0  \n",
       "6839             1            0              0         0    0  -1.0     0.0  \n",
       "6840             0            1              0         0    0  -1.0     1.0  \n",
       "6841             0            0              1         0    0  -1.0     0.0  \n",
       "6842             0            0              0         1    1  -1.0     0.0  \n",
       "\n",
       "[6843 rows x 11910 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "def is_url_in_tweet(tweet):\n",
    "    \"\"\"\n",
    "    If tweet contains http then return 1\n",
    "    :param tweet:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    tweet = str(tweet)\n",
    "    if \"http\" in tweet:\n",
    "        return 1\n",
    "    return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def capital_word_ratio(tweet):\n",
    "    \"\"\"\n",
    "    Returns capital_count / total_word_count\n",
    "    :param tweet:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    tweet = str(tweet)\n",
    "    words = tweet.split()\n",
    "    word_count = len(words)\n",
    "    capital_count = 0\n",
    "    for word in words:\n",
    "        if word.isupper():\n",
    "            capital_count += 1\n",
    "\n",
    "    return capital_count / word_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_vector2['Capitals_Ratio'] = 0\n",
    "df_vector2['URI_Included'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>#inaugurationday</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "      <th>usdx</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Capitals_Ratio</th>\n",
       "      <th>URI_Included</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 11912 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   nothing  democrats  fail  american  people  continue  impeachment  scam  \\\n",
       "0        1          1     1         1       1         1            1     1   \n",
       "1        0          0     0         1       0         0            0     0   \n",
       "2        1          1     0         0       0         0            1     0   \n",
       "3        0          0     0         0       0         0            0     0   \n",
       "4        0          0     0         0       1         0            0     0   \n",
       "\n",
       "   administration  deliver  ...  #inaugurationday  transferring  swearing-in  \\\n",
       "0               1        1  ...                 0             0            0   \n",
       "1               0        0  ...                 0             0            0   \n",
       "2               0        0  ...                 0             0            0   \n",
       "3               0        0  ...                 0             0            0   \n",
       "4               0        0  ...                 0             0            0   \n",
       "\n",
       "   #inauguration  tonight-  am…  usdx  rt_fav  Capitals_Ratio  URI_Included  \n",
       "0              0         0    0  -1.0     1.0               0             0  \n",
       "1              0         0    0  -1.0     0.0               0             0  \n",
       "2              0         0    0  -1.0     0.0               0             0  \n",
       "3              0         0    0  -1.0     1.0               0             0  \n",
       "4              0         0    0  -1.0     0.0               0             0  \n",
       "\n",
       "[5 rows x 11912 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6842/6843\r"
     ]
    }
   ],
   "source": [
    "i = 0\n",
    "for line in df_merged2['text']:\n",
    "    df_vector2.iloc[i, df_vector2.columns.get_loc('Capitals_Ratio')] = capital_word_ratio(str(line))\n",
    "    df_vector2.iloc[i, df_vector2.columns.get_loc('URI_Included')] = is_url_in_tweet(str(line))\n",
    "    print(str(i) + \"/\" + str(len(df_merged2['text'])), end=\"\\r\", flush=True)\n",
    "    i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>#inaugurationday</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "      <th>usdx</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Capitals_Ratio</th>\n",
       "      <th>URI_Included</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020408</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 11912 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   nothing  democrats  fail  american  people  continue  impeachment  scam  \\\n",
       "0        1          1     1         1       1         1            1     1   \n",
       "1        0          0     0         1       0         0            0     0   \n",
       "2        1          1     0         0       0         0            1     0   \n",
       "3        0          0     0         0       0         0            0     0   \n",
       "4        0          0     0         0       1         0            0     0   \n",
       "\n",
       "   administration  deliver  ...  #inaugurationday  transferring  swearing-in  \\\n",
       "0               1        1  ...                 0             0            0   \n",
       "1               0        0  ...                 0             0            0   \n",
       "2               0        0  ...                 0             0            0   \n",
       "3               0        0  ...                 0             0            0   \n",
       "4               0        0  ...                 0             0            0   \n",
       "\n",
       "   #inauguration  tonight-  am…  usdx  rt_fav  Capitals_Ratio  URI_Included  \n",
       "0              0         0    0  -1.0     1.0        0.100000             1  \n",
       "1              0         0    0  -1.0     0.0        0.000000             0  \n",
       "2              0         0    0  -1.0     0.0        0.000000             0  \n",
       "3              0         0    0  -1.0     1.0        0.000000             0  \n",
       "4              0         0    0  -1.0     0.0        0.020408             0  \n",
       "\n",
       "[5 rows x 11912 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from textblob import TextBlob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sentiment_analysis(tweet):\n",
    "    tweet = str(tweet)\n",
    "    blob = TextBlob(tweet)\n",
    "\n",
    "    if blob.sentiment.polarity > 0:\n",
    "        return 1\n",
    "    elif blob.sentiment.polarity == 0:\n",
    "        return 0\n",
    "    else:\n",
    "        return -1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df_tweet = pd.read_csv(\"trump_tweets.csv\", delimiter=\",\")\n",
    "#tweets = df_merged2[\"text\"]\n",
    "sentiment_list = []\n",
    "df_vector2['Sentiment_Result'] = 0\n",
    "i = 0\n",
    "for tweet in df_merged2[\"text\"]:\n",
    "    sentiment = sentiment_analysis(tweet)\n",
    "    df_vector2.iloc[i, df_vector2.columns.get_loc('Sentiment_Result')] = sentiment\n",
    "    print(str(i) + \"/\" + str(len(df_merged2['text'])), end=\"\\r\", flush=True)\n",
    "    i +=1 \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>nothing</th>\n",
       "      <th>democrats</th>\n",
       "      <th>fail</th>\n",
       "      <th>american</th>\n",
       "      <th>people</th>\n",
       "      <th>continue</th>\n",
       "      <th>impeachment</th>\n",
       "      <th>scam</th>\n",
       "      <th>administration</th>\n",
       "      <th>deliver</th>\n",
       "      <th>...</th>\n",
       "      <th>transferring</th>\n",
       "      <th>swearing-in</th>\n",
       "      <th>#inauguration</th>\n",
       "      <th>tonight-</th>\n",
       "      <th>am…</th>\n",
       "      <th>usdx</th>\n",
       "      <th>rt_fav</th>\n",
       "      <th>Capitals_Ratio</th>\n",
       "      <th>URI_Included</th>\n",
       "      <th>Sentiment_Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020408</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 11913 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   nothing  democrats  fail  american  people  continue  impeachment  scam  \\\n",
       "0        1          1     1         1       1         1            1     1   \n",
       "1        0          0     0         1       0         0            0     0   \n",
       "2        1          1     0         0       0         0            1     0   \n",
       "3        0          0     0         0       0         0            0     0   \n",
       "4        0          0     0         0       1         0            0     0   \n",
       "\n",
       "   administration  deliver  ...  transferring  swearing-in  #inauguration  \\\n",
       "0               1        1  ...             0            0              0   \n",
       "1               0        0  ...             0            0              0   \n",
       "2               0        0  ...             0            0              0   \n",
       "3               0        0  ...             0            0              0   \n",
       "4               0        0  ...             0            0              0   \n",
       "\n",
       "   tonight-  am…  usdx  rt_fav  Capitals_Ratio  URI_Included  Sentiment_Result  \n",
       "0         0    0  -1.0     1.0        0.100000             1                -1  \n",
       "1         0    0  -1.0     0.0        0.000000             0                 1  \n",
       "2         0    0  -1.0     0.0        0.000000             0                 0  \n",
       "3         0    0  -1.0     1.0        0.000000             0                 1  \n",
       "4         0    0  -1.0     0.0        0.020408             0                -1  \n",
       "\n",
       "[5 rows x 11913 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vector2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_vector2.to_csv(\"result_trump_tweets.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning Algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gaussian Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Import Gaussian Naive Bayes model\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "df_all = pd.read_csv(\"result_trump_tweets.csv\", delimiter=',')\n",
    "df_all.drop([\"Unnamed: 0\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all = df_all.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_y = df_all['usdx']\n",
    "df_X = df_all.drop(columns=['usdx'])\n",
    "X_train_orig, X_test, y_train_orig, y_test = train_test_split(df_X, df_y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test_valid, y_train, y_test_valid = train_test_split(X_train_orig, y_train_orig, test_size=0.2, random_state=42)\n",
    "df_labels = y_train\n",
    "df_features = X_train\n",
    "df_features_test = X_test_valid\n",
    "df_labels_test = y_test_valid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_prediction_stats(actual, prediction):\n",
    "    correct = 0\n",
    "    assert(len(actual) == len(prediction))\n",
    "    for i in range(len(actual)):\n",
    "        if actual.iloc[i] == prediction[i]:\n",
    "            correct += 1\n",
    "    print(\"Total predictions made:\\t\", len(actual))\n",
    "    print(\"Correct predictions: \\t\", correct)\n",
    "    print(\"Wrong predictions:   \\t\", len(actual) - correct)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate accuracy percentage\n",
    "def accuracy_metric(actual, predicted):\n",
    "    correct = 0\n",
    "    assert(len(actual) == len(predicted))\n",
    "    for i in range(len(actual)):\n",
    "        if actual.iloc[i] == predicted[i]:\n",
    "            correct += 1\n",
    "    return correct / float(len(actual)) * 100.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.3689497716894977\n",
      "Total predictions made:\t 1095\n",
      "Correct predictions: \t 404\n",
      "Wrong predictions:   \t 691\n"
     ]
    }
   ],
   "source": [
    "#Create a Gaussian Classifier\n",
    "model = GaussianNB()\n",
    "\n",
    "# Train the model using the training sets\n",
    "model.fit(df_features,df_labels)\n",
    "\n",
    "#Predict Output\n",
    "predicted= model.predict(df_features_test) # 0:Overcast, 2:Mild\n",
    "\n",
    "print(\"Accuracy:\",metrics.accuracy_score(df_labels_test, predicted))\n",
    "print_prediction_stats(df_labels_test, predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "import seaborn as sns\n",
    "sns.set()\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import pandas as pd\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr = LogisticRegression()\n",
    "lr.fit(df_features, df_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = lr.predict(df_features_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 74, 184,  44],\n",
       "       [ 66, 377,  71],\n",
       "       [ 47, 171,  61]], dtype=int64)"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(df_labels_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.46757990867579907\n",
      "Total predictions made:\t 1095\n",
      "Correct predictions: \t 512\n",
      "Wrong predictions:   \t 583\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy:\",metrics.accuracy_score(df_labels_test, y_pred))\n",
    "print_prediction_stats(df_labels_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   C_parameter  Accuracy\n",
      "0        0.001  0.469406\n",
      "1        0.010   0.47032\n",
      "2        0.100  0.483105\n",
      "3        1.000   0.46758\n",
      "4       10.000  0.443836\n",
      "5      100.000  0.418265\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "\n",
    "C_param_range = [0.001,0.01,0.1,1,10,100]\n",
    "sepal_acc_table = pd.DataFrame(columns = ['C_parameter','Accuracy'])\n",
    "sepal_acc_table['C_parameter'] = C_param_range\n",
    "\n",
    "j = 0\n",
    "for i in C_param_range:\n",
    "    \n",
    "    # Apply logistic regression model to training data\n",
    "    lr = LogisticRegression(penalty = 'l2', C = i, random_state = 0)\n",
    "    lr.fit(df_features, df_labels)\n",
    "    \n",
    "    # Predict using model\n",
    "    y_pred = lr.predict(df_features_test)\n",
    "    \n",
    "    # Saving accuracy score in table\n",
    "    sepal_acc_table.iloc[j,1] = accuracy_score(df_labels_test, y_pred)\n",
    "    j += 1\n",
    "    \n",
    "    # Printing decision regions\n",
    "\n",
    "print(sepal_acc_table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SGD Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------FITTING MODELS-------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   8 | elapsed:    9.5s remaining:   28.7s\n",
      "[Parallel(n_jobs=-1)]: Done   3 out of   8 | elapsed:    9.6s remaining:   16.0s\n",
      "[Parallel(n_jobs=-1)]: Done   4 out of   8 | elapsed:   10.4s remaining:   10.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   8 | elapsed:   10.8s remaining:    6.4s\n",
      "[Parallel(n_jobs=-1)]: Done   6 out of   8 | elapsed:   11.5s remaining:    3.8s\n",
      "[Parallel(n_jobs=-1)]: Done   8 out of   8 | elapsed:   12.2s remaining:    0.0s\n",
      "[Parallel(n_jobs=-1)]: Done   8 out of   8 | elapsed:   12.2s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------SCORING MODELS-------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   8 | elapsed:    0.3s remaining:    1.2s\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "multiclass format is not supported",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31m_RemoteTraceback\u001b[0m                          Traceback (most recent call last)",
      "\u001b[1;31m_RemoteTraceback\u001b[0m: \n\"\"\"\nTraceback (most recent call last):\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\parfit\\score.py\", line 39, in scoreOne\n    return metric(y, model.predict_proba(X)[:, 1])\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\ranking.py\", line 355, in roc_auc_score\n    sample_weight=sample_weight)\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\base.py\", line 73, in _average_binary_score\n    raise ValueError(\"{0} format is not supported\".format(y_type))\nValueError: multiclass format is not supported\n\nDuring handling of the above exception, another exception occurred:\n\nTraceback (most recent call last):\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\joblib\\externals\\loky\\process_executor.py\", line 418, in _process_worker\n    r = call_item()\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\joblib\\externals\\loky\\process_executor.py\", line 272, in __call__\n    return self.fn(*self.args, **self.kwargs)\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\joblib\\_parallel_backends.py\", line 567, in __call__\n    return self.func(*args, **kwargs)\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\", line 225, in __call__\n    for func, args, kwargs in self.items]\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\", line 225, in <listcomp>\n    for func, args, kwargs in self.items]\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\parfit\\score.py\", line 42, in scoreOne\n    return metric(y, model.predict(X))\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\ranking.py\", line 355, in roc_auc_score\n    sample_weight=sample_weight)\n  File \"C:\\Users\\John\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\base.py\", line 73, in _average_binary_score\n    raise ValueError(\"{0} format is not supported\".format(y_type))\nValueError: multiclass format is not supported\n\"\"\"",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-181-1c5a39273ba6>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     18\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     19\u001b[0m bestModel, bestScore, allModels, allScores = pf.bestFit(SGDClassifier, paramGrid,\n\u001b[1;32m---> 20\u001b[1;33m            df_features, df_labels, df_features_test, df_labels_test)\n\u001b[0m\u001b[0;32m     21\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     22\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mbestModel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbestScore\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\parfit\\parfit.py\u001b[0m in \u001b[0;36mbestFit\u001b[1;34m(model, paramGrid, X_train, y_train, X_val, y_val, nfolds, metric, greater_is_better, predict_proba, showPlot, scoreLabel, vrange, cmap, n_jobs, verbose)\u001b[0m\n\u001b[0;32m     65\u001b[0m         \u001b[0mmodels\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfitModels\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mparamGrid\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     66\u001b[0m         \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"-------------SCORING MODELS-------------\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 67\u001b[1;33m         \u001b[0mscores\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mscoreModels\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodels\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX_val\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_val\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmetric\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpredict_proba\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     68\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mshowPlot\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     69\u001b[0m         \u001b[0mplotScores\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mscores\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mparamGrid\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscoreLabel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mgreater_is_better\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvrange\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcmap\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\parfit\\score.py\u001b[0m in \u001b[0;36mscoreModels\u001b[1;34m(models, X, y, metric, predict_proba, n_jobs, verbose)\u001b[0m\n\u001b[0;32m     66\u001b[0m     \"\"\"\n\u001b[0;32m     67\u001b[0m     return Parallel(n_jobs=n_jobs, verbose=np.ceil(verbose / 10))(\n\u001b[1;32m---> 68\u001b[1;33m         \u001b[0mdelayed\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mscoreOne\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mm\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmetric\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpredict_proba\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mm\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mmodels\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     69\u001b[0m     )\n\u001b[0;32m     70\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m    932\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    933\u001b[0m             \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieval_context\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 934\u001b[1;33m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieve\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    935\u001b[0m             \u001b[1;31m# Make sure that we get a last message telling us we are done\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    936\u001b[0m             \u001b[0melapsed_time\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_start_time\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mretrieve\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    831\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    832\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'supports_timeout'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 833\u001b[1;33m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    834\u001b[0m                 \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    835\u001b[0m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\_parallel_backends.py\u001b[0m in \u001b[0;36mwrap_future_result\u001b[1;34m(future, timeout)\u001b[0m\n\u001b[0;32m    519\u001b[0m         AsyncResults.get from multiprocessing.\"\"\"\n\u001b[0;32m    520\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 521\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mfuture\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mresult\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    522\u001b[0m         \u001b[1;32mexcept\u001b[0m \u001b[0mLokyTimeoutError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    523\u001b[0m             \u001b[1;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\concurrent\\futures\\_base.py\u001b[0m in \u001b[0;36mresult\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    433\u001b[0m                 \u001b[1;32mraise\u001b[0m \u001b[0mCancelledError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    434\u001b[0m             \u001b[1;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_state\u001b[0m \u001b[1;33m==\u001b[0m \u001b[0mFINISHED\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 435\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__get_result\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    436\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    437\u001b[0m                 \u001b[1;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\concurrent\\futures\\_base.py\u001b[0m in \u001b[0;36m__get_result\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    382\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m__get_result\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    383\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_exception\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 384\u001b[1;33m             \u001b[1;32mraise\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_exception\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    385\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    386\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_result\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: multiclass format is not supported"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ParameterGrid\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.metrics import roc_auc_score\n",
    "import parfit.parfit as pf\n",
    "from sklearn.preprocessing import label_binarize\n",
    "\n",
    "#df_labels = label_binarize(df_labels, classes=[-1, 0, 1])\n",
    "#df_labels_test = label_binarize(df_labels_test, classes=[-1, 0, 1])\n",
    "\n",
    "grid = {\n",
    "    'alpha': [1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3], # learning rate\n",
    "    'max_iter': [1000], # number of epochs\n",
    "    'loss': ['log'], # logistic regression\n",
    "    'penalty': ['l2'],\n",
    "    'n_jobs': [-1]\n",
    "}\n",
    "paramGrid = ParameterGrid(grid)\n",
    "\n",
    "bestModel, bestScore, allModels, allScores = pf.bestFit(SGDClassifier, paramGrid,\n",
    "           df_features, df_labels, df_features_test, df_labels_test)\n",
    "\n",
    "print(bestModel, bestScore)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score of LogisticRegression: 0.44866858378147567\n",
      "Accuracy score of SGDClassifier: 0.43441699248722304\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "numFolds = 10\n",
    "kf = KFold(numFolds, shuffle=True)\n",
    "\n",
    "# These are \"Class objects\". For each Class, find the AUC through\n",
    "# 10 fold cross validation.\n",
    "Model = SGDClassifier\n",
    "\n",
    "# Added max_iter here\n",
    "params = []\n",
    "alphas = [1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3]\n",
    "param = { \n",
    "               \"loss\": \"log\", \n",
    "               \"penalty\": \"l2\", \n",
    "               'max_iter':1000\n",
    "        }\n",
    "for alpha in alphas:\n",
    "    param['alpha'] = alpha\n",
    "    params.append(param)\n",
    "\n",
    "\n",
    "list(itertools.product(a, b))\n",
    "\n",
    "for param in params:\n",
    "    total = 0\n",
    "    for train_indices, test_indices in kf.split(X_train_orig):\n",
    "        train_X = X_train_orig.values[train_indices, :]; train_Y = y_train_orig.values[train_indices]\n",
    "        test_X = X_train_orig.values[test_indices, :]; test_Y = y_train_orig.values[test_indices]\n",
    "        reg = Model(**param)\n",
    "        reg.fit(train_X, train_Y)\n",
    "        predictions = reg.predict(test_X)\n",
    "        total += accuracy_score(test_Y, predictions)\n",
    "\n",
    "    accuracy = total / numFolds\n",
    "    print(\"Accuracy score of {0} with Alpha{2}: {1}\".format(Model.__name__, accuracy, param['alpha']))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameters currently in use:\n",
      "\n",
      "{'bootstrap': True,\n",
      " 'class_weight': None,\n",
      " 'criterion': 'gini',\n",
      " 'max_depth': None,\n",
      " 'max_features': 'auto',\n",
      " 'max_leaf_nodes': None,\n",
      " 'min_impurity_decrease': 0.0,\n",
      " 'min_impurity_split': None,\n",
      " 'min_samples_leaf': 1,\n",
      " 'min_samples_split': 2,\n",
      " 'min_weight_fraction_leaf': 0.0,\n",
      " 'n_estimators': 'warn',\n",
      " 'n_jobs': None,\n",
      " 'oob_score': False,\n",
      " 'random_state': 42,\n",
      " 'verbose': 0,\n",
      " 'warm_start': False}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rf = RandomForestClassifier(random_state = 42)\n",
    "from pprint import pprint\n",
    "# Look at parameters used by our current forest\n",
    "print('Parameters currently in use:\\n')\n",
    "pprint(rf.get_params())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': [True, False],\n",
      " 'max_depth': [10, 60, 110],\n",
      " 'max_features': ['auto', 'sqrt'],\n",
      " 'min_samples_leaf': [2, 4],\n",
      " 'min_samples_split': [2, 5, 10],\n",
      " 'n_estimators': [200, 1100, 2000]}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "# Number of trees in random forest\n",
    "n_estimators = [int(x) for x in np.linspace(start = 200, stop = 2000, num = 3)]\n",
    "# Number of features to consider at every split\n",
    "max_features = ['auto', 'sqrt']\n",
    "# Maximum number of levels in tree\n",
    "max_depth = [int(x) for x in np.linspace(10, 110, num = 3)]\n",
    "#max_depth.append(None)\n",
    "# Minimum number of samples required to split a node\n",
    "min_samples_split = [2, 5, 10]\n",
    "# Minimum number of samples required at each leaf node\n",
    "min_samples_leaf = [2, 4]\n",
    "# Method of selecting samples for training each tree\n",
    "bootstrap = [True, False]\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_features': max_features,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split,\n",
    "               'min_samples_leaf': min_samples_leaf,\n",
    "               'bootstrap': bootstrap}\n",
    "pprint(random_grid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 4 folds for each of 10 candidates, totalling 40 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 tasks      | elapsed:   41.6s\n",
      "[Parallel(n_jobs=-1)]: Done   9 tasks      | elapsed:   48.0s\n",
      "[Parallel(n_jobs=-1)]: Done  16 tasks      | elapsed:   53.3s\n",
      "[Parallel(n_jobs=-1)]: Done  25 tasks      | elapsed:  3.9min\n",
      "[Parallel(n_jobs=-1)]: Done  30 out of  40 | elapsed:  8.3min remaining:  2.8min\n",
      "[Parallel(n_jobs=-1)]: Done  35 out of  40 | elapsed:  8.9min remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done  40 out of  40 | elapsed:  9.0min remaining:    0.0s\n",
      "[Parallel(n_jobs=-1)]: Done  40 out of  40 | elapsed:  9.0min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=4, error_score='raise-deprecating',\n",
       "                   estimator=RandomForestClassifier(bootstrap=True,\n",
       "                                                    class_weight=None,\n",
       "                                                    criterion='gini',\n",
       "                                                    max_depth=None,\n",
       "                                                    max_features='auto',\n",
       "                                                    max_leaf_nodes=None,\n",
       "                                                    min_impurity_decrease=0.0,\n",
       "                                                    min_impurity_split=None,\n",
       "                                                    min_samples_leaf=1,\n",
       "                                                    min_samples_split=2,\n",
       "                                                    min_weight_fraction_leaf=0.0,\n",
       "                                                    n_estimators='warn',\n",
       "                                                    n_jobs=None,\n",
       "                                                    oob_sc...\n",
       "                                                    random_state=None,\n",
       "                                                    verbose=0,\n",
       "                                                    warm_start=False),\n",
       "                   iid='warn', n_iter=10, n_jobs=-1,\n",
       "                   param_distributions={'bootstrap': [True, False],\n",
       "                                        'max_depth': [10, 60, 110],\n",
       "                                        'max_features': ['auto', 'sqrt'],\n",
       "                                        'min_samples_leaf': [2, 4],\n",
       "                                        'min_samples_split': [2, 5, 10],\n",
       "                                        'n_estimators': [200, 1100, 2000]},\n",
       "                   pre_dispatch='2*n_jobs', random_state=42, refit=True,\n",
       "                   return_train_score=False, scoring=None, verbose=10)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use the random grid to search for best hyperparameters\n",
    "# First create the base model to tune\n",
    "rf = RandomForestClassifier()\n",
    "# Random search of parameters, using 3 fold cross validation, \n",
    "# search across 100 different combinations, and use all available cores\n",
    "rf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, n_iter = 10, cv = 4, verbose=10, random_state=42, n_jobs = -1)\n",
    "# Fit the random search model\n",
    "rf_random.fit(df_features, df_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 200,\n",
       " 'min_samples_split': 5,\n",
       " 'min_samples_leaf': 2,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 110,\n",
       " 'bootstrap': False}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_random.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.48412879652888785"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_random.best_score_"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}