Data Preparation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compensation Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "from collections import defaultdict\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from patsy import dmatrix\n",
    "\n",
    "from fuzzywuzzy import fuzz, process\n",
    "\n",
    "pd.set_option('display.max_rows', 50)\n",
    "plt.rcParams[\"figure.figsize\"] = (20,10)\n",
    "plt.rcParams[\"font.weight\"] = \"bold\"\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>578.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>387.5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Uber</td>\n",
       "      <td>Senior</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>173.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Vancouver, BC, Canada</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>190.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>156.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>62</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>201.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>63</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>137.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>59</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>191.0</td>\n",
       "      <td>152.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>63</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>218.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>220.0</td>\n",
       "      <td>155.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Facebook</td>\n",
       "      <td>34</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>430.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>Facebook</td>\n",
       "      <td>E6</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus    company   level  \\\n",
       "entry_id                                                        \n",
       "4              578.0   180.0  387.5    9.0       Uber  Senior   \n",
       "6              173.0   120.0    0.0   53.0     Amazon      L5   \n",
       "10             190.0   110.0   80.0    0.0     Amazon      L5   \n",
       "13             156.0   135.0    8.0   13.0  Microsoft      62   \n",
       "16             201.0   157.0   26.0   28.0  Microsoft      63   \n",
       "17             137.0   110.0   16.0   11.0  Microsoft      59   \n",
       "21             191.0   152.0   17.0   22.0  Microsoft      63   \n",
       "25             218.0   150.0    7.0   61.0     Amazon      L5   \n",
       "26             220.0   155.0   50.0   15.0   Facebook      34   \n",
       "28             430.0   200.0  180.0   50.0   Facebook      E6   \n",
       "\n",
       "                       location  years_experience  years_company  tag  \n",
       "entry_id                                                               \n",
       "4             San Francisco, CA              10.0            2.0  NaN  \n",
       "6         Vancouver, BC, Canada              11.0            1.0  NaN  \n",
       "10                  Seattle, WA               3.0            3.0  NaN  \n",
       "13                  Seattle, WA               4.0            4.0  NaN  \n",
       "16                  Seattle, WA              12.0            6.0  NaN  \n",
       "17                  Seattle, WA               1.0            1.0  NaN  \n",
       "21                  Seattle, WA               7.0            7.0  NaN  \n",
       "25                  Seattle, WA              10.0            0.0  NaN  \n",
       "26            San Francisco, CA               2.0            0.0  NaN  \n",
       "28            San Francisco, CA               7.0            7.0  NaN  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comp = pd.read_csv('data/clean_comp_data.csv', index_col=\"entry_id\")\n",
    "comp.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "comp.level = comp.level.replace(to_replace = np.NaN, value = '')\n",
    "comp.tag = comp.tag.replace(to_replace = np.NaN, value = '')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Determines similar groupings in categorical values using \"trust the majority\" mentality.\n",
    "# Most people will put \"Amazon\" but some might input \"amazon\" or \"Amazon Web Services\"\n",
    "# This function uses fuzzy string matching + manually provided special cases to generate these mappings.\n",
    "def get_similar_grouping(series, special_cases = [], ratio_func = fuzz.ratio, tolerance = 90):\n",
    "    counts = series.value_counts()\n",
    "    \n",
    "    mapping = defaultdict(list)\n",
    "    already_mapped = set()\n",
    "    \n",
    "    items = list(counts.items())\n",
    "    for i in range(len(items)):\n",
    "        key1, count1 = items[i]\n",
    "        \n",
    "        if key1 in already_mapped:\n",
    "            # print(f'Ignoring {key1}. Already mapped.')\n",
    "            continue\n",
    "            \n",
    "        # print(f'Checking {key1} with count {count1}')\n",
    "        mapping[key1].append(key1)\n",
    "        for j in range(i+1, len(items)):\n",
    "            key2, count2 = items[j]\n",
    "            \n",
    "            if key2.lower() in special_cases:\n",
    "                continue\n",
    "            \n",
    "            # print(key2, count2)\n",
    "            ratio = ratio_func(key1.lower(), key2.lower())\n",
    "            if ratio > tolerance:\n",
    "                # print(f'Adding {key2} to the list of {key1}.')\n",
    "                mapping[key1].append(key2)\n",
    "                already_mapped.add(key2)\n",
    "                \n",
    "    for combination_group in special_cases:\n",
    "        most_common = combination_group[0]\n",
    "        for i in range(1, len(combination_group)):\n",
    "            less_common_company = combination_group[i]\n",
    "            for variation in mapping[less_common_company]:\n",
    "                mapping[most_common].append(variation)\n",
    "            del mapping[less_common_company]\n",
    "    \n",
    "    return mapping\n",
    "\n",
    "# Takes a particular datapoint plus a mapping generated from the above function\n",
    "# Returns the \"most common\" grouping that it should be instead\n",
    "def match_datapoint_to_common_variation(datapoint, datapoint_mapping):    \n",
    "    for possible_variation, possible_variation_group in datapoint_mapping.items():\n",
    "        for possible_variation_option in possible_variation_group:\n",
    "            if datapoint == possible_variation_option:\n",
    "                return possible_variation\n",
    "    return datapoint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dummy Creation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Non-Interaction Terms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Intercept</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30133</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Intercept\n",
       "entry_id           \n",
       "4               1.0\n",
       "6               1.0\n",
       "10              1.0\n",
       "13              1.0\n",
       "16              1.0\n",
       "...             ...\n",
       "30133           1.0\n",
       "30139           1.0\n",
       "30140           1.0\n",
       "30142           1.0\n",
       "30143           1.0\n",
       "\n",
       "[14557 rows x 1 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intercept = dmatrix(\"1\", comp, return_type='dataframe')\n",
    "intercept"
   ]
  },
  {
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "      <th>location</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>578.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>387.5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Uber</td>\n",
       "      <td>Senior</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td></td>\n",
       "      <td>SF Bay Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>173.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td></td>\n",
       "      <td>Vancouver, BC, Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>190.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td></td>\n",
       "      <td>Seattle Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>156.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>62</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td></td>\n",
       "      <td>Seattle Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>201.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>63</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td></td>\n",
       "      <td>Seattle Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>115.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>IBM</td>\n",
       "      <td>Advisory Engineer</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "      <td>Madison, WI</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>370.0</td>\n",
       "      <td>187.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>Senior SWE</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>iOS</td>\n",
       "      <td>NYC Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>212.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>Bloomberg</td>\n",
       "      <td>Senior Software Engineer</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "      <td>NYC Area</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>185.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L4</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "      <td>London, EN, United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>180.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Broadcom</td>\n",
       "      <td>ICB 3</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Networking</td>\n",
       "      <td>Tel Aviv, TA, Israel</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus    company  \\\n",
       "entry_id                                                \n",
       "4              578.0   180.0  387.5    9.0       Uber   \n",
       "6              173.0   120.0    0.0   53.0     Amazon   \n",
       "10             190.0   110.0   80.0    0.0     Amazon   \n",
       "13             156.0   135.0    8.0   13.0  Microsoft   \n",
       "16             201.0   157.0   26.0   28.0  Microsoft   \n",
       "...              ...     ...    ...    ...        ...   \n",
       "30133          115.0   115.0    0.0    0.0        IBM   \n",
       "30139          370.0   187.0  150.0   33.0    Twitter   \n",
       "30140          212.0   177.0    0.0   35.0  Bloomberg   \n",
       "30142          185.0    98.0   57.0   30.0     Google   \n",
       "30143          180.0   110.0   55.0   15.0   Broadcom   \n",
       "\n",
       "                             level  years_experience  years_company  \\\n",
       "entry_id                                                              \n",
       "4                           Senior              10.0            2.0   \n",
       "6                               L5              11.0            1.0   \n",
       "10                              L5               3.0            3.0   \n",
       "13                              62               4.0            4.0   \n",
       "16                              63              12.0            6.0   \n",
       "...                            ...               ...            ...   \n",
       "30133            Advisory Engineer               6.0            1.0   \n",
       "30139                   Senior SWE              10.0            6.0   \n",
       "30140     Senior Software Engineer               2.0            0.0   \n",
       "30142                           L4               4.0            2.0   \n",
       "30143                        ICB 3               7.0            7.0   \n",
       "\n",
       "                                     tag                    location  \n",
       "entry_id                                                              \n",
       "4                                                        SF Bay Area  \n",
       "6                                              Vancouver, BC, Canada  \n",
       "10                                                      Seattle Area  \n",
       "13                                                      Seattle Area  \n",
       "16                                                      Seattle Area  \n",
       "...                                  ...                         ...  \n",
       "30133     Distributed Systems (Back-End)                 Madison, WI  \n",
       "30139                                iOS                    NYC Area  \n",
       "30140     Distributed Systems (Back-End)                    NYC Area  \n",
       "30142         API Development (Back-End)  London, EN, United Kingdom  \n",
       "30143                         Networking        Tel Aviv, TA, Israel  \n",
       "\n",
       "[14557 rows x 10 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Provided directly from Levels.fyi site code (found in comp.js)\n",
    "location_map = {\n",
    "    \"SF Bay Area\": [\"San Francisco, CA\", \"Silicon Valley\", \"SF Bay Area\", \"Bay Area\", \"San Francisco\", \"San Francisco Bay Area\", \"San Mateo, CA\", \"Redwood City, CA\", \"Menlo Park, CA\", \"Fremont, CA\", \"Palo Alto, CA\", \"Santa Clara, CA\", \"Sunnyvale, CA\", \"San Jose, CA\", \"Berkeley, CA\", \"Oakland, CA\", \"Mountain View, CA\", \"Burlingame, CA\", \"Marin, CA\", \"Union City, CA\", \"Pleasanton, CA\", \"Cupertino, CA\"],\n",
    "    \"Seattle Area\": [\"Seattle\", \"Seattle, WA\", \"Kirkland, WA\", \"Redmond, WA\", \"Bellevue, WA\"],\n",
    "    \"NYC Area\": [\"New York\", \"New York, NY\", \"NYC\", \"NYC, NY\", \"Manhattan, NY\", \"Brooklyn, NY\"],\n",
    "    \"LA Area\": [\"Los Angeles\", \"Los Angeles, CA\", \"Santa Monica, CA\", \"Santa Monica\", \"Pasadena, CA\", \"Pasadena\", \"Irvine, CA\", \"Orange County, CA\", \"Anaheim, CA\", \"Venice, CA\", \"Burbank, CA\", \"Long Beach, CA\", \"Malibu, CA\", \"Culver City, CA\"],\n",
    "    \"Boston Area\": [\"Boston\", \"Boston, MA\", \"Cambridge\", \"Cambridge, MA\", \"Somerville, MA\", \"Newton, MA\", \"Quincy, MA\", \"Brookline, MA\", \"Malden, MA\", \"Waltham, MA\", \"Medford, MA\", \"Revere, MA\", \"Watertown, MA\", \"Lexington, MA\"],\n",
    "    \"Chicago Area\": [\"Chicago\", \"Chicago, IL\", \"Joliet, IL\", \"Naperville, IL\", \"Schaumburg, IL\", \"Evanston, IL\"],\n",
    "    \"Boulder Area\": [\"Boulder\", \"Boulder, CO\", \"Denver, CO\"]\n",
    "}\n",
    "clean_locations = comp.location.apply(match_datapoint_to_common_variation, args=(location_map,))\n",
    "pd.concat([comp.drop(columns=['location']), clean_locations], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SF Bay Area             5515\n",
       "Seattle Area            3652\n",
       "NYC Area                1028\n",
       "Boston Area              319\n",
       "Austin, TX               265\n",
       "                        ... \n",
       "San Ramon, CA             19\n",
       "Reston, VA                18\n",
       "Newark, NJ                16\n",
       "Salt Lake City, UT        15\n",
       "Montreal, QC, Canada      15\n",
       "Name: location, Length: 58, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_locations = clean_locations.value_counts()\n",
    "filtered_locations = filtered_locations[filtered_locations >= 15]\n",
    "filtered_locations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "location[Amsterdam, NH, Netherlands]     47.0\n",
       "location[Arlington, VA]                  26.0\n",
       "location[Atlanta, GA]                    66.0\n",
       "location[Austin, TX]                    265.0\n",
       "location[Bangalore, KA, India]          119.0\n",
       "                                        ...  \n",
       "location[Tokyo, TY, Japan]               19.0\n",
       "location[Toronto, ON, Canada]            72.0\n",
       "location[Vancouver, BC, Canada]         123.0\n",
       "location[Washington, DC]                 72.0\n",
       "location[Zurich, ZH, Switzerland]        59.0\n",
       "Length: 58, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>location[Amsterdam, NH, Netherlands]</th>\n",
       "      <th>location[Arlington, VA]</th>\n",
       "      <th>location[Atlanta, GA]</th>\n",
       "      <th>location[Austin, TX]</th>\n",
       "      <th>location[Bangalore, KA, India]</th>\n",
       "      <th>location[Bengaluru, KA, India]</th>\n",
       "      <th>location[Berlin, BE, Germany]</th>\n",
       "      <th>location[Boston Area]</th>\n",
       "      <th>location[Boulder Area]</th>\n",
       "      <th>location[Cambridge, EN, United Kingdom]</th>\n",
       "      <th>...</th>\n",
       "      <th>location[Seattle Area]</th>\n",
       "      <th>location[Singapore, SG, Singapore]</th>\n",
       "      <th>location[Sydney, NS, Australia]</th>\n",
       "      <th>location[Tel Aviv, TA, Israel]</th>\n",
       "      <th>location[Tempe, AZ]</th>\n",
       "      <th>location[Tokyo, TY, Japan]</th>\n",
       "      <th>location[Toronto, ON, Canada]</th>\n",
       "      <th>location[Vancouver, BC, Canada]</th>\n",
       "      <th>location[Washington, DC]</th>\n",
       "      <th>location[Zurich, ZH, Switzerland]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 58 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          location[Amsterdam, NH, Netherlands]  location[Arlington, VA]  \\\n",
       "entry_id                                                                  \n",
       "4                                          0.0                      0.0   \n",
       "6                                          0.0                      0.0   \n",
       "10                                         0.0                      0.0   \n",
       "13                                         0.0                      0.0   \n",
       "16                                         0.0                      0.0   \n",
       "...                                        ...                      ...   \n",
       "30133                                      0.0                      0.0   \n",
       "30139                                      0.0                      0.0   \n",
       "30140                                      0.0                      0.0   \n",
       "30142                                      0.0                      0.0   \n",
       "30143                                      0.0                      0.0   \n",
       "\n",
       "          location[Atlanta, GA]  location[Austin, TX]  \\\n",
       "entry_id                                                \n",
       "4                           0.0                   0.0   \n",
       "6                           0.0                   0.0   \n",
       "10                          0.0                   0.0   \n",
       "13                          0.0                   0.0   \n",
       "16                          0.0                   0.0   \n",
       "...                         ...                   ...   \n",
       "30133                       0.0                   0.0   \n",
       "30139                       0.0                   0.0   \n",
       "30140                       0.0                   0.0   \n",
       "30142                       0.0                   0.0   \n",
       "30143                       0.0                   0.0   \n",
       "\n",
       "          location[Bangalore, KA, India]  location[Bengaluru, KA, India]  \\\n",
       "entry_id                                                                   \n",
       "4                                    0.0                             0.0   \n",
       "6                                    0.0                             0.0   \n",
       "10                                   0.0                             0.0   \n",
       "13                                   0.0                             0.0   \n",
       "16                                   0.0                             0.0   \n",
       "...                                  ...                             ...   \n",
       "30133                                0.0                             0.0   \n",
       "30139                                0.0                             0.0   \n",
       "30140                                0.0                             0.0   \n",
       "30142                                0.0                             0.0   \n",
       "30143                                0.0                             0.0   \n",
       "\n",
       "          location[Berlin, BE, Germany]  location[Boston Area]  \\\n",
       "entry_id                                                         \n",
       "4                                   0.0                    0.0   \n",
       "6                                   0.0                    0.0   \n",
       "10                                  0.0                    0.0   \n",
       "13                                  0.0                    0.0   \n",
       "16                                  0.0                    0.0   \n",
       "...                                 ...                    ...   \n",
       "30133                               0.0                    0.0   \n",
       "30139                               0.0                    0.0   \n",
       "30140                               0.0                    0.0   \n",
       "30142                               0.0                    0.0   \n",
       "30143                               0.0                    0.0   \n",
       "\n",
       "          location[Boulder Area]  location[Cambridge, EN, United Kingdom]  \\\n",
       "entry_id                                                                    \n",
       "4                            0.0                                      0.0   \n",
       "6                            0.0                                      0.0   \n",
       "10                           0.0                                      0.0   \n",
       "13                           0.0                                      0.0   \n",
       "16                           0.0                                      0.0   \n",
       "...                          ...                                      ...   \n",
       "30133                        0.0                                      0.0   \n",
       "30139                        0.0                                      0.0   \n",
       "30140                        0.0                                      0.0   \n",
       "30142                        0.0                                      0.0   \n",
       "30143                        0.0                                      0.0   \n",
       "\n",
       "          ...  location[Seattle Area]  location[Singapore, SG, Singapore]  \\\n",
       "entry_id  ...                                                               \n",
       "4         ...                     0.0                                 0.0   \n",
       "6         ...                     0.0                                 0.0   \n",
       "10        ...                     1.0                                 0.0   \n",
       "13        ...                     1.0                                 0.0   \n",
       "16        ...                     1.0                                 0.0   \n",
       "...       ...                     ...                                 ...   \n",
       "30133     ...                     0.0                                 0.0   \n",
       "30139     ...                     0.0                                 0.0   \n",
       "30140     ...                     0.0                                 0.0   \n",
       "30142     ...                     0.0                                 0.0   \n",
       "30143     ...                     0.0                                 0.0   \n",
       "\n",
       "          location[Sydney, NS, Australia]  location[Tel Aviv, TA, Israel]  \\\n",
       "entry_id                                                                    \n",
       "4                                     0.0                             0.0   \n",
       "6                                     0.0                             0.0   \n",
       "10                                    0.0                             0.0   \n",
       "13                                    0.0                             0.0   \n",
       "16                                    0.0                             0.0   \n",
       "...                                   ...                             ...   \n",
       "30133                                 0.0                             0.0   \n",
       "30139                                 0.0                             0.0   \n",
       "30140                                 0.0                             0.0   \n",
       "30142                                 0.0                             0.0   \n",
       "30143                                 0.0                             1.0   \n",
       "\n",
       "          location[Tempe, AZ]  location[Tokyo, TY, Japan]  \\\n",
       "entry_id                                                    \n",
       "4                         0.0                         0.0   \n",
       "6                         0.0                         0.0   \n",
       "10                        0.0                         0.0   \n",
       "13                        0.0                         0.0   \n",
       "16                        0.0                         0.0   \n",
       "...                       ...                         ...   \n",
       "30133                     0.0                         0.0   \n",
       "30139                     0.0                         0.0   \n",
       "30140                     0.0                         0.0   \n",
       "30142                     0.0                         0.0   \n",
       "30143                     0.0                         0.0   \n",
       "\n",
       "          location[Toronto, ON, Canada]  location[Vancouver, BC, Canada]  \\\n",
       "entry_id                                                                   \n",
       "4                                   0.0                              0.0   \n",
       "6                                   0.0                              1.0   \n",
       "10                                  0.0                              0.0   \n",
       "13                                  0.0                              0.0   \n",
       "16                                  0.0                              0.0   \n",
       "...                                 ...                              ...   \n",
       "30133                               0.0                              0.0   \n",
       "30139                               0.0                              0.0   \n",
       "30140                               0.0                              0.0   \n",
       "30142                               0.0                              0.0   \n",
       "30143                               0.0                              0.0   \n",
       "\n",
       "          location[Washington, DC]  location[Zurich, ZH, Switzerland]  \n",
       "entry_id                                                               \n",
       "4                              0.0                                0.0  \n",
       "6                              0.0                                0.0  \n",
       "10                             0.0                                0.0  \n",
       "13                             0.0                                0.0  \n",
       "16                             0.0                                0.0  \n",
       "...                            ...                                ...  \n",
       "30133                          0.0                                0.0  \n",
       "30139                          0.0                                0.0  \n",
       "30140                          0.0                                0.0  \n",
       "30142                          0.0                                0.0  \n",
       "30143                          0.0                                0.0  \n",
       "\n",
       "[14557 rows x 58 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummies_locations = dmatrix(\"0 + location\", pd.DataFrame({ 'location': clean_locations }) , return_type='dataframe')\n",
    "filtered_dummies_locations = dummies_locations.loc[:, (dummies_locations.sum(axis=0) >= 15)]\n",
    "filtered_dummies_locations.sum(axis=0)\n",
    "filtered_dummies_locations"
   ]
  },
  {
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "      <th>company</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>578.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>387.5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Senior</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td></td>\n",
       "      <td>Uber</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>173.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>L5</td>\n",
       "      <td>Vancouver, BC, Canada</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td></td>\n",
       "      <td>Amazon</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>190.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td></td>\n",
       "      <td>Amazon</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>156.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>62</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td></td>\n",
       "      <td>Microsoft</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>201.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>63</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td></td>\n",
       "      <td>Microsoft</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>115.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Advisory Engineer</td>\n",
       "      <td>Madison, WI</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "      <td>IBM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>370.0</td>\n",
       "      <td>187.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Senior SWE</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>iOS</td>\n",
       "      <td>Twitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>212.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>Senior Software Engineer</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "      <td>Bloomberg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>185.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>L4</td>\n",
       "      <td>London, EN, United Kingdom</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "      <td>Google</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>180.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>ICB 3</td>\n",
       "      <td>Tel Aviv, TA, Israel</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Networking</td>\n",
       "      <td>Broadcom</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus                     level  \\\n",
       "entry_id                                                               \n",
       "4              578.0   180.0  387.5    9.0                    Senior   \n",
       "6              173.0   120.0    0.0   53.0                        L5   \n",
       "10             190.0   110.0   80.0    0.0                        L5   \n",
       "13             156.0   135.0    8.0   13.0                        62   \n",
       "16             201.0   157.0   26.0   28.0                        63   \n",
       "...              ...     ...    ...    ...                       ...   \n",
       "30133          115.0   115.0    0.0    0.0         Advisory Engineer   \n",
       "30139          370.0   187.0  150.0   33.0                Senior SWE   \n",
       "30140          212.0   177.0    0.0   35.0  Senior Software Engineer   \n",
       "30142          185.0    98.0   57.0   30.0                        L4   \n",
       "30143          180.0   110.0   55.0   15.0                     ICB 3   \n",
       "\n",
       "                            location  years_experience  years_company  \\\n",
       "entry_id                                                                \n",
       "4                  San Francisco, CA              10.0            2.0   \n",
       "6              Vancouver, BC, Canada              11.0            1.0   \n",
       "10                       Seattle, WA               3.0            3.0   \n",
       "13                       Seattle, WA               4.0            4.0   \n",
       "16                       Seattle, WA              12.0            6.0   \n",
       "...                              ...               ...            ...   \n",
       "30133                    Madison, WI               6.0            1.0   \n",
       "30139                   New York, NY              10.0            6.0   \n",
       "30140                   New York, NY               2.0            0.0   \n",
       "30142     London, EN, United Kingdom               4.0            2.0   \n",
       "30143           Tel Aviv, TA, Israel               7.0            7.0   \n",
       "\n",
       "                                     tag    company  \n",
       "entry_id                                             \n",
       "4                                              Uber  \n",
       "6                                            Amazon  \n",
       "10                                           Amazon  \n",
       "13                                        Microsoft  \n",
       "16                                        Microsoft  \n",
       "...                                  ...        ...  \n",
       "30133     Distributed Systems (Back-End)        IBM  \n",
       "30139                                iOS    Twitter  \n",
       "30140     Distributed Systems (Back-End)  Bloomberg  \n",
       "30142         API Development (Back-End)     Google  \n",
       "30143                         Networking   Broadcom  \n",
       "\n",
       "[14557 rows x 10 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "company_special_cases = [\n",
    "    ['Amazon', 'Amazon Web Services', 'AWS'],\n",
    "    ['Akamai', 'Akamai Technologies'],\n",
    "    ['Cisco', 'Cisco Systems', 'Cisco Meraki'],\n",
    "    ['Disney', 'The Walt Disney Company'],\n",
    "    ['Dell', 'Dell Technologies'],\n",
    "    ['EPAM Systems', 'Epam'],\n",
    "    ['Expedia', 'Expedia Group'],\n",
    "    ['GE Digital', 'GE'],\n",
    "    ['HP', 'HP Inc', 'Hewlett Packard Enterprise'],\n",
    "    ['Intel', 'Intel Corporation'],\n",
    "    ['SAP', 'SAP Concur'],\n",
    "    ['Samsung', 'Samsung Research America'],\n",
    "    ['Snap', 'Snapchat'],\n",
    "    ['Verizon', 'Verizon Media'],\n",
    "    ['Walmart', 'Walmart Labs'],\n",
    "    ['Zillow', 'Zillow Group']\n",
    "]\n",
    "company_map = get_similar_grouping(comp.company, company_special_cases)\n",
    "clean_companies = comp.company.apply(match_datapoint_to_common_variation, args=(company_map,))\n",
    "pd.concat([comp.drop(columns=['company']), clean_companies], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Amazon                1963\n",
       "Microsoft             1525\n",
       "Google                1423\n",
       "Facebook               849\n",
       "Apple                  595\n",
       "                      ... \n",
       "Citrix Systems Inc      15\n",
       "MathWorks               15\n",
       "Deloitte                15\n",
       "AT&T                    15\n",
       "Coinbase                15\n",
       "Name: company, Length: 114, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_companies = clean_companies.value_counts()\n",
    "filtered_companies = filtered_companies[filtered_companies >= 15]\n",
    "filtered_companies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "company[AT&T]           15.0\n",
       "company[Accenture]      28.0\n",
       "company[Adobe]          88.0\n",
       "company[Airbnb]         96.0\n",
       "company[Amazon]       1963.0\n",
       "                       ...  \n",
       "company[Yahoo]          95.0\n",
       "company[Yandex]         59.0\n",
       "company[Yelp]           96.0\n",
       "company[Zillow]         64.0\n",
       "company[eBay]          124.0\n",
       "Length: 114, dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>company[AT&amp;T]</th>\n",
       "      <th>company[Accenture]</th>\n",
       "      <th>company[Adobe]</th>\n",
       "      <th>company[Airbnb]</th>\n",
       "      <th>company[Amazon]</th>\n",
       "      <th>company[American Express]</th>\n",
       "      <th>company[Andela]</th>\n",
       "      <th>company[Apple]</th>\n",
       "      <th>company[Argo AI]</th>\n",
       "      <th>company[Atlassian]</th>\n",
       "      <th>...</th>\n",
       "      <th>company[Visa]</th>\n",
       "      <th>company[Walmart]</th>\n",
       "      <th>company[Wayfair]</th>\n",
       "      <th>company[WeWork]</th>\n",
       "      <th>company[Workday]</th>\n",
       "      <th>company[Yahoo]</th>\n",
       "      <th>company[Yandex]</th>\n",
       "      <th>company[Yelp]</th>\n",
       "      <th>company[Zillow]</th>\n",
       "      <th>company[eBay]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 114 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          company[AT&T]  company[Accenture]  company[Adobe]  company[Airbnb]  \\\n",
       "entry_id                                                                       \n",
       "4                   0.0                 0.0             0.0              0.0   \n",
       "6                   0.0                 0.0             0.0              0.0   \n",
       "10                  0.0                 0.0             0.0              0.0   \n",
       "13                  0.0                 0.0             0.0              0.0   \n",
       "16                  0.0                 0.0             0.0              0.0   \n",
       "...                 ...                 ...             ...              ...   \n",
       "30133               0.0                 0.0             0.0              0.0   \n",
       "30139               0.0                 0.0             0.0              0.0   \n",
       "30140               0.0                 0.0             0.0              0.0   \n",
       "30142               0.0                 0.0             0.0              0.0   \n",
       "30143               0.0                 0.0             0.0              0.0   \n",
       "\n",
       "          company[Amazon]  company[American Express]  company[Andela]  \\\n",
       "entry_id                                                                \n",
       "4                     0.0                        0.0              0.0   \n",
       "6                     1.0                        0.0              0.0   \n",
       "10                    1.0                        0.0              0.0   \n",
       "13                    0.0                        0.0              0.0   \n",
       "16                    0.0                        0.0              0.0   \n",
       "...                   ...                        ...              ...   \n",
       "30133                 0.0                        0.0              0.0   \n",
       "30139                 0.0                        0.0              0.0   \n",
       "30140                 0.0                        0.0              0.0   \n",
       "30142                 0.0                        0.0              0.0   \n",
       "30143                 0.0                        0.0              0.0   \n",
       "\n",
       "          company[Apple]  company[Argo AI]  company[Atlassian]  ...  \\\n",
       "entry_id                                                        ...   \n",
       "4                    0.0               0.0                 0.0  ...   \n",
       "6                    0.0               0.0                 0.0  ...   \n",
       "10                   0.0               0.0                 0.0  ...   \n",
       "13                   0.0               0.0                 0.0  ...   \n",
       "16                   0.0               0.0                 0.0  ...   \n",
       "...                  ...               ...                 ...  ...   \n",
       "30133                0.0               0.0                 0.0  ...   \n",
       "30139                0.0               0.0                 0.0  ...   \n",
       "30140                0.0               0.0                 0.0  ...   \n",
       "30142                0.0               0.0                 0.0  ...   \n",
       "30143                0.0               0.0                 0.0  ...   \n",
       "\n",
       "          company[Visa]  company[Walmart]  company[Wayfair]  company[WeWork]  \\\n",
       "entry_id                                                                       \n",
       "4                   0.0               0.0               0.0              0.0   \n",
       "6                   0.0               0.0               0.0              0.0   \n",
       "10                  0.0               0.0               0.0              0.0   \n",
       "13                  0.0               0.0               0.0              0.0   \n",
       "16                  0.0               0.0               0.0              0.0   \n",
       "...                 ...               ...               ...              ...   \n",
       "30133               0.0               0.0               0.0              0.0   \n",
       "30139               0.0               0.0               0.0              0.0   \n",
       "30140               0.0               0.0               0.0              0.0   \n",
       "30142               0.0               0.0               0.0              0.0   \n",
       "30143               0.0               0.0               0.0              0.0   \n",
       "\n",
       "          company[Workday]  company[Yahoo]  company[Yandex]  company[Yelp]  \\\n",
       "entry_id                                                                     \n",
       "4                      0.0             0.0              0.0            0.0   \n",
       "6                      0.0             0.0              0.0            0.0   \n",
       "10                     0.0             0.0              0.0            0.0   \n",
       "13                     0.0             0.0              0.0            0.0   \n",
       "16                     0.0             0.0              0.0            0.0   \n",
       "...                    ...             ...              ...            ...   \n",
       "30133                  0.0             0.0              0.0            0.0   \n",
       "30139                  0.0             0.0              0.0            0.0   \n",
       "30140                  0.0             0.0              0.0            0.0   \n",
       "30142                  0.0             0.0              0.0            0.0   \n",
       "30143                  0.0             0.0              0.0            0.0   \n",
       "\n",
       "          company[Zillow]  company[eBay]  \n",
       "entry_id                                  \n",
       "4                     0.0            0.0  \n",
       "6                     0.0            0.0  \n",
       "10                    0.0            0.0  \n",
       "13                    0.0            0.0  \n",
       "16                    0.0            0.0  \n",
       "...                   ...            ...  \n",
       "30133                 0.0            0.0  \n",
       "30139                 0.0            0.0  \n",
       "30140                 0.0            0.0  \n",
       "30142                 0.0            0.0  \n",
       "30143                 0.0            0.0  \n",
       "\n",
       "[14557 rows x 114 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummies_companies = dmatrix(\"0 + company\", pd.DataFrame({ 'company': clean_companies }), return_type='dataframe')\n",
    "filtered_dummies_companies = dummies_companies.loc[:, (dummies_companies.sum(axis=0) >= 15)]\n",
    "filtered_dummies_companies.sum(axis=0)\n",
    "filtered_dummies_companies"
   ]
  },
  {
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>578.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>387.5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Uber</td>\n",
       "      <td>Senior</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>173.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Vancouver, BC, Canada</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>190.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>156.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>62</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>201.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>63</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>115.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>IBM</td>\n",
       "      <td>Advisory Engineer</td>\n",
       "      <td>Madison, WI</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>370.0</td>\n",
       "      <td>187.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>Senior SWE</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>iOS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>212.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>Bloomberg</td>\n",
       "      <td>Senior Software Engineer</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>185.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L4</td>\n",
       "      <td>London, EN, United Kingdom</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>180.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Broadcom</td>\n",
       "      <td>ICB 3</td>\n",
       "      <td>Tel Aviv, TA, Israel</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Networking</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus    company  \\\n",
       "entry_id                                                \n",
       "4              578.0   180.0  387.5    9.0       Uber   \n",
       "6              173.0   120.0    0.0   53.0     Amazon   \n",
       "10             190.0   110.0   80.0    0.0     Amazon   \n",
       "13             156.0   135.0    8.0   13.0  Microsoft   \n",
       "16             201.0   157.0   26.0   28.0  Microsoft   \n",
       "...              ...     ...    ...    ...        ...   \n",
       "30133          115.0   115.0    0.0    0.0        IBM   \n",
       "30139          370.0   187.0  150.0   33.0    Twitter   \n",
       "30140          212.0   177.0    0.0   35.0  Bloomberg   \n",
       "30142          185.0    98.0   57.0   30.0     Google   \n",
       "30143          180.0   110.0   55.0   15.0   Broadcom   \n",
       "\n",
       "                             level                    location  \\\n",
       "entry_id                                                         \n",
       "4                           Senior           San Francisco, CA   \n",
       "6                               L5       Vancouver, BC, Canada   \n",
       "10                              L5                 Seattle, WA   \n",
       "13                              62                 Seattle, WA   \n",
       "16                              63                 Seattle, WA   \n",
       "...                            ...                         ...   \n",
       "30133            Advisory Engineer                 Madison, WI   \n",
       "30139                   Senior SWE                New York, NY   \n",
       "30140     Senior Software Engineer                New York, NY   \n",
       "30142                           L4  London, EN, United Kingdom   \n",
       "30143                        ICB 3        Tel Aviv, TA, Israel   \n",
       "\n",
       "          years_experience  years_company                             tag  \n",
       "entry_id                                                                   \n",
       "4                     10.0            2.0                                  \n",
       "6                     11.0            1.0                                  \n",
       "10                     3.0            3.0                                  \n",
       "13                     4.0            4.0                                  \n",
       "16                    12.0            6.0                                  \n",
       "...                    ...            ...                             ...  \n",
       "30133                  6.0            1.0  Distributed Systems (Back-End)  \n",
       "30139                 10.0            6.0                             iOS  \n",
       "30140                  2.0            0.0  Distributed Systems (Back-End)  \n",
       "30142                  4.0            2.0      API Development (Back-End)  \n",
       "30143                  7.0            7.0                      Networking  \n",
       "\n",
       "[14557 rows x 10 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tag_special_cases = [\n",
    "    ['Embedded', 'Embedded Systems', 'Embedded Software'],\n",
    "    ['Operating Systems', 'OS'],\n",
    "    ['Data', 'Data Engineering']\n",
    "]\n",
    "tag_map = get_similar_grouping(comp.tag, tag_special_cases)\n",
    "clean_tags = comp.tag.apply(match_datapoint_to_common_variation, args=(tag_map,))\n",
    "pd.concat([comp.drop(columns=['tag']), clean_tags], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Distributed Systems (Back-End)    3591\n",
       "Full Stack                        3294\n",
       "API Development (Back-End)        2024\n",
       "ML / AI                           1044\n",
       "Web Development (Front-End)        813\n",
       "DevOps                             463\n",
       "iOS                                426\n",
       "Android                            347\n",
       "Networking                         315\n",
       "Security                           268\n",
       "Testing (SDET)                     226\n",
       "Mobile (iOS + Android)             208\n",
       "Embedded                            96\n",
       "Site Reliability (SRE)              90\n",
       "Firmware                            50\n",
       "Data                                40\n",
       "Operating Systems                   29\n",
       "Name: tag, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_tags = clean_tags.value_counts()\n",
    "filtered_tags = filtered_tags[filtered_tags >= 25]\n",
    "filtered_tags = filtered_tags.drop('')\n",
    "filtered_tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tag[API Development (Back-End)]        2024.0\n",
       "tag[Android]                            347.0\n",
       "tag[Data]                                40.0\n",
       "tag[DevOps]                             463.0\n",
       "tag[Distributed Systems (Back-End)]    3591.0\n",
       "tag[Embedded]                            96.0\n",
       "tag[Firmware]                            50.0\n",
       "tag[Full Stack]                        3294.0\n",
       "tag[ML / AI]                           1044.0\n",
       "tag[Mobile (iOS + Android)]             208.0\n",
       "tag[Networking]                         315.0\n",
       "tag[Operating Systems]                   29.0\n",
       "tag[Security]                           268.0\n",
       "tag[Site Reliability (SRE)]              90.0\n",
       "tag[Testing (SDET)]                     226.0\n",
       "tag[Web Development (Front-End)]        813.0\n",
       "tag[iOS]                                426.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>tag[API Development (Back-End)]</th>\n",
       "      <th>tag[Android]</th>\n",
       "      <th>tag[Data]</th>\n",
       "      <th>tag[DevOps]</th>\n",
       "      <th>tag[Distributed Systems (Back-End)]</th>\n",
       "      <th>tag[Embedded]</th>\n",
       "      <th>tag[Firmware]</th>\n",
       "      <th>tag[Full Stack]</th>\n",
       "      <th>tag[ML / AI]</th>\n",
       "      <th>tag[Mobile (iOS + Android)]</th>\n",
       "      <th>tag[Networking]</th>\n",
       "      <th>tag[Operating Systems]</th>\n",
       "      <th>tag[Security]</th>\n",
       "      <th>tag[Site Reliability (SRE)]</th>\n",
       "      <th>tag[Testing (SDET)]</th>\n",
       "      <th>tag[Web Development (Front-End)]</th>\n",
       "      <th>tag[iOS]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          tag[API Development (Back-End)]  tag[Android]  tag[Data]  \\\n",
       "entry_id                                                             \n",
       "4                                     0.0           0.0        0.0   \n",
       "6                                     0.0           0.0        0.0   \n",
       "10                                    0.0           0.0        0.0   \n",
       "13                                    0.0           0.0        0.0   \n",
       "16                                    0.0           0.0        0.0   \n",
       "...                                   ...           ...        ...   \n",
       "30133                                 0.0           0.0        0.0   \n",
       "30139                                 0.0           0.0        0.0   \n",
       "30140                                 0.0           0.0        0.0   \n",
       "30142                                 1.0           0.0        0.0   \n",
       "30143                                 0.0           0.0        0.0   \n",
       "\n",
       "          tag[DevOps]  tag[Distributed Systems (Back-End)]  tag[Embedded]  \\\n",
       "entry_id                                                                    \n",
       "4                 0.0                                  0.0            0.0   \n",
       "6                 0.0                                  0.0            0.0   \n",
       "10                0.0                                  0.0            0.0   \n",
       "13                0.0                                  0.0            0.0   \n",
       "16                0.0                                  0.0            0.0   \n",
       "...               ...                                  ...            ...   \n",
       "30133             0.0                                  1.0            0.0   \n",
       "30139             0.0                                  0.0            0.0   \n",
       "30140             0.0                                  1.0            0.0   \n",
       "30142             0.0                                  0.0            0.0   \n",
       "30143             0.0                                  0.0            0.0   \n",
       "\n",
       "          tag[Firmware]  tag[Full Stack]  tag[ML / AI]  \\\n",
       "entry_id                                                 \n",
       "4                   0.0              0.0           0.0   \n",
       "6                   0.0              0.0           0.0   \n",
       "10                  0.0              0.0           0.0   \n",
       "13                  0.0              0.0           0.0   \n",
       "16                  0.0              0.0           0.0   \n",
       "...                 ...              ...           ...   \n",
       "30133               0.0              0.0           0.0   \n",
       "30139               0.0              0.0           0.0   \n",
       "30140               0.0              0.0           0.0   \n",
       "30142               0.0              0.0           0.0   \n",
       "30143               0.0              0.0           0.0   \n",
       "\n",
       "          tag[Mobile (iOS + Android)]  tag[Networking]  \\\n",
       "entry_id                                                 \n",
       "4                                 0.0              0.0   \n",
       "6                                 0.0              0.0   \n",
       "10                                0.0              0.0   \n",
       "13                                0.0              0.0   \n",
       "16                                0.0              0.0   \n",
       "...                               ...              ...   \n",
       "30133                             0.0              0.0   \n",
       "30139                             0.0              0.0   \n",
       "30140                             0.0              0.0   \n",
       "30142                             0.0              0.0   \n",
       "30143                             0.0              1.0   \n",
       "\n",
       "          tag[Operating Systems]  tag[Security]  tag[Site Reliability (SRE)]  \\\n",
       "entry_id                                                                       \n",
       "4                            0.0            0.0                          0.0   \n",
       "6                            0.0            0.0                          0.0   \n",
       "10                           0.0            0.0                          0.0   \n",
       "13                           0.0            0.0                          0.0   \n",
       "16                           0.0            0.0                          0.0   \n",
       "...                          ...            ...                          ...   \n",
       "30133                        0.0            0.0                          0.0   \n",
       "30139                        0.0            0.0                          0.0   \n",
       "30140                        0.0            0.0                          0.0   \n",
       "30142                        0.0            0.0                          0.0   \n",
       "30143                        0.0            0.0                          0.0   \n",
       "\n",
       "          tag[Testing (SDET)]  tag[Web Development (Front-End)]  tag[iOS]  \n",
       "entry_id                                                                   \n",
       "4                         0.0                               0.0       0.0  \n",
       "6                         0.0                               0.0       0.0  \n",
       "10                        0.0                               0.0       0.0  \n",
       "13                        0.0                               0.0       0.0  \n",
       "16                        0.0                               0.0       0.0  \n",
       "...                       ...                               ...       ...  \n",
       "30133                     0.0                               0.0       0.0  \n",
       "30139                     0.0                               0.0       1.0  \n",
       "30140                     0.0                               0.0       0.0  \n",
       "30142                     0.0                               0.0       0.0  \n",
       "30143                     0.0                               0.0       0.0  \n",
       "\n",
       "[14557 rows x 17 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummies_tags = dmatrix(\"0 + tag\", pd.DataFrame({ 'tag': clean_tags }), return_type='dataframe')\n",
    "filtered_dummies_tags = dummies_tags.loc[:, (dummies_tags.sum(axis=0) >= 25)]\n",
    "filtered_dummies_tags = filtered_dummies_tags.drop(columns=['tag[]'])\n",
    "filtered_dummies_tags.sum(axis=0)\n",
    "filtered_dummies_tags"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Interaction Terms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "company_level[Adobe_Senior Software Engineer (4)]    20.0\n",
       "company_level[Adobe_Senior Software Engineer (5)]    12.0\n",
       "company_level[Adobe_Software Engineer 3]             19.0\n",
       "company_level[Airbnb_L3]                             14.0\n",
       "company_level[Airbnb_L4]                             34.0\n",
       "                                                     ... \n",
       "company_level[Zillow_P3]                             17.0\n",
       "company_level[Zillow_P4]                             22.0\n",
       "company_level[eBay_MTS 1]                            17.0\n",
       "company_level[eBay_MTS 2]                            17.0\n",
       "company_level[eBay_SE 3]                             31.0\n",
       "Length: 227, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>company_level[Adobe_Senior Software Engineer (4)]</th>\n",
       "      <th>company_level[Adobe_Senior Software Engineer (5)]</th>\n",
       "      <th>company_level[Adobe_Software Engineer 3]</th>\n",
       "      <th>company_level[Airbnb_L3]</th>\n",
       "      <th>company_level[Airbnb_L4]</th>\n",
       "      <th>company_level[Airbnb_L5]</th>\n",
       "      <th>company_level[Airbnb_L6]</th>\n",
       "      <th>company_level[Amazon_5]</th>\n",
       "      <th>company_level[Amazon_L4]</th>\n",
       "      <th>company_level[Amazon_L5]</th>\n",
       "      <th>...</th>\n",
       "      <th>company_level[Yandex_G18]</th>\n",
       "      <th>company_level[Yelp_IC1]</th>\n",
       "      <th>company_level[Yelp_IC2]</th>\n",
       "      <th>company_level[Yelp_IC3]</th>\n",
       "      <th>company_level[Yelp_IC4]</th>\n",
       "      <th>company_level[Zillow_P3]</th>\n",
       "      <th>company_level[Zillow_P4]</th>\n",
       "      <th>company_level[eBay_MTS 1]</th>\n",
       "      <th>company_level[eBay_MTS 2]</th>\n",
       "      <th>company_level[eBay_SE 3]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 227 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          company_level[Adobe_Senior Software Engineer (4)]  \\\n",
       "entry_id                                                      \n",
       "4                                                       0.0   \n",
       "6                                                       0.0   \n",
       "10                                                      0.0   \n",
       "13                                                      0.0   \n",
       "16                                                      0.0   \n",
       "...                                                     ...   \n",
       "30133                                                   0.0   \n",
       "30139                                                   0.0   \n",
       "30140                                                   0.0   \n",
       "30142                                                   0.0   \n",
       "30143                                                   0.0   \n",
       "\n",
       "          company_level[Adobe_Senior Software Engineer (5)]  \\\n",
       "entry_id                                                      \n",
       "4                                                       0.0   \n",
       "6                                                       0.0   \n",
       "10                                                      0.0   \n",
       "13                                                      0.0   \n",
       "16                                                      0.0   \n",
       "...                                                     ...   \n",
       "30133                                                   0.0   \n",
       "30139                                                   0.0   \n",
       "30140                                                   0.0   \n",
       "30142                                                   0.0   \n",
       "30143                                                   0.0   \n",
       "\n",
       "          company_level[Adobe_Software Engineer 3]  company_level[Airbnb_L3]  \\\n",
       "entry_id                                                                       \n",
       "4                                              0.0                       0.0   \n",
       "6                                              0.0                       0.0   \n",
       "10                                             0.0                       0.0   \n",
       "13                                             0.0                       0.0   \n",
       "16                                             0.0                       0.0   \n",
       "...                                            ...                       ...   \n",
       "30133                                          0.0                       0.0   \n",
       "30139                                          0.0                       0.0   \n",
       "30140                                          0.0                       0.0   \n",
       "30142                                          0.0                       0.0   \n",
       "30143                                          0.0                       0.0   \n",
       "\n",
       "          company_level[Airbnb_L4]  company_level[Airbnb_L5]  \\\n",
       "entry_id                                                       \n",
       "4                              0.0                       0.0   \n",
       "6                              0.0                       0.0   \n",
       "10                             0.0                       0.0   \n",
       "13                             0.0                       0.0   \n",
       "16                             0.0                       0.0   \n",
       "...                            ...                       ...   \n",
       "30133                          0.0                       0.0   \n",
       "30139                          0.0                       0.0   \n",
       "30140                          0.0                       0.0   \n",
       "30142                          0.0                       0.0   \n",
       "30143                          0.0                       0.0   \n",
       "\n",
       "          company_level[Airbnb_L6]  company_level[Amazon_5]  \\\n",
       "entry_id                                                      \n",
       "4                              0.0                      0.0   \n",
       "6                              0.0                      0.0   \n",
       "10                             0.0                      0.0   \n",
       "13                             0.0                      0.0   \n",
       "16                             0.0                      0.0   \n",
       "...                            ...                      ...   \n",
       "30133                          0.0                      0.0   \n",
       "30139                          0.0                      0.0   \n",
       "30140                          0.0                      0.0   \n",
       "30142                          0.0                      0.0   \n",
       "30143                          0.0                      0.0   \n",
       "\n",
       "          company_level[Amazon_L4]  company_level[Amazon_L5]  ...  \\\n",
       "entry_id                                                      ...   \n",
       "4                              0.0                       0.0  ...   \n",
       "6                              0.0                       1.0  ...   \n",
       "10                             0.0                       1.0  ...   \n",
       "13                             0.0                       0.0  ...   \n",
       "16                             0.0                       0.0  ...   \n",
       "...                            ...                       ...  ...   \n",
       "30133                          0.0                       0.0  ...   \n",
       "30139                          0.0                       0.0  ...   \n",
       "30140                          0.0                       0.0  ...   \n",
       "30142                          0.0                       0.0  ...   \n",
       "30143                          0.0                       0.0  ...   \n",
       "\n",
       "          company_level[Yandex_G18]  company_level[Yelp_IC1]  \\\n",
       "entry_id                                                       \n",
       "4                               0.0                      0.0   \n",
       "6                               0.0                      0.0   \n",
       "10                              0.0                      0.0   \n",
       "13                              0.0                      0.0   \n",
       "16                              0.0                      0.0   \n",
       "...                             ...                      ...   \n",
       "30133                           0.0                      0.0   \n",
       "30139                           0.0                      0.0   \n",
       "30140                           0.0                      0.0   \n",
       "30142                           0.0                      0.0   \n",
       "30143                           0.0                      0.0   \n",
       "\n",
       "          company_level[Yelp_IC2]  company_level[Yelp_IC3]  \\\n",
       "entry_id                                                     \n",
       "4                             0.0                      0.0   \n",
       "6                             0.0                      0.0   \n",
       "10                            0.0                      0.0   \n",
       "13                            0.0                      0.0   \n",
       "16                            0.0                      0.0   \n",
       "...                           ...                      ...   \n",
       "30133                         0.0                      0.0   \n",
       "30139                         0.0                      0.0   \n",
       "30140                         0.0                      0.0   \n",
       "30142                         0.0                      0.0   \n",
       "30143                         0.0                      0.0   \n",
       "\n",
       "          company_level[Yelp_IC4]  company_level[Zillow_P3]  \\\n",
       "entry_id                                                      \n",
       "4                             0.0                       0.0   \n",
       "6                             0.0                       0.0   \n",
       "10                            0.0                       0.0   \n",
       "13                            0.0                       0.0   \n",
       "16                            0.0                       0.0   \n",
       "...                           ...                       ...   \n",
       "30133                         0.0                       0.0   \n",
       "30139                         0.0                       0.0   \n",
       "30140                         0.0                       0.0   \n",
       "30142                         0.0                       0.0   \n",
       "30143                         0.0                       0.0   \n",
       "\n",
       "          company_level[Zillow_P4]  company_level[eBay_MTS 1]  \\\n",
       "entry_id                                                        \n",
       "4                              0.0                        0.0   \n",
       "6                              0.0                        0.0   \n",
       "10                             0.0                        0.0   \n",
       "13                             0.0                        0.0   \n",
       "16                             0.0                        0.0   \n",
       "...                            ...                        ...   \n",
       "30133                          0.0                        0.0   \n",
       "30139                          0.0                        0.0   \n",
       "30140                          0.0                        0.0   \n",
       "30142                          0.0                        0.0   \n",
       "30143                          0.0                        0.0   \n",
       "\n",
       "          company_level[eBay_MTS 2]  company_level[eBay_SE 3]  \n",
       "entry_id                                                       \n",
       "4                               0.0                       0.0  \n",
       "6                               0.0                       0.0  \n",
       "10                              0.0                       0.0  \n",
       "13                              0.0                       0.0  \n",
       "16                              0.0                       0.0  \n",
       "...                             ...                       ...  \n",
       "30133                           0.0                       0.0  \n",
       "30139                           0.0                       0.0  \n",
       "30140                           0.0                       0.0  \n",
       "30142                           0.0                       0.0  \n",
       "30143                           0.0                       0.0  \n",
       "\n",
       "[14557 rows x 227 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "company_levels = pd.DataFrame({ 'company_level': clean_companies + \"_\" + comp[\"level\"] }) \n",
    "dummies_company_levels = dmatrix(\"0 + company_level\", company_levels, return_type='dataframe')\n",
    "filtered_dummies_company_levels = dummies_company_levels.loc[:, (dummies_company_levels.sum(axis=0) >= 10)]\n",
    "filtered_dummies_company_levels.sum(axis=0)\n",
    "filtered_dummies_company_levels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "company_location[Adobe_NYC Area]          10.0\n",
       "company_location[Adobe_SF Bay Area]       40.0\n",
       "company_location[Airbnb_SF Bay Area]      80.0\n",
       "company_location[Airbnb_Seattle Area]     10.0\n",
       "company_location[Amazon_Arlington, VA]    13.0\n",
       "                                          ... \n",
       "company_location[Yelp_SF Bay Area]        92.0\n",
       "company_location[Zillow_SF Bay Area]      12.0\n",
       "company_location[Zillow_Seattle Area]     42.0\n",
       "company_location[eBay_SF Bay Area]        97.0\n",
       "company_location[eBay_Seattle Area]       12.0\n",
       "Length: 197, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>company_location[Adobe_NYC Area]</th>\n",
       "      <th>company_location[Adobe_SF Bay Area]</th>\n",
       "      <th>company_location[Airbnb_SF Bay Area]</th>\n",
       "      <th>company_location[Airbnb_Seattle Area]</th>\n",
       "      <th>company_location[Amazon_Arlington, VA]</th>\n",
       "      <th>company_location[Amazon_Austin, TX]</th>\n",
       "      <th>company_location[Amazon_Bangalore, KA, India]</th>\n",
       "      <th>company_location[Amazon_Boston Area]</th>\n",
       "      <th>company_location[Amazon_Boulder Area]</th>\n",
       "      <th>company_location[Amazon_Dublin, DN, Ireland]</th>\n",
       "      <th>...</th>\n",
       "      <th>company_location[WeWork_NYC Area]</th>\n",
       "      <th>company_location[WeWork_SF Bay Area]</th>\n",
       "      <th>company_location[Workday_SF Bay Area]</th>\n",
       "      <th>company_location[Yahoo_SF Bay Area]</th>\n",
       "      <th>company_location[Yandex_Moscow, MC, Russia]</th>\n",
       "      <th>company_location[Yelp_SF Bay Area]</th>\n",
       "      <th>company_location[Zillow_SF Bay Area]</th>\n",
       "      <th>company_location[Zillow_Seattle Area]</th>\n",
       "      <th>company_location[eBay_SF Bay Area]</th>\n",
       "      <th>company_location[eBay_Seattle Area]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 197 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          company_location[Adobe_NYC Area]  \\\n",
       "entry_id                                     \n",
       "4                                      0.0   \n",
       "6                                      0.0   \n",
       "10                                     0.0   \n",
       "13                                     0.0   \n",
       "16                                     0.0   \n",
       "...                                    ...   \n",
       "30133                                  0.0   \n",
       "30139                                  0.0   \n",
       "30140                                  0.0   \n",
       "30142                                  0.0   \n",
       "30143                                  0.0   \n",
       "\n",
       "          company_location[Adobe_SF Bay Area]  \\\n",
       "entry_id                                        \n",
       "4                                         0.0   \n",
       "6                                         0.0   \n",
       "10                                        0.0   \n",
       "13                                        0.0   \n",
       "16                                        0.0   \n",
       "...                                       ...   \n",
       "30133                                     0.0   \n",
       "30139                                     0.0   \n",
       "30140                                     0.0   \n",
       "30142                                     0.0   \n",
       "30143                                     0.0   \n",
       "\n",
       "          company_location[Airbnb_SF Bay Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Airbnb_Seattle Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[Amazon_Arlington, VA]  \\\n",
       "entry_id                                           \n",
       "4                                            0.0   \n",
       "6                                            0.0   \n",
       "10                                           0.0   \n",
       "13                                           0.0   \n",
       "16                                           0.0   \n",
       "...                                          ...   \n",
       "30133                                        0.0   \n",
       "30139                                        0.0   \n",
       "30140                                        0.0   \n",
       "30142                                        0.0   \n",
       "30143                                        0.0   \n",
       "\n",
       "          company_location[Amazon_Austin, TX]  \\\n",
       "entry_id                                        \n",
       "4                                         0.0   \n",
       "6                                         0.0   \n",
       "10                                        0.0   \n",
       "13                                        0.0   \n",
       "16                                        0.0   \n",
       "...                                       ...   \n",
       "30133                                     0.0   \n",
       "30139                                     0.0   \n",
       "30140                                     0.0   \n",
       "30142                                     0.0   \n",
       "30143                                     0.0   \n",
       "\n",
       "          company_location[Amazon_Bangalore, KA, India]  \\\n",
       "entry_id                                                  \n",
       "4                                                   0.0   \n",
       "6                                                   0.0   \n",
       "10                                                  0.0   \n",
       "13                                                  0.0   \n",
       "16                                                  0.0   \n",
       "...                                                 ...   \n",
       "30133                                               0.0   \n",
       "30139                                               0.0   \n",
       "30140                                               0.0   \n",
       "30142                                               0.0   \n",
       "30143                                               0.0   \n",
       "\n",
       "          company_location[Amazon_Boston Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Amazon_Boulder Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[Amazon_Dublin, DN, Ireland]  ...  \\\n",
       "entry_id                                                ...   \n",
       "4                                                  0.0  ...   \n",
       "6                                                  0.0  ...   \n",
       "10                                                 0.0  ...   \n",
       "13                                                 0.0  ...   \n",
       "16                                                 0.0  ...   \n",
       "...                                                ...  ...   \n",
       "30133                                              0.0  ...   \n",
       "30139                                              0.0  ...   \n",
       "30140                                              0.0  ...   \n",
       "30142                                              0.0  ...   \n",
       "30143                                              0.0  ...   \n",
       "\n",
       "          company_location[WeWork_NYC Area]  \\\n",
       "entry_id                                      \n",
       "4                                       0.0   \n",
       "6                                       0.0   \n",
       "10                                      0.0   \n",
       "13                                      0.0   \n",
       "16                                      0.0   \n",
       "...                                     ...   \n",
       "30133                                   0.0   \n",
       "30139                                   0.0   \n",
       "30140                                   0.0   \n",
       "30142                                   0.0   \n",
       "30143                                   0.0   \n",
       "\n",
       "          company_location[WeWork_SF Bay Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Workday_SF Bay Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[Yahoo_SF Bay Area]  \\\n",
       "entry_id                                        \n",
       "4                                         0.0   \n",
       "6                                         0.0   \n",
       "10                                        0.0   \n",
       "13                                        0.0   \n",
       "16                                        0.0   \n",
       "...                                       ...   \n",
       "30133                                     0.0   \n",
       "30139                                     0.0   \n",
       "30140                                     0.0   \n",
       "30142                                     0.0   \n",
       "30143                                     0.0   \n",
       "\n",
       "          company_location[Yandex_Moscow, MC, Russia]  \\\n",
       "entry_id                                                \n",
       "4                                                 0.0   \n",
       "6                                                 0.0   \n",
       "10                                                0.0   \n",
       "13                                                0.0   \n",
       "16                                                0.0   \n",
       "...                                               ...   \n",
       "30133                                             0.0   \n",
       "30139                                             0.0   \n",
       "30140                                             0.0   \n",
       "30142                                             0.0   \n",
       "30143                                             0.0   \n",
       "\n",
       "          company_location[Yelp_SF Bay Area]  \\\n",
       "entry_id                                       \n",
       "4                                        0.0   \n",
       "6                                        0.0   \n",
       "10                                       0.0   \n",
       "13                                       0.0   \n",
       "16                                       0.0   \n",
       "...                                      ...   \n",
       "30133                                    0.0   \n",
       "30139                                    0.0   \n",
       "30140                                    0.0   \n",
       "30142                                    0.0   \n",
       "30143                                    0.0   \n",
       "\n",
       "          company_location[Zillow_SF Bay Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Zillow_Seattle Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[eBay_SF Bay Area]  \\\n",
       "entry_id                                       \n",
       "4                                        0.0   \n",
       "6                                        0.0   \n",
       "10                                       0.0   \n",
       "13                                       0.0   \n",
       "16                                       0.0   \n",
       "...                                      ...   \n",
       "30133                                    0.0   \n",
       "30139                                    0.0   \n",
       "30140                                    0.0   \n",
       "30142                                    0.0   \n",
       "30143                                    0.0   \n",
       "\n",
       "          company_location[eBay_Seattle Area]  \n",
       "entry_id                                       \n",
       "4                                         0.0  \n",
       "6                                         0.0  \n",
       "10                                        0.0  \n",
       "13                                        0.0  \n",
       "16                                        0.0  \n",
       "...                                       ...  \n",
       "30133                                     0.0  \n",
       "30139                                     0.0  \n",
       "30140                                     0.0  \n",
       "30142                                     0.0  \n",
       "30143                                     0.0  \n",
       "\n",
       "[14557 rows x 197 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "company_locations = pd.DataFrame({ 'company_location': clean_companies + \"_\" + clean_locations }) \n",
    "dummies_company_locations = dmatrix(\"0 + company_location\", company_locations, return_type='dataframe')\n",
    "filtered_dummies_company_locations = dummies_company_locations.loc[:, (dummies_company_locations.sum(axis=0) >= 10)]\n",
    "filtered_dummies_company_locations.sum(axis=0)\n",
    "filtered_dummies_company_locations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1042</th>\n",
       "      <td>310.0</td>\n",
       "      <td>146.0</td>\n",
       "      <td>164.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>8.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6488</th>\n",
       "      <td>248.0</td>\n",
       "      <td>153.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21860</th>\n",
       "      <td>247.0</td>\n",
       "      <td>149.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L4</td>\n",
       "      <td>Mountain View, CA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7900</th>\n",
       "      <td>237.0</td>\n",
       "      <td>139.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4392</th>\n",
       "      <td>330.0</td>\n",
       "      <td>173.0</td>\n",
       "      <td>119.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L5</td>\n",
       "      <td>Mountain View, CA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>28116</th>\n",
       "      <td>176.0</td>\n",
       "      <td>153.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>Bloomberg</td>\n",
       "      <td>Software Engineer</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13550</th>\n",
       "      <td>205.0</td>\n",
       "      <td>152.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8130</th>\n",
       "      <td>448.0</td>\n",
       "      <td>210.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L6</td>\n",
       "      <td>Zurich, ZH, Switzerland</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>ML / AI</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13210</th>\n",
       "      <td>295.0</td>\n",
       "      <td>160.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Lyft</td>\n",
       "      <td>T4</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4479</th>\n",
       "      <td>480.0</td>\n",
       "      <td>209.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>Airbnb</td>\n",
       "      <td>L5</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>13101 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus    company              level  \\\n",
       "entry_id                                                                   \n",
       "1042           310.0   146.0  164.0    0.0     Amazon                 L5   \n",
       "6488           248.0   153.0   95.0    0.0     Amazon                 L5   \n",
       "21860          247.0   149.0   65.0   33.0     Google                 L4   \n",
       "7900           237.0   139.0   98.0    0.0     Amazon                 L5   \n",
       "4392           330.0   173.0  119.0   38.0     Google                 L5   \n",
       "...              ...     ...    ...    ...        ...                ...   \n",
       "28116          176.0   153.0    0.0   23.0  Bloomberg  Software Engineer   \n",
       "13550          205.0   152.0   33.0   21.0     Amazon                 L5   \n",
       "8130           448.0   210.0  190.0   48.0     Google                 L6   \n",
       "13210          295.0   160.0  130.0    5.0       Lyft                 T4   \n",
       "4479           480.0   209.0  200.0   70.0     Airbnb                 L5   \n",
       "\n",
       "                         location  years_experience  years_company  \\\n",
       "entry_id                                                             \n",
       "1042                  Seattle, WA               8.0            3.0   \n",
       "6488            San Francisco, CA               4.0            4.0   \n",
       "21860           Mountain View, CA               2.0            2.0   \n",
       "7900                  Seattle, WA               5.0            5.0   \n",
       "4392            Mountain View, CA               4.0            4.0   \n",
       "...                           ...               ...            ...   \n",
       "28116                New York, NY               2.0            2.0   \n",
       "13550                New York, NY               6.0            1.0   \n",
       "8130      Zurich, ZH, Switzerland               9.0            2.0   \n",
       "13210           San Francisco, CA               2.0            0.0   \n",
       "4479            San Francisco, CA               8.0            1.0   \n",
       "\n",
       "                                     tag  \n",
       "entry_id                                  \n",
       "1042      Distributed Systems (Back-End)  \n",
       "6488      Distributed Systems (Back-End)  \n",
       "21860         API Development (Back-End)  \n",
       "7900                          Full Stack  \n",
       "4392                          Full Stack  \n",
       "...                                  ...  \n",
       "28116                         Full Stack  \n",
       "13550                         Full Stack  \n",
       "8130                             ML / AI  \n",
       "13210         API Development (Back-End)  \n",
       "4479          API Development (Back-End)  \n",
       "\n",
       "[13101 rows x 10 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23789</th>\n",
       "      <td>225.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>Oracle</td>\n",
       "      <td>IC-2</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11706</th>\n",
       "      <td>130.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>60</td>\n",
       "      <td>Bellevue, WA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310</th>\n",
       "      <td>275.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Airbnb</td>\n",
       "      <td>L4</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>815</th>\n",
       "      <td>152.5</td>\n",
       "      <td>115.0</td>\n",
       "      <td>17.5</td>\n",
       "      <td>20.0</td>\n",
       "      <td>Apple</td>\n",
       "      <td>ICT3</td>\n",
       "      <td>Cupertino, CA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Networking</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5982</th>\n",
       "      <td>263.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>LinkedIn</td>\n",
       "      <td>Senior Software Engineer</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>DevOps</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>25211</th>\n",
       "      <td>118.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Capital One</td>\n",
       "      <td>Associate Software Eng</td>\n",
       "      <td>Washington, DC</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25710</th>\n",
       "      <td>150.0</td>\n",
       "      <td>147.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>ThoughtWorks</td>\n",
       "      <td>Senior</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Full Stack</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13958</th>\n",
       "      <td>332.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L5</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4446</th>\n",
       "      <td>300.0</td>\n",
       "      <td>155.0</td>\n",
       "      <td>152.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L6</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8299</th>\n",
       "      <td>320.0</td>\n",
       "      <td>156.0</td>\n",
       "      <td>160.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L6</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>15.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Data Engineer</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1456 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus       company  \\\n",
       "entry_id                                                   \n",
       "23789          225.0   150.0   55.0   20.0        Oracle   \n",
       "11706          130.0   117.0   10.0    3.0     Microsoft   \n",
       "310            275.0   170.0  105.0    0.0        Airbnb   \n",
       "815            152.5   115.0   17.5   20.0         Apple   \n",
       "5982           263.0   170.0   75.0   17.0      LinkedIn   \n",
       "...              ...     ...    ...    ...           ...   \n",
       "25211          118.0   113.0    0.0    5.0   Capital One   \n",
       "25710          150.0   147.0    0.0    3.0  ThoughtWorks   \n",
       "13958          332.0   180.0  125.0   27.0        Google   \n",
       "4446           300.0   155.0  152.0    0.0        Amazon   \n",
       "8299           320.0   156.0  160.0    0.0        Amazon   \n",
       "\n",
       "                             level           location  years_experience  \\\n",
       "entry_id                                                                  \n",
       "23789                         IC-2        Seattle, WA               3.0   \n",
       "11706                           60       Bellevue, WA              10.0   \n",
       "310                             L4  San Francisco, CA               3.0   \n",
       "815                           ICT3      Cupertino, CA               0.0   \n",
       "5982      Senior Software Engineer       New York, NY               8.0   \n",
       "...                            ...                ...               ...   \n",
       "25211       Associate Software Eng     Washington, DC               3.0   \n",
       "25710                       Senior       New York, NY               3.0   \n",
       "13958                           L5       New York, NY              11.0   \n",
       "4446                            L6        Seattle, WA               8.0   \n",
       "8299                            L6       New York, NY              15.0   \n",
       "\n",
       "          years_company                             tag  \n",
       "entry_id                                                 \n",
       "23789               0.0      API Development (Back-End)  \n",
       "11706               6.0                      Full Stack  \n",
       "310                 1.0                                  \n",
       "815                 0.0                      Networking  \n",
       "5982                0.0                          DevOps  \n",
       "...                 ...                             ...  \n",
       "25211               0.0                      Full Stack  \n",
       "25710               1.0                      Full Stack  \n",
       "13958               0.0  Distributed Systems (Back-End)  \n",
       "4446                6.0  Distributed Systems (Back-End)  \n",
       "8299                4.0                   Data Engineer  \n",
       "\n",
       "[1456 rows x 10 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RANDOM_STATE = 721 # Lucky number; will be reused for future training/test split\n",
    "X_train_dry_run, X_test_dry_run, _, _ = train_test_split(comp, comp, test_size=0.1, random_state=RANDOM_STATE)\n",
    "X_train_dry_run\n",
    "X_test_dry_run"
   ]
  },
  {
   "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>scaled_years_experience</th>\n",
       "      <th>scaled_years_company</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.712125</td>\n",
       "      <td>-0.123292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.896446</td>\n",
       "      <td>-0.467863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>-0.578121</td>\n",
       "      <td>0.221278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>-0.393800</td>\n",
       "      <td>0.565848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.080767</td>\n",
       "      <td>1.254988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30133</th>\n",
       "      <td>-0.025158</td>\n",
       "      <td>-0.467863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>0.712125</td>\n",
       "      <td>1.254988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>-0.762442</td>\n",
       "      <td>-0.812433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>-0.393800</td>\n",
       "      <td>-0.123292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>0.159163</td>\n",
       "      <td>1.599558</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          scaled_years_experience  scaled_years_company\n",
       "entry_id                                               \n",
       "4                        0.712125             -0.123292\n",
       "6                        0.896446             -0.467863\n",
       "10                      -0.578121              0.221278\n",
       "13                      -0.393800              0.565848\n",
       "16                       1.080767              1.254988\n",
       "...                           ...                   ...\n",
       "30133                   -0.025158             -0.467863\n",
       "30139                    0.712125              1.254988\n",
       "30140                   -0.762442             -0.812433\n",
       "30142                   -0.393800             -0.123292\n",
       "30143                    0.159163              1.599558\n",
       "\n",
       "[14557 rows x 2 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler_basis_data = X_train_dry_run[['years_experience', 'years_company']]\n",
    "scaler_years = StandardScaler().fit(scaler_basis_data)\n",
    "scaled_raw_data = scaler_years.transform(comp[['years_experience', 'years_company']])\n",
    "\n",
    "scaled_data = pd.DataFrame(scaled_raw_data, index = comp.index, columns = ['scaled_years_experience', 'scaled_years_company'])\n",
    "scaled_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Target Transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>log_total_comp</th>\n",
       "      <th>log_salary</th>\n",
       "      <th>log_stock</th>\n",
       "      <th>log_bonus</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.359576</td>\n",
       "      <td>5.192962</td>\n",
       "      <td>5.959718</td>\n",
       "      <td>2.197336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>5.153297</td>\n",
       "      <td>4.787500</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.970311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>5.247029</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.382039</td>\n",
       "      <td>-6.907755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>5.049862</td>\n",
       "      <td>4.905282</td>\n",
       "      <td>2.079567</td>\n",
       "      <td>2.565026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5.303310</td>\n",
       "      <td>5.056252</td>\n",
       "      <td>3.258135</td>\n",
       "      <td>3.332240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30133</th>\n",
       "      <td>4.744941</td>\n",
       "      <td>4.744941</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-6.907755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>5.913506</td>\n",
       "      <td>5.231114</td>\n",
       "      <td>5.010642</td>\n",
       "      <td>3.496538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>5.356591</td>\n",
       "      <td>5.176155</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.555377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>5.220361</td>\n",
       "      <td>4.584978</td>\n",
       "      <td>4.043069</td>\n",
       "      <td>3.401231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>5.192962</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.007351</td>\n",
       "      <td>2.708117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          log_total_comp  log_salary  log_stock  log_bonus\n",
       "entry_id                                                  \n",
       "4               6.359576    5.192962   5.959718   2.197336\n",
       "6               5.153297    4.787500  -6.907755   3.970311\n",
       "10              5.247029    4.700489   4.382039  -6.907755\n",
       "13              5.049862    4.905282   2.079567   2.565026\n",
       "16              5.303310    5.056252   3.258135   3.332240\n",
       "...                  ...         ...        ...        ...\n",
       "30133           4.744941    4.744941  -6.907755  -6.907755\n",
       "30139           5.913506    5.231114   5.010642   3.496538\n",
       "30140           5.356591    5.176155  -6.907755   3.555377\n",
       "30142           5.220361    4.584978   4.043069   3.401231\n",
       "30143           5.192962    4.700489   4.007351   2.708117\n",
       "\n",
       "[14557 rows x 4 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "translation = 0.001 # Necessary to be able to log transform 0 values\n",
    "log_total_comp = np.log(comp.total_comp + translation).rename('log_total_comp')\n",
    "log_salary = np.log(comp.salary + translation).rename('log_salary')\n",
    "log_stock = np.log(comp.stock + translation).rename('log_stock')\n",
    "log_bonus = np.log(comp.bonus + translation).rename('log_bonus')\n",
    "pd.concat([log_total_comp, log_salary, log_stock, log_bonus], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Final Dataset Combination"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>log_total_comp</th>\n",
       "      <th>log_salary</th>\n",
       "      <th>log_stock</th>\n",
       "      <th>log_bonus</th>\n",
       "      <th>scaled_years_experience</th>\n",
       "      <th>scaled_years_company</th>\n",
       "      <th>company[AT&amp;T]</th>\n",
       "      <th>company[Accenture]</th>\n",
       "      <th>company[Adobe]</th>\n",
       "      <th>company[Airbnb]</th>\n",
       "      <th>...</th>\n",
       "      <th>tag[Full Stack]</th>\n",
       "      <th>tag[ML / AI]</th>\n",
       "      <th>tag[Mobile (iOS + Android)]</th>\n",
       "      <th>tag[Networking]</th>\n",
       "      <th>tag[Operating Systems]</th>\n",
       "      <th>tag[Security]</th>\n",
       "      <th>tag[Site Reliability (SRE)]</th>\n",
       "      <th>tag[Testing (SDET)]</th>\n",
       "      <th>tag[Web Development (Front-End)]</th>\n",
       "      <th>tag[iOS]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.359576</td>\n",
       "      <td>5.192962</td>\n",
       "      <td>5.959718</td>\n",
       "      <td>2.197336</td>\n",
       "      <td>0.712125</td>\n",
       "      <td>-0.123292</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>5.153297</td>\n",
       "      <td>4.787500</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.970311</td>\n",
       "      <td>0.896446</td>\n",
       "      <td>-0.467863</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>5.247029</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.382039</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-0.578121</td>\n",
       "      <td>0.221278</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>5.049862</td>\n",
       "      <td>4.905282</td>\n",
       "      <td>2.079567</td>\n",
       "      <td>2.565026</td>\n",
       "      <td>-0.393800</td>\n",
       "      <td>0.565848</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5.303310</td>\n",
       "      <td>5.056252</td>\n",
       "      <td>3.258135</td>\n",
       "      <td>3.332240</td>\n",
       "      <td>1.080767</td>\n",
       "      <td>1.254988</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>4.744941</td>\n",
       "      <td>4.744941</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-0.025158</td>\n",
       "      <td>-0.467863</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>5.913506</td>\n",
       "      <td>5.231114</td>\n",
       "      <td>5.010642</td>\n",
       "      <td>3.496538</td>\n",
       "      <td>0.712125</td>\n",
       "      <td>1.254988</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>5.356591</td>\n",
       "      <td>5.176155</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.555377</td>\n",
       "      <td>-0.762442</td>\n",
       "      <td>-0.812433</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>5.220361</td>\n",
       "      <td>4.584978</td>\n",
       "      <td>4.043069</td>\n",
       "      <td>3.401231</td>\n",
       "      <td>-0.393800</td>\n",
       "      <td>-0.123292</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>5.192962</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.007351</td>\n",
       "      <td>2.708117</td>\n",
       "      <td>0.159163</td>\n",
       "      <td>1.599558</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 195 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          log_total_comp  log_salary  log_stock  log_bonus  \\\n",
       "entry_id                                                     \n",
       "4               6.359576    5.192962   5.959718   2.197336   \n",
       "6               5.153297    4.787500  -6.907755   3.970311   \n",
       "10              5.247029    4.700489   4.382039  -6.907755   \n",
       "13              5.049862    4.905282   2.079567   2.565026   \n",
       "16              5.303310    5.056252   3.258135   3.332240   \n",
       "...                  ...         ...        ...        ...   \n",
       "30133           4.744941    4.744941  -6.907755  -6.907755   \n",
       "30139           5.913506    5.231114   5.010642   3.496538   \n",
       "30140           5.356591    5.176155  -6.907755   3.555377   \n",
       "30142           5.220361    4.584978   4.043069   3.401231   \n",
       "30143           5.192962    4.700489   4.007351   2.708117   \n",
       "\n",
       "          scaled_years_experience  scaled_years_company  company[AT&T]  \\\n",
       "entry_id                                                                 \n",
       "4                        0.712125             -0.123292            0.0   \n",
       "6                        0.896446             -0.467863            0.0   \n",
       "10                      -0.578121              0.221278            0.0   \n",
       "13                      -0.393800              0.565848            0.0   \n",
       "16                       1.080767              1.254988            0.0   \n",
       "...                           ...                   ...            ...   \n",
       "30133                   -0.025158             -0.467863            0.0   \n",
       "30139                    0.712125              1.254988            0.0   \n",
       "30140                   -0.762442             -0.812433            0.0   \n",
       "30142                   -0.393800             -0.123292            0.0   \n",
       "30143                    0.159163              1.599558            0.0   \n",
       "\n",
       "          company[Accenture]  company[Adobe]  company[Airbnb]  ...  \\\n",
       "entry_id                                                       ...   \n",
       "4                        0.0             0.0              0.0  ...   \n",
       "6                        0.0             0.0              0.0  ...   \n",
       "10                       0.0             0.0              0.0  ...   \n",
       "13                       0.0             0.0              0.0  ...   \n",
       "16                       0.0             0.0              0.0  ...   \n",
       "...                      ...             ...              ...  ...   \n",
       "30133                    0.0             0.0              0.0  ...   \n",
       "30139                    0.0             0.0              0.0  ...   \n",
       "30140                    0.0             0.0              0.0  ...   \n",
       "30142                    0.0             0.0              0.0  ...   \n",
       "30143                    0.0             0.0              0.0  ...   \n",
       "\n",
       "          tag[Full Stack]  tag[ML / AI]  tag[Mobile (iOS + Android)]  \\\n",
       "entry_id                                                               \n",
       "4                     0.0           0.0                          0.0   \n",
       "6                     0.0           0.0                          0.0   \n",
       "10                    0.0           0.0                          0.0   \n",
       "13                    0.0           0.0                          0.0   \n",
       "16                    0.0           0.0                          0.0   \n",
       "...                   ...           ...                          ...   \n",
       "30133                 0.0           0.0                          0.0   \n",
       "30139                 0.0           0.0                          0.0   \n",
       "30140                 0.0           0.0                          0.0   \n",
       "30142                 0.0           0.0                          0.0   \n",
       "30143                 0.0           0.0                          0.0   \n",
       "\n",
       "          tag[Networking]  tag[Operating Systems]  tag[Security]  \\\n",
       "entry_id                                                           \n",
       "4                     0.0                     0.0            0.0   \n",
       "6                     0.0                     0.0            0.0   \n",
       "10                    0.0                     0.0            0.0   \n",
       "13                    0.0                     0.0            0.0   \n",
       "16                    0.0                     0.0            0.0   \n",
       "...                   ...                     ...            ...   \n",
       "30133                 0.0                     0.0            0.0   \n",
       "30139                 0.0                     0.0            0.0   \n",
       "30140                 0.0                     0.0            0.0   \n",
       "30142                 0.0                     0.0            0.0   \n",
       "30143                 1.0                     0.0            0.0   \n",
       "\n",
       "          tag[Site Reliability (SRE)]  tag[Testing (SDET)]  \\\n",
       "entry_id                                                     \n",
       "4                                 0.0                  0.0   \n",
       "6                                 0.0                  0.0   \n",
       "10                                0.0                  0.0   \n",
       "13                                0.0                  0.0   \n",
       "16                                0.0                  0.0   \n",
       "...                               ...                  ...   \n",
       "30133                             0.0                  0.0   \n",
       "30139                             0.0                  0.0   \n",
       "30140                             0.0                  0.0   \n",
       "30142                             0.0                  0.0   \n",
       "30143                             0.0                  0.0   \n",
       "\n",
       "          tag[Web Development (Front-End)]  tag[iOS]  \n",
       "entry_id                                              \n",
       "4                                      0.0       0.0  \n",
       "6                                      0.0       0.0  \n",
       "10                                     0.0       0.0  \n",
       "13                                     0.0       0.0  \n",
       "16                                     0.0       0.0  \n",
       "...                                    ...       ...  \n",
       "30133                                  0.0       0.0  \n",
       "30139                                  0.0       1.0  \n",
       "30140                                  0.0       0.0  \n",
       "30142                                  0.0       0.0  \n",
       "30143                                  0.0       0.0  \n",
       "\n",
       "[14557 rows x 195 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comp_final = pd.concat([\n",
    "    log_total_comp,log_salary, log_stock, log_bonus,\n",
    "    scaled_data, filtered_dummies_companies, filtered_dummies_locations, filtered_dummies_tags,\n",
    "], axis=1)\n",
    "comp_final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>log_total_comp</th>\n",
       "      <th>log_salary</th>\n",
       "      <th>log_stock</th>\n",
       "      <th>log_bonus</th>\n",
       "      <th>scaled_years_experience</th>\n",
       "      <th>scaled_years_company</th>\n",
       "      <th>company[AT&amp;T]</th>\n",
       "      <th>company[Accenture]</th>\n",
       "      <th>company[Adobe]</th>\n",
       "      <th>company[Airbnb]</th>\n",
       "      <th>...</th>\n",
       "      <th>company_location[WeWork_NYC Area]</th>\n",
       "      <th>company_location[WeWork_SF Bay Area]</th>\n",
       "      <th>company_location[Workday_SF Bay Area]</th>\n",
       "      <th>company_location[Yahoo_SF Bay Area]</th>\n",
       "      <th>company_location[Yandex_Moscow, MC, Russia]</th>\n",
       "      <th>company_location[Yelp_SF Bay Area]</th>\n",
       "      <th>company_location[Zillow_SF Bay Area]</th>\n",
       "      <th>company_location[Zillow_Seattle Area]</th>\n",
       "      <th>company_location[eBay_SF Bay Area]</th>\n",
       "      <th>company_location[eBay_Seattle Area]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.359576</td>\n",
       "      <td>5.192962</td>\n",
       "      <td>5.959718</td>\n",
       "      <td>2.197336</td>\n",
       "      <td>0.712125</td>\n",
       "      <td>-0.123292</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>5.153297</td>\n",
       "      <td>4.787500</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.970311</td>\n",
       "      <td>0.896446</td>\n",
       "      <td>-0.467863</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>5.247029</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.382039</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-0.578121</td>\n",
       "      <td>0.221278</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>5.049862</td>\n",
       "      <td>4.905282</td>\n",
       "      <td>2.079567</td>\n",
       "      <td>2.565026</td>\n",
       "      <td>-0.393800</td>\n",
       "      <td>0.565848</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5.303310</td>\n",
       "      <td>5.056252</td>\n",
       "      <td>3.258135</td>\n",
       "      <td>3.332240</td>\n",
       "      <td>1.080767</td>\n",
       "      <td>1.254988</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>4.744941</td>\n",
       "      <td>4.744941</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>-0.025158</td>\n",
       "      <td>-0.467863</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>5.913506</td>\n",
       "      <td>5.231114</td>\n",
       "      <td>5.010642</td>\n",
       "      <td>3.496538</td>\n",
       "      <td>0.712125</td>\n",
       "      <td>1.254988</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>5.356591</td>\n",
       "      <td>5.176155</td>\n",
       "      <td>-6.907755</td>\n",
       "      <td>3.555377</td>\n",
       "      <td>-0.762442</td>\n",
       "      <td>-0.812433</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>5.220361</td>\n",
       "      <td>4.584978</td>\n",
       "      <td>4.043069</td>\n",
       "      <td>3.401231</td>\n",
       "      <td>-0.393800</td>\n",
       "      <td>-0.123292</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>5.192962</td>\n",
       "      <td>4.700489</td>\n",
       "      <td>4.007351</td>\n",
       "      <td>2.708117</td>\n",
       "      <td>0.159163</td>\n",
       "      <td>1.599558</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 619 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          log_total_comp  log_salary  log_stock  log_bonus  \\\n",
       "entry_id                                                     \n",
       "4               6.359576    5.192962   5.959718   2.197336   \n",
       "6               5.153297    4.787500  -6.907755   3.970311   \n",
       "10              5.247029    4.700489   4.382039  -6.907755   \n",
       "13              5.049862    4.905282   2.079567   2.565026   \n",
       "16              5.303310    5.056252   3.258135   3.332240   \n",
       "...                  ...         ...        ...        ...   \n",
       "30133           4.744941    4.744941  -6.907755  -6.907755   \n",
       "30139           5.913506    5.231114   5.010642   3.496538   \n",
       "30140           5.356591    5.176155  -6.907755   3.555377   \n",
       "30142           5.220361    4.584978   4.043069   3.401231   \n",
       "30143           5.192962    4.700489   4.007351   2.708117   \n",
       "\n",
       "          scaled_years_experience  scaled_years_company  company[AT&T]  \\\n",
       "entry_id                                                                 \n",
       "4                        0.712125             -0.123292            0.0   \n",
       "6                        0.896446             -0.467863            0.0   \n",
       "10                      -0.578121              0.221278            0.0   \n",
       "13                      -0.393800              0.565848            0.0   \n",
       "16                       1.080767              1.254988            0.0   \n",
       "...                           ...                   ...            ...   \n",
       "30133                   -0.025158             -0.467863            0.0   \n",
       "30139                    0.712125              1.254988            0.0   \n",
       "30140                   -0.762442             -0.812433            0.0   \n",
       "30142                   -0.393800             -0.123292            0.0   \n",
       "30143                    0.159163              1.599558            0.0   \n",
       "\n",
       "          company[Accenture]  company[Adobe]  company[Airbnb]  ...  \\\n",
       "entry_id                                                       ...   \n",
       "4                        0.0             0.0              0.0  ...   \n",
       "6                        0.0             0.0              0.0  ...   \n",
       "10                       0.0             0.0              0.0  ...   \n",
       "13                       0.0             0.0              0.0  ...   \n",
       "16                       0.0             0.0              0.0  ...   \n",
       "...                      ...             ...              ...  ...   \n",
       "30133                    0.0             0.0              0.0  ...   \n",
       "30139                    0.0             0.0              0.0  ...   \n",
       "30140                    0.0             0.0              0.0  ...   \n",
       "30142                    0.0             0.0              0.0  ...   \n",
       "30143                    0.0             0.0              0.0  ...   \n",
       "\n",
       "          company_location[WeWork_NYC Area]  \\\n",
       "entry_id                                      \n",
       "4                                       0.0   \n",
       "6                                       0.0   \n",
       "10                                      0.0   \n",
       "13                                      0.0   \n",
       "16                                      0.0   \n",
       "...                                     ...   \n",
       "30133                                   0.0   \n",
       "30139                                   0.0   \n",
       "30140                                   0.0   \n",
       "30142                                   0.0   \n",
       "30143                                   0.0   \n",
       "\n",
       "          company_location[WeWork_SF Bay Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Workday_SF Bay Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[Yahoo_SF Bay Area]  \\\n",
       "entry_id                                        \n",
       "4                                         0.0   \n",
       "6                                         0.0   \n",
       "10                                        0.0   \n",
       "13                                        0.0   \n",
       "16                                        0.0   \n",
       "...                                       ...   \n",
       "30133                                     0.0   \n",
       "30139                                     0.0   \n",
       "30140                                     0.0   \n",
       "30142                                     0.0   \n",
       "30143                                     0.0   \n",
       "\n",
       "          company_location[Yandex_Moscow, MC, Russia]  \\\n",
       "entry_id                                                \n",
       "4                                                 0.0   \n",
       "6                                                 0.0   \n",
       "10                                                0.0   \n",
       "13                                                0.0   \n",
       "16                                                0.0   \n",
       "...                                               ...   \n",
       "30133                                             0.0   \n",
       "30139                                             0.0   \n",
       "30140                                             0.0   \n",
       "30142                                             0.0   \n",
       "30143                                             0.0   \n",
       "\n",
       "          company_location[Yelp_SF Bay Area]  \\\n",
       "entry_id                                       \n",
       "4                                        0.0   \n",
       "6                                        0.0   \n",
       "10                                       0.0   \n",
       "13                                       0.0   \n",
       "16                                       0.0   \n",
       "...                                      ...   \n",
       "30133                                    0.0   \n",
       "30139                                    0.0   \n",
       "30140                                    0.0   \n",
       "30142                                    0.0   \n",
       "30143                                    0.0   \n",
       "\n",
       "          company_location[Zillow_SF Bay Area]  \\\n",
       "entry_id                                         \n",
       "4                                          0.0   \n",
       "6                                          0.0   \n",
       "10                                         0.0   \n",
       "13                                         0.0   \n",
       "16                                         0.0   \n",
       "...                                        ...   \n",
       "30133                                      0.0   \n",
       "30139                                      0.0   \n",
       "30140                                      0.0   \n",
       "30142                                      0.0   \n",
       "30143                                      0.0   \n",
       "\n",
       "          company_location[Zillow_Seattle Area]  \\\n",
       "entry_id                                          \n",
       "4                                           0.0   \n",
       "6                                           0.0   \n",
       "10                                          0.0   \n",
       "13                                          0.0   \n",
       "16                                          0.0   \n",
       "...                                         ...   \n",
       "30133                                       0.0   \n",
       "30139                                       0.0   \n",
       "30140                                       0.0   \n",
       "30142                                       0.0   \n",
       "30143                                       0.0   \n",
       "\n",
       "          company_location[eBay_SF Bay Area]  \\\n",
       "entry_id                                       \n",
       "4                                        0.0   \n",
       "6                                        0.0   \n",
       "10                                       0.0   \n",
       "13                                       0.0   \n",
       "16                                       0.0   \n",
       "...                                      ...   \n",
       "30133                                    0.0   \n",
       "30139                                    0.0   \n",
       "30140                                    0.0   \n",
       "30142                                    0.0   \n",
       "30143                                    0.0   \n",
       "\n",
       "          company_location[eBay_Seattle Area]  \n",
       "entry_id                                       \n",
       "4                                         0.0  \n",
       "6                                         0.0  \n",
       "10                                        0.0  \n",
       "13                                        0.0  \n",
       "16                                        0.0  \n",
       "...                                       ...  \n",
       "30133                                     0.0  \n",
       "30139                                     0.0  \n",
       "30140                                     0.0  \n",
       "30142                                     0.0  \n",
       "30143                                     0.0  \n",
       "\n",
       "[14557 rows x 619 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comp_final_interactions = pd.concat([\n",
    "    comp_final, filtered_dummies_company_levels, filtered_dummies_company_locations\n",
    "], axis=1)\n",
    "comp_final_interactions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "comp_final.to_csv('data/prepped_comp_data.csv')\n",
    "comp_final_interactions.to_csv('data/prepped_comp_data_interactions.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Minor Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>total_comp</th>\n",
       "      <th>salary</th>\n",
       "      <th>stock</th>\n",
       "      <th>bonus</th>\n",
       "      <th>company</th>\n",
       "      <th>level</th>\n",
       "      <th>location</th>\n",
       "      <th>years_experience</th>\n",
       "      <th>years_company</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>578.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>387.5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Uber</td>\n",
       "      <td>Senior</td>\n",
       "      <td>SF Bay Area</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>173.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Vancouver, BC, Canada</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>190.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Amazon</td>\n",
       "      <td>L5</td>\n",
       "      <td>Seattle Area</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>156.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>62</td>\n",
       "      <td>Seattle Area</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>201.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>63</td>\n",
       "      <td>Seattle Area</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>30133</th>\n",
       "      <td>115.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>IBM</td>\n",
       "      <td>Advisory Engineer</td>\n",
       "      <td>Madison, WI</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30139</th>\n",
       "      <td>370.0</td>\n",
       "      <td>187.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>Senior SWE</td>\n",
       "      <td>NYC Area</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>iOS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30140</th>\n",
       "      <td>212.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>Bloomberg</td>\n",
       "      <td>Senior Software Engineer</td>\n",
       "      <td>NYC Area</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Distributed Systems (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30142</th>\n",
       "      <td>185.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>Google</td>\n",
       "      <td>L4</td>\n",
       "      <td>London, EN, United Kingdom</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>API Development (Back-End)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30143</th>\n",
       "      <td>180.0</td>\n",
       "      <td>110.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Broadcom</td>\n",
       "      <td>ICB 3</td>\n",
       "      <td>Tel Aviv, TA, Israel</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Networking</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14557 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          total_comp  salary  stock  bonus    company  \\\n",
       "entry_id                                                \n",
       "4              578.0   180.0  387.5    9.0       Uber   \n",
       "6              173.0   120.0    0.0   53.0     Amazon   \n",
       "10             190.0   110.0   80.0    0.0     Amazon   \n",
       "13             156.0   135.0    8.0   13.0  Microsoft   \n",
       "16             201.0   157.0   26.0   28.0  Microsoft   \n",
       "...              ...     ...    ...    ...        ...   \n",
       "30133          115.0   115.0    0.0    0.0        IBM   \n",
       "30139          370.0   187.0  150.0   33.0    Twitter   \n",
       "30140          212.0   177.0    0.0   35.0  Bloomberg   \n",
       "30142          185.0    98.0   57.0   30.0     Google   \n",
       "30143          180.0   110.0   55.0   15.0   Broadcom   \n",
       "\n",
       "                             level                    location  \\\n",
       "entry_id                                                         \n",
       "4                           Senior                 SF Bay Area   \n",
       "6                               L5       Vancouver, BC, Canada   \n",
       "10                              L5                Seattle Area   \n",
       "13                              62                Seattle Area   \n",
       "16                              63                Seattle Area   \n",
       "...                            ...                         ...   \n",
       "30133            Advisory Engineer                 Madison, WI   \n",
       "30139                   Senior SWE                    NYC Area   \n",
       "30140     Senior Software Engineer                    NYC Area   \n",
       "30142                           L4  London, EN, United Kingdom   \n",
       "30143                        ICB 3        Tel Aviv, TA, Israel   \n",
       "\n",
       "          years_experience  years_company                             tag  \n",
       "entry_id                                                                   \n",
       "4                     10.0            2.0                                  \n",
       "6                     11.0            1.0                                  \n",
       "10                     3.0            3.0                                  \n",
       "13                     4.0            4.0                                  \n",
       "16                    12.0            6.0                                  \n",
       "...                    ...            ...                             ...  \n",
       "30133                  6.0            1.0  Distributed Systems (Back-End)  \n",
       "30139                 10.0            6.0                             iOS  \n",
       "30140                  2.0            0.0  Distributed Systems (Back-End)  \n",
       "30142                  4.0            2.0      API Development (Back-End)  \n",
       "30143                  7.0            7.0                      Networking  \n",
       "\n",
       "[14557 rows x 10 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comp.location = clean_locations\n",
    "comp.company = clean_companies\n",
    "comp.tag = clean_tags\n",
    "comp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def boxplot_sorted(df, by, column, above = 0):\n",
    "    df2 = pd.DataFrame({col:vals[column] for col, vals in df.groupby(by)})\n",
    "    meds = df2.median().sort_values()\n",
    "    return df2[meds[meds > above].index].boxplot(rot=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "salary_plot = boxplot_sorted(comp, ['company'], 'salary', 150)\n",
    "salary_plot = salary_plot.set_ylim([125, 600])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "stock_plot = boxplot_sorted(comp, ['company'], 'stock', 50)\n",
    "stock_plot = stock_plot.set_ylim([25, 500])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bonus_plot = boxplot_sorted(comp, ['company'], 'bonus', 15)\n",
    "bonus_plot = bonus_plot.set_ylim([0, 350])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "total_comp_plot = boxplot_sorted(comp, ['company'], 'total_comp', 200)\n",
    "total_comp_plot = total_comp_plot.set_ylim([150, 1000])"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}