feature_engineering/dirty_cat_example/employee_salaries.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": "[![View on GitHub](https://img.shields.io/badge/GitHub-View_on_GitHub-blue?logo=GitHub)](https://github.com/khuyentran1401/Data-science/blob/master/feature_engineering/dirty_cat_example/employee_salaries.ipynb)\n\n[<img src=\"https://deepnote.com/buttons/launch-in-deepnote.svg\">](https://deepnote.com/project/Data-science-hxlyJpi-QrKFJziQgoMSmQ/%2FData-science%2Ffeature_engineering%2Fdirty_cat_example%2Femployee_salaries.ipynb)",
   "metadata": {
    "tags": [],
    "cell_id": "00000-0d1bb351-b2b9-452e-ac61-71c220e1516a",
    "deepnote_cell_type": "markdown"
   },
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "source": "!pip install dirty_cat==0.1.0",
   "metadata": {
    "tags": [],
    "cell_id": "00001-9e493fdd-47cd-4063-8e06-70a9d8a4024c",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "d55a34bc",
    "execution_start": 1634005693783,
    "execution_millis": 3398,
    "output_cleared": true,
    "deepnote_cell_type": "code"
   },
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "source": "%load_ext watermark\n%watermark --iversions",
   "metadata": {
    "tags": [],
    "cell_id": "00002-be7d30eb-eb94-4f08-bf02-f8563ea79e5d",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "2c397fdc",
    "execution_start": 1634005803466,
    "execution_millis": 17,
    "deepnote_cell_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "text": "The watermark extension is already loaded. To reload it, use:\n  %reload_ext watermark\npandas  : 1.2.5\nnumpy   : 1.19.5\nseaborn : 0.11.2\npsycopg2: 2.9.1 (dt dec pq3 ext lo64)\n\n",
     "output_type": "stream"
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "# Import data",
   "metadata": {
    "cell_id": "00000-dae778d4-343f-4a7e-aab7-f262f56edb2c",
    "deepnote_cell_type": "markdown"
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.345845Z",
     "start_time": "2021-05-02T23:10:27.843477Z"
    },
    "cell_id": "00001-a6f27d5f-01f5-496c-875f-a33e813f4358",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "21ff7199",
    "execution_start": 1634005796229,
    "execution_millis": 1,
    "deepnote_cell_type": "code"
   },
   "source": "from dirty_cat.datasets import fetch_employee_salaries\nfrom dirty_cat import SimilarityEncoder\n\nimport pandas as pd\nimport numpy as np\nimport seaborn as sns",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.648254Z",
     "start_time": "2021-05-02T23:10:30.347346Z"
    },
    "cell_id": "00002-439b9e49-8259-4b11-9b7e-0aeced2b9c51",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "2a0fa147",
    "execution_start": 1634005703077,
    "execution_millis": 2444,
    "deepnote_cell_type": "code"
   },
   "source": "df = fetch_employee_salaries()[\"data\"]",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.680101Z",
     "start_time": "2021-05-02T23:10:30.651576Z"
    },
    "cell_id": "00003-6dcadf35-8c24-47ba-91e3-020fa49738d0",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "f30f989a",
    "execution_start": 1634005705532,
    "execution_millis": 92,
    "deepnote_cell_type": "code"
   },
   "source": "df.head(10)",
   "outputs": [
    {
     "output_type": "execute_result",
     "execution_count": 5,
     "data": {
      "application/vnd.deepnote.dataframe.v2+json": {
       "row_count": 10,
       "column_count": 13,
       "columns": [
        {
         "name": "full_name",
         "dtype": "object",
         "stats": {
          "unique_count": 10,
          "nan_count": 0,
          "categories": [
           {
            "name": "Aarhus, Pam J.",
            "count": 1
           },
           {
            "name": "Aaron, David J.",
            "count": 1
           },
           {
            "name": "8 others",
            "count": 8
           }
          ]
         }
        },
        {
         "name": "gender",
         "dtype": "category",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "categories": [
           {
            "name": "M",
            "count": 6
           },
           {
            "name": "F",
            "count": 4
           }
          ]
         }
        },
        {
         "name": "2016_gross_pay_received",
         "dtype": "float64",
         "stats": {
          "unique_count": 9,
          "nan_count": 1,
          "min": "27820.02",
          "max": "107000.24",
          "histogram": [
           {
            "bin_start": 27820.02,
            "bin_end": 35738.042,
            "count": 1
           },
           {
            "bin_start": 35738.042,
            "bin_end": 43656.064,
            "count": 0
           },
           {
            "bin_start": 43656.064,
            "bin_end": 51574.085999999996,
            "count": 1
           },
           {
            "bin_start": 51574.085999999996,
            "bin_end": 59492.108,
            "count": 1
           },
           {
            "bin_start": 59492.108,
            "bin_end": 67410.13,
            "count": 1
           },
           {
            "bin_start": 67410.13,
            "bin_end": 75328.152,
            "count": 1
           },
           {
            "bin_start": 75328.152,
            "bin_end": 83246.174,
            "count": 1
           },
           {
            "bin_start": 83246.174,
            "bin_end": 91164.196,
            "count": 0
           },
           {
            "bin_start": 91164.196,
            "bin_end": 99082.21800000001,
            "count": 1
           },
           {
            "bin_start": 99082.21800000001,
            "bin_end": 107000.24,
            "count": 2
           }
          ]
         }
        },
        {
         "name": "2016_overtime_pay",
         "dtype": "float64",
         "stats": {
          "unique_count": 6,
          "nan_count": 4,
          "min": "275.03",
          "max": "9111.09",
          "histogram": [
           {
            "bin_start": 275.03,
            "bin_end": 1158.636,
            "count": 2
           },
           {
            "bin_start": 1158.636,
            "bin_end": 2042.242,
            "count": 1
           },
           {
            "bin_start": 2042.242,
            "bin_end": 2925.848,
            "count": 0
           },
           {
            "bin_start": 2925.848,
            "bin_end": 3809.4539999999997,
            "count": 2
           },
           {
            "bin_start": 3809.4539999999997,
            "bin_end": 4693.0599999999995,
            "count": 0
           },
           {
            "bin_start": 4693.0599999999995,
            "bin_end": 5576.666,
            "count": 0
           },
           {
            "bin_start": 5576.666,
            "bin_end": 6460.272,
            "count": 0
           },
           {
            "bin_start": 6460.272,
            "bin_end": 7343.878,
            "count": 0
           },
           {
            "bin_start": 7343.878,
            "bin_end": 8227.484,
            "count": 0
           },
           {
            "bin_start": 8227.484,
            "bin_end": 9111.09,
            "count": 1
           }
          ]
         }
        },
        {
         "name": "department",
         "dtype": "category",
         "stats": {
          "unique_count": 5,
          "nan_count": 0,
          "categories": [
           {
            "name": "POL",
            "count": 4
           },
           {
            "name": "HHS",
            "count": 2
           },
           {
            "name": "3 others",
            "count": 4
           }
          ]
         }
        },
        {
         "name": "department_name",
         "dtype": "category",
         "stats": {
          "unique_count": 5,
          "nan_count": 0,
          "categories": [
           {
            "name": "Department of Police",
            "count": 4
           },
           {
            "name": "Department of Health and Human Services",
            "count": 2
           },
           {
            "name": "3 others",
            "count": 4
           }
          ]
         }
        },
        {
         "name": "division",
         "dtype": "object",
         "stats": {
          "unique_count": 10,
          "nan_count": 0,
          "categories": [
           {
            "name": "MSB Information Mgmt and Tech Division Records Management Section",
            "count": 1
           },
           {
            "name": "ISB Major Crimes Division Fugitive Section",
            "count": 1
           },
           {
            "name": "8 others",
            "count": 8
           }
          ]
         }
        },
        {
         "name": "assignment_category",
         "dtype": "category",
         "stats": {
          "unique_count": 1,
          "nan_count": 0,
          "categories": [
           {
            "name": "Fulltime-Regular",
            "count": 10
           }
          ]
         }
        },
        {
         "name": "employee_position_title",
         "dtype": "object",
         "stats": {
          "unique_count": 10,
          "nan_count": 0,
          "categories": [
           {
            "name": "Office Services Coordinator",
            "count": 1
           },
           {
            "name": "Master Police Officer",
            "count": 1
           },
           {
            "name": "8 others",
            "count": 8
           }
          ]
         }
        },
        {
         "name": "underfilled_job_title",
         "dtype": "object",
         "stats": {
          "unique_count": 1,
          "nan_count": 9,
          "categories": [
           {
            "name": "Firefighter/Rescuer I (Recruit)",
            "count": 1
           },
           {
            "name": "Missing",
            "count": 9
           }
          ]
         }
        },
        {
         "name": "date_first_hired",
         "dtype": "object",
         "stats": {
          "unique_count": 10,
          "nan_count": 0,
          "categories": [
           {
            "name": "09/22/1986",
            "count": 1
           },
           {
            "name": "09/12/1988",
            "count": 1
           },
           {
            "name": "8 others",
            "count": 8
           }
          ]
         }
        },
        {
         "name": "year_first_hired",
         "dtype": "float64",
         "stats": {
          "unique_count": 6,
          "nan_count": 0,
          "min": "1986.0",
          "max": "2016.0",
          "histogram": [
           {
            "bin_start": 1986,
            "bin_end": 1989,
            "count": 2
           },
           {
            "bin_start": 1989,
            "bin_end": 1992,
            "count": 1
           },
           {
            "bin_start": 1992,
            "bin_end": 1995,
            "count": 0
           },
           {
            "bin_start": 1995,
            "bin_end": 1998,
            "count": 0
           },
           {
            "bin_start": 1998,
            "bin_end": 2001,
            "count": 0
           },
           {
            "bin_start": 2001,
            "bin_end": 2004,
            "count": 0
           },
           {
            "bin_start": 2004,
            "bin_end": 2007,
            "count": 0
           },
           {
            "bin_start": 2007,
            "bin_end": 2010,
            "count": 3
           },
           {
            "bin_start": 2010,
            "bin_end": 2013,
            "count": 0
           },
           {
            "bin_start": 2013,
            "bin_end": 2016,
            "count": 4
           }
          ]
         }
        },
        {
         "name": "Current Annual Salary",
         "dtype": "float64",
         "stats": {
          "unique_count": 10,
          "nan_count": 0,
          "min": "45261.0",
          "max": "104717.28",
          "histogram": [
           {
            "bin_start": 45261,
            "bin_end": 51206.628,
            "count": 2
           },
           {
            "bin_start": 51206.628,
            "bin_end": 57152.256,
            "count": 1
           },
           {
            "bin_start": 57152.256,
            "bin_end": 63097.884,
            "count": 1
           },
           {
            "bin_start": 63097.884,
            "bin_end": 69043.512,
            "count": 1
           },
           {
            "bin_start": 69043.512,
            "bin_end": 74989.14,
            "count": 2
           },
           {
            "bin_start": 74989.14,
            "bin_end": 80934.768,
            "count": 0
           },
           {
            "bin_start": 80934.768,
            "bin_end": 86880.39600000001,
            "count": 0
           },
           {
            "bin_start": 86880.39600000001,
            "bin_end": 92826.024,
            "count": 0
           },
           {
            "bin_start": 92826.024,
            "bin_end": 98771.652,
            "count": 2
           },
           {
            "bin_start": 98771.652,
            "bin_end": 104717.28,
            "count": 1
           }
          ]
         }
        },
        {
         "name": "_deepnote_index_column",
         "dtype": "int64"
        }
       ],
       "rows_top": [
        {
         "full_name": "Aarhus, Pam J.",
         "gender": "F",
         "2016_gross_pay_received": 71225.98,
         "2016_overtime_pay": 416.1,
         "department": "POL",
         "department_name": "Department of Police",
         "division": "MSB Information Mgmt and Tech Division Records Management Section",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Office Services Coordinator",
         "underfilled_job_title": "None",
         "date_first_hired": "09/22/1986",
         "year_first_hired": 1986,
         "Current Annual Salary": 69222.18,
         "_deepnote_index_column": 0
        },
        {
         "full_name": "Aaron, David J.",
         "gender": "M",
         "2016_gross_pay_received": 103088.48,
         "2016_overtime_pay": 3326.19,
         "department": "POL",
         "department_name": "Department of Police",
         "division": "ISB Major Crimes Division Fugitive Section",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Master Police Officer",
         "underfilled_job_title": "None",
         "date_first_hired": "09/12/1988",
         "year_first_hired": 1988,
         "Current Annual Salary": 97392.47,
         "_deepnote_index_column": 1
        },
        {
         "full_name": "Aaron, Marsha M.",
         "gender": "F",
         "2016_gross_pay_received": 107000.24,
         "2016_overtime_pay": 1353.32,
         "department": "HHS",
         "department_name": "Department of Health and Human Services",
         "division": "Adult Protective and Case Management Services",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Social Worker IV",
         "underfilled_job_title": "None",
         "date_first_hired": "11/19/1989",
         "year_first_hired": 1989,
         "Current Annual Salary": 104717.28,
         "_deepnote_index_column": 2
        },
        {
         "full_name": "Ababio, Godfred A.",
         "gender": "M",
         "2016_gross_pay_received": 57819.04,
         "2016_overtime_pay": 3423.07,
         "department": "COR",
         "department_name": "Correction and Rehabilitation",
         "division": "PRRS Facility and Security",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Resident Supervisor II",
         "underfilled_job_title": "None",
         "date_first_hired": "05/05/2014",
         "year_first_hired": 2014,
         "Current Annual Salary": 52734.57,
         "_deepnote_index_column": 3
        },
        {
         "full_name": "Ababu, Essayas",
         "gender": "M",
         "2016_gross_pay_received": 95815.17,
         "2016_overtime_pay": "nan",
         "department": "HCA",
         "department_name": "Department of Housing and Community Affairs",
         "division": "Affordable Housing Programs",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Planning Specialist III",
         "underfilled_job_title": "None",
         "date_first_hired": "03/05/2007",
         "year_first_hired": 2007,
         "Current Annual Salary": 93396,
         "_deepnote_index_column": 4
        },
        {
         "full_name": "Abbamonte, Drew B.",
         "gender": "M",
         "2016_gross_pay_received": 80926.43,
         "2016_overtime_pay": 9111.09,
         "department": "POL",
         "department_name": "Department of Police",
         "division": "PSB 6th District Special Assignment Team",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Police Officer III",
         "underfilled_job_title": "None",
         "date_first_hired": "07/16/2007",
         "year_first_hired": 2007,
         "Current Annual Salary": 70435,
         "_deepnote_index_column": 5
        },
        {
         "full_name": "Abdalla, Eiman M.",
         "gender": "F",
         "2016_gross_pay_received": 27820.02,
         "2016_overtime_pay": "nan",
         "department": "FRS",
         "department_name": "Fire and Rescue Services",
         "division": "EMS Billing",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Accountant/Auditor II",
         "underfilled_job_title": "None",
         "date_first_hired": "06/27/2016",
         "year_first_hired": 2016,
         "Current Annual Salary": 60300,
         "_deepnote_index_column": 6
        },
        {
         "full_name": "Abdelmoniem, Marwan M.",
         "gender": "M",
         "2016_gross_pay_received": 62588.67,
         "2016_overtime_pay": "nan",
         "department": "HHS",
         "department_name": "Department of Health and Human Services",
         "division": "Head Start",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Administrative Specialist II",
         "underfilled_job_title": "None",
         "date_first_hired": "11/17/2014",
         "year_first_hired": 2014,
         "Current Annual Salary": 64788.65,
         "_deepnote_index_column": 7
        },
        {
         "full_name": "Abdool, Daniel Z.",
         "gender": "M",
         "2016_gross_pay_received": "nan",
         "2016_overtime_pay": "nan",
         "department": "FRS",
         "department_name": "Fire and Rescue Services",
         "division": "Recruit Training",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Firefighter/Rescuer III",
         "underfilled_job_title": "Firefighter/Rescuer I (Recruit)",
         "date_first_hired": "12/12/2016",
         "year_first_hired": 2016,
         "Current Annual Salary": 45261,
         "_deepnote_index_column": 8
        },
        {
         "full_name": "Abdul-Ghani, Hasinah J.",
         "gender": "F",
         "2016_gross_pay_received": 49336.12,
         "2016_overtime_pay": 275.03,
         "department": "POL",
         "department_name": "Department of Police",
         "division": "FSB Traffic Division Automated Traffic Enforcement Section",
         "assignment_category": "Fulltime-Regular",
         "employee_position_title": "Police Aide",
         "underfilled_job_title": "None",
         "date_first_hired": "02/05/2007",
         "year_first_hired": 2007,
         "Current Annual Salary": 47670.09,
         "_deepnote_index_column": 9
        }
       ],
       "rows_bottom": null
      },
      "text/plain": "                 full_name gender  2016_gross_pay_received  2016_overtime_pay  \\\n0           Aarhus, Pam J.      F                 71225.98             416.10   \n1          Aaron, David J.      M                103088.48            3326.19   \n2         Aaron, Marsha M.      F                107000.24            1353.32   \n3       Ababio, Godfred A.      M                 57819.04            3423.07   \n4           Ababu, Essayas      M                 95815.17                NaN   \n5       Abbamonte, Drew B.      M                 80926.43            9111.09   \n6        Abdalla, Eiman M.      F                 27820.02                NaN   \n7   Abdelmoniem, Marwan M.      M                 62588.67                NaN   \n8        Abdool, Daniel Z.      M                      NaN                NaN   \n9  Abdul-Ghani, Hasinah J.      F                 49336.12             275.03   \n\n  department                              department_name  \\\n0        POL                         Department of Police   \n1        POL                         Department of Police   \n2        HHS      Department of Health and Human Services   \n3        COR                Correction and Rehabilitation   \n4        HCA  Department of Housing and Community Affairs   \n5        POL                         Department of Police   \n6        FRS                     Fire and Rescue Services   \n7        HHS      Department of Health and Human Services   \n8        FRS                     Fire and Rescue Services   \n9        POL                         Department of Police   \n\n                                            division assignment_category  \\\n0  MSB Information Mgmt and Tech Division Records...    Fulltime-Regular   \n1         ISB Major Crimes Division Fugitive Section    Fulltime-Regular   \n2      Adult Protective and Case Management Services    Fulltime-Regular   \n3                         PRRS Facility and Security    Fulltime-Regular   \n4                        Affordable Housing Programs    Fulltime-Regular   \n5           PSB 6th District Special Assignment Team    Fulltime-Regular   \n6                                        EMS Billing    Fulltime-Regular   \n7                                         Head Start    Fulltime-Regular   \n8                                   Recruit Training    Fulltime-Regular   \n9  FSB Traffic Division Automated Traffic Enforce...    Fulltime-Regular   \n\n        employee_position_title            underfilled_job_title  \\\n0   Office Services Coordinator                             None   \n1         Master Police Officer                             None   \n2              Social Worker IV                             None   \n3        Resident Supervisor II                             None   \n4       Planning Specialist III                             None   \n5            Police Officer III                             None   \n6         Accountant/Auditor II                             None   \n7  Administrative Specialist II                             None   \n8       Firefighter/Rescuer III  Firefighter/Rescuer I (Recruit)   \n9                   Police Aide                             None   \n\n  date_first_hired  year_first_hired  Current Annual Salary  \n0       09/22/1986            1986.0               69222.18  \n1       09/12/1988            1988.0               97392.47  \n2       11/19/1989            1989.0              104717.28  \n3       05/05/2014            2014.0               52734.57  \n4       03/05/2007            2007.0               93396.00  \n5       07/16/2007            2007.0               70435.00  \n6       06/27/2016            2016.0               60300.00  \n7       11/17/2014            2014.0               64788.65  \n8       12/12/2016            2016.0               45261.00  \n9       02/05/2007            2007.0               47670.09  ",
      "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>full_name</th>\n      <th>gender</th>\n      <th>2016_gross_pay_received</th>\n      <th>2016_overtime_pay</th>\n      <th>department</th>\n      <th>department_name</th>\n      <th>division</th>\n      <th>assignment_category</th>\n      <th>employee_position_title</th>\n      <th>underfilled_job_title</th>\n      <th>date_first_hired</th>\n      <th>year_first_hired</th>\n      <th>Current Annual Salary</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Aarhus, Pam J.</td>\n      <td>F</td>\n      <td>71225.98</td>\n      <td>416.10</td>\n      <td>POL</td>\n      <td>Department of Police</td>\n      <td>MSB Information Mgmt and Tech Division Records...</td>\n      <td>Fulltime-Regular</td>\n      <td>Office Services Coordinator</td>\n      <td>None</td>\n      <td>09/22/1986</td>\n      <td>1986.0</td>\n      <td>69222.18</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Aaron, David J.</td>\n      <td>M</td>\n      <td>103088.48</td>\n      <td>3326.19</td>\n      <td>POL</td>\n      <td>Department of Police</td>\n      <td>ISB Major Crimes Division Fugitive Section</td>\n      <td>Fulltime-Regular</td>\n      <td>Master Police Officer</td>\n      <td>None</td>\n      <td>09/12/1988</td>\n      <td>1988.0</td>\n      <td>97392.47</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Aaron, Marsha M.</td>\n      <td>F</td>\n      <td>107000.24</td>\n      <td>1353.32</td>\n      <td>HHS</td>\n      <td>Department of Health and Human Services</td>\n      <td>Adult Protective and Case Management Services</td>\n      <td>Fulltime-Regular</td>\n      <td>Social Worker IV</td>\n      <td>None</td>\n      <td>11/19/1989</td>\n      <td>1989.0</td>\n      <td>104717.28</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Ababio, Godfred A.</td>\n      <td>M</td>\n      <td>57819.04</td>\n      <td>3423.07</td>\n      <td>COR</td>\n      <td>Correction and Rehabilitation</td>\n      <td>PRRS Facility and Security</td>\n      <td>Fulltime-Regular</td>\n      <td>Resident Supervisor II</td>\n      <td>None</td>\n      <td>05/05/2014</td>\n      <td>2014.0</td>\n      <td>52734.57</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Ababu, Essayas</td>\n      <td>M</td>\n      <td>95815.17</td>\n      <td>NaN</td>\n      <td>HCA</td>\n      <td>Department of Housing and Community Affairs</td>\n      <td>Affordable Housing Programs</td>\n      <td>Fulltime-Regular</td>\n      <td>Planning Specialist III</td>\n      <td>None</td>\n      <td>03/05/2007</td>\n      <td>2007.0</td>\n      <td>93396.00</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>Abbamonte, Drew B.</td>\n      <td>M</td>\n      <td>80926.43</td>\n      <td>9111.09</td>\n      <td>POL</td>\n      <td>Department of Police</td>\n      <td>PSB 6th District Special Assignment Team</td>\n      <td>Fulltime-Regular</td>\n      <td>Police Officer III</td>\n      <td>None</td>\n      <td>07/16/2007</td>\n      <td>2007.0</td>\n      <td>70435.00</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>Abdalla, Eiman M.</td>\n      <td>F</td>\n      <td>27820.02</td>\n      <td>NaN</td>\n      <td>FRS</td>\n      <td>Fire and Rescue Services</td>\n      <td>EMS Billing</td>\n      <td>Fulltime-Regular</td>\n      <td>Accountant/Auditor II</td>\n      <td>None</td>\n      <td>06/27/2016</td>\n      <td>2016.0</td>\n      <td>60300.00</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>Abdelmoniem, Marwan M.</td>\n      <td>M</td>\n      <td>62588.67</td>\n      <td>NaN</td>\n      <td>HHS</td>\n      <td>Department of Health and Human Services</td>\n      <td>Head Start</td>\n      <td>Fulltime-Regular</td>\n      <td>Administrative Specialist II</td>\n      <td>None</td>\n      <td>11/17/2014</td>\n      <td>2014.0</td>\n      <td>64788.65</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>Abdool, Daniel Z.</td>\n      <td>M</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>FRS</td>\n      <td>Fire and Rescue Services</td>\n      <td>Recruit Training</td>\n      <td>Fulltime-Regular</td>\n      <td>Firefighter/Rescuer III</td>\n      <td>Firefighter/Rescuer I (Recruit)</td>\n      <td>12/12/2016</td>\n      <td>2016.0</td>\n      <td>45261.00</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>Abdul-Ghani, Hasinah J.</td>\n      <td>F</td>\n      <td>49336.12</td>\n      <td>275.03</td>\n      <td>POL</td>\n      <td>Department of Police</td>\n      <td>FSB Traffic Division Automated Traffic Enforce...</td>\n      <td>Fulltime-Regular</td>\n      <td>Police Aide</td>\n      <td>None</td>\n      <td>02/05/2007</td>\n      <td>2007.0</td>\n      <td>47670.09</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {}
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.720884Z",
     "start_time": "2021-05-02T23:10:30.681151Z"
    },
    "cell_id": "00004-d73ca8b9-75ce-4324-896b-a98f7b4e2f15",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "6ba4a766",
    "execution_start": 1634005707139,
    "execution_millis": 44,
    "deepnote_cell_type": "code"
   },
   "source": "df[\"Date First Hired\"] = pd.to_datetime(df[\"date_first_hired\"])\ndf[\"Year First Hired\"] = df[\"Date First Hired\"].apply(lambda x: x.year)\n# drop rows with NaN in gender\ndf.dropna(subset=[\"gender\"], inplace=True)",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "# Encoding",
   "metadata": {
    "cell_id": "00005-834bd0e3-c8be-4b32-bd27-694ecb6c50da",
    "deepnote_cell_type": "markdown"
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.724677Z",
     "start_time": "2021-05-02T23:10:30.721798Z"
    },
    "cell_id": "00006-863d6902-569b-49de-bdcd-098f5671ef43",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "dd8d9e17",
    "execution_start": 1634005710682,
    "execution_millis": 31,
    "deepnote_cell_type": "code"
   },
   "source": "dirty_column = \"employee_position_title\"\nX_dirty = df[dirty_column].values\nsample = X_dirty[:10]\nsample",
   "outputs": [
    {
     "output_type": "execute_result",
     "execution_count": 7,
     "data": {
      "text/plain": "array(['Office Services Coordinator', 'Master Police Officer',\n       'Social Worker IV', 'Resident Supervisor II',\n       'Planning Specialist III', 'Police Officer III',\n       'Accountant/Auditor II', 'Administrative Specialist II',\n       'Firefighter/Rescuer III', 'Police Aide'], dtype=object)"
     },
     "metadata": {}
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "## One-hot Encoding",
   "metadata": {
    "cell_id": "00007-06abab3a-3df6-40cd-83e3-326bbfc14bf8",
    "deepnote_cell_type": "markdown"
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.727552Z",
     "start_time": "2021-05-02T23:10:30.725644Z"
    },
    "cell_id": "00008-2f06e7d0-deaa-4568-8d3b-9d46c7129b01",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "e1e6f593",
    "execution_start": 1634005713347,
    "execution_millis": 2,
    "deepnote_cell_type": "code"
   },
   "source": "from sklearn.preprocessing import OneHotEncoder",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.733961Z",
     "start_time": "2021-05-02T23:10:30.729479Z"
    },
    "cell_id": "00009-eb404dc6-5292-4567-89bd-73e0dcc8ecc5",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "ba922db5",
    "execution_start": 1634005713963,
    "execution_millis": 15,
    "deepnote_cell_type": "code"
   },
   "source": "enc = OneHotEncoder(sparse=False)\nonehot = enc.fit_transform(sample.reshape(-1, 1))\nonehot",
   "outputs": [
    {
     "output_type": "execute_result",
     "execution_count": 9,
     "data": {
      "text/plain": "array([[0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n       [0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n       [0., 0., 0., 0., 0., 0., 0., 0., 0., 1.],\n       [0., 0., 0., 0., 0., 0., 0., 0., 1., 0.],\n       [0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n       [0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n       [1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n       [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n       [0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n       [0., 0., 0., 0., 0., 0., 1., 0., 0., 0.]])"
     },
     "metadata": {}
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.738014Z",
     "start_time": "2021-05-02T23:10:30.734974Z"
    },
    "cell_id": "00010-8f893c8e-cbfb-4eec-95a0-7966b937e6ac",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "fbcbca79",
    "execution_start": 1634005715732,
    "execution_millis": 6,
    "deepnote_cell_type": "code"
   },
   "source": "onehot_df = pd.DataFrame(onehot, index=sample.ravel())",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.753533Z",
     "start_time": "2021-05-02T23:10:30.738932Z"
    },
    "cell_id": "00011-50f892ae-3017-4ff0-9204-febfaa50c1fd",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "fcb8b8d1",
    "execution_start": 1634005716233,
    "execution_millis": 43,
    "deepnote_cell_type": "code"
   },
   "source": "onehot_df",
   "outputs": [
    {
     "output_type": "execute_result",
     "execution_count": 11,
     "data": {
      "application/vnd.deepnote.dataframe.v2+json": {
       "row_count": 10,
       "column_count": 10,
       "columns": [
        {
         "name": 0,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 1,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 2,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 3,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 4,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 5,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 6,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 7,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 8,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": 9,
         "dtype": "float64",
         "stats": {
          "unique_count": 2,
          "nan_count": 0,
          "min": "0.0",
          "max": "1.0",
          "histogram": [
           {
            "bin_start": 0,
            "bin_end": 0.1,
            "count": 9
           },
           {
            "bin_start": 0.1,
            "bin_end": 0.2,
            "count": 0
           },
           {
            "bin_start": 0.2,
            "bin_end": 0.30000000000000004,
            "count": 0
           },
           {
            "bin_start": 0.30000000000000004,
            "bin_end": 0.4,
            "count": 0
           },
           {
            "bin_start": 0.4,
            "bin_end": 0.5,
            "count": 0
           },
           {
            "bin_start": 0.5,
            "bin_end": 0.6000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.6000000000000001,
            "bin_end": 0.7000000000000001,
            "count": 0
           },
           {
            "bin_start": 0.7000000000000001,
            "bin_end": 0.8,
            "count": 0
           },
           {
            "bin_start": 0.8,
            "bin_end": 0.9,
            "count": 0
           },
           {
            "bin_start": 0.9,
            "bin_end": 1,
            "count": 1
           }
          ]
         }
        },
        {
         "name": "_deepnote_index_column",
         "dtype": "object"
        }
       ],
       "rows_top": [
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 1,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Office Services Coordinator"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 1,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Master Police Officer"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 1,
         "_deepnote_index_column": "Social Worker IV"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 1,
         "9": 0,
         "_deepnote_index_column": "Resident Supervisor II"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 1,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Planning Specialist III"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 1,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Police Officer III"
        },
        {
         "0": 1,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Accountant/Auditor II"
        },
        {
         "0": 0,
         "1": 1,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Administrative Specialist II"
        },
        {
         "0": 0,
         "1": 0,
         "2": 1,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 0,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Firefighter/Rescuer III"
        },
        {
         "0": 0,
         "1": 0,
         "2": 0,
         "3": 0,
         "4": 0,
         "5": 0,
         "6": 1,
         "7": 0,
         "8": 0,
         "9": 0,
         "_deepnote_index_column": "Police Aide"
        }
       ],
       "rows_bottom": null
      },
      "text/plain": "                                0    1    2    3    4    5    6    7    8    9\nOffice Services Coordinator   0.0  0.0  0.0  0.0  1.0  0.0  0.0  0.0  0.0  0.0\nMaster Police Officer         0.0  0.0  0.0  1.0  0.0  0.0  0.0  0.0  0.0  0.0\nSocial Worker IV              0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  1.0\nResident Supervisor II        0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  1.0  0.0\nPlanning Specialist III       0.0  0.0  0.0  0.0  0.0  1.0  0.0  0.0  0.0  0.0\nPolice Officer III            0.0  0.0  0.0  0.0  0.0  0.0  0.0  1.0  0.0  0.0\nAccountant/Auditor II         1.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0\nAdministrative Specialist II  0.0  1.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0\nFirefighter/Rescuer III       0.0  0.0  1.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0\nPolice Aide                   0.0  0.0  0.0  0.0  0.0  0.0  1.0  0.0  0.0  0.0",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>0</th>\n      <th>1</th>\n      <th>2</th>\n      <th>3</th>\n      <th>4</th>\n      <th>5</th>\n      <th>6</th>\n      <th>7</th>\n      <th>8</th>\n      <th>9</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>Office Services Coordinator</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    </tr>\n    <tr>\n      <th>Master Police Officer</th>\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    <tr>\n      <th>Social Worker IV</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    </tr>\n    <tr>\n      <th>Resident Supervisor II</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>1.0</td>\n      <td>0.0</td>\n    </tr>\n    <tr>\n      <th>Planning Specialist III</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>1.0</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>Police Officer III</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>1.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n    <tr>\n      <th>Accountant/Auditor II</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    </tr>\n    <tr>\n      <th>Administrative Specialist II</th>\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    </tr>\n    <tr>\n      <th>Firefighter/Rescuer III</th>\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    </tr>\n    <tr>\n      <th>Police Aide</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>1.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {}
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "## Similarity Encoding",
   "metadata": {
    "cell_id": "00012-25688db1-695b-4ed4-a9ec-5af1e3ab80fb",
    "deepnote_cell_type": "markdown"
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.762839Z",
     "start_time": "2021-05-02T23:10:30.754344Z"
    },
    "cell_id": "00013-9a0209f2-4e5c-430b-a4f5-894cd41b9402",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "c6a3b891",
    "execution_start": 1634005719189,
    "execution_millis": 19,
    "deepnote_cell_type": "code"
   },
   "source": "enc = SimilarityEncoder(similarity=\"ngram\")\nX_enc = enc.fit_transform(sample.reshape(-1, 1))\nX_enc",
   "outputs": [
    {
     "output_type": "execute_result",
     "execution_count": 12,
     "data": {
      "text/plain": "array([[0.05882353, 0.03125   , 0.02739726, 0.19008264, 1.        ,\n        0.01351351, 0.05555556, 0.20535714, 0.08088235, 0.032     ],\n       [0.008     , 0.02083333, 0.056     , 1.        , 0.19008264,\n        0.02325581, 0.23076923, 0.56      , 0.01574803, 0.02777778],\n       [0.03738318, 0.07317073, 0.05405405, 0.02777778, 0.032     ,\n        0.0733945 , 0.        , 0.0625    , 0.06542056, 1.        ],\n       [0.11206897, 0.07142857, 0.09756098, 0.01574803, 0.08088235,\n        0.07142857, 0.03125   , 0.08108108, 1.        , 0.06542056],\n       [0.04761905, 0.3539823 , 0.06976744, 0.02325581, 0.01351351,\n        1.        , 0.02      , 0.09821429, 0.07142857, 0.0733945 ],\n       [0.0733945 , 0.05343511, 0.14953271, 0.56      , 0.20535714,\n        0.09821429, 0.26086957, 1.        , 0.08108108, 0.0625    ],\n       [1.        , 0.05      , 0.06451613, 0.008     , 0.05882353,\n        0.04761905, 0.01052632, 0.0733945 , 0.11206897, 0.03738318],\n       [0.05      , 1.        , 0.03378378, 0.02083333, 0.03125   ,\n        0.3539823 , 0.02631579, 0.05343511, 0.07142857, 0.07317073],\n       [0.06451613, 0.03378378, 1.        , 0.056     , 0.02739726,\n        0.06976744, 0.        , 0.14953271, 0.09756098, 0.05405405],\n       [0.01052632, 0.02631579, 0.        , 0.23076923, 0.05555556,\n        0.02      , 1.        , 0.26086957, 0.03125   , 0.        ]])"
     },
     "metadata": {}
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.765123Z",
     "start_time": "2021-05-02T23:10:30.763586Z"
    },
    "cell_id": "00014-7de3a034-e387-4d2e-88e9-59deaf1109d8",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "92125e66",
    "execution_start": 1634005721311,
    "execution_millis": 2,
    "deepnote_cell_type": "code"
   },
   "source": "from sklearn.preprocessing import normalize",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:30.769022Z",
     "start_time": "2021-05-02T23:10:30.765947Z"
    },
    "cell_id": "00015-5a55f840-e52c-4c26-8020-f86f170bc915",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "669861b8",
    "execution_start": 1634005721767,
    "execution_millis": 3,
    "deepnote_cell_type": "code"
   },
   "source": "def plot_similarity(labels, features):\n    normalized_features = normalize(features)\n    corr = np.inner(normalized_features, normalized_features)\n    sns.set(font_scale=1.2)\n    g = sns.heatmap(\n        corr,\n        xticklabels=labels,\n        yticklabels=labels,\n        vmin=0,\n        vmax=1,\n        cmap=\"YlOrRd\",\n        annot=True,\n        annot_kws={\"size\": 10},\n    )\n    g.set_xticklabels(labels, rotation=90)\n    g.set_title(\"Similarity\")\n\n\ndef encode_and_plot(labels):\n    enc = SimilarityEncoder(similarity=\"ngram\")\n    X_enc = enc.fit_transform(labels.reshape(-1, 1))\n    plot_similarity(labels, X_enc)",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.222503Z",
     "start_time": "2021-05-02T23:10:30.769900Z"
    },
    "cell_id": "00016-960fd1dd-baca-4585-a796-f116c4642fbc",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "923cc1db",
    "execution_start": 1634005723592,
    "execution_millis": 1282,
    "deepnote_cell_type": "code"
   },
   "source": "encode_and_plot(sample)",
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 2 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "image/png": {
       "width": 532,
       "height": 440
      }
     },
     "output_type": "display_data"
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "# One-hot Encoding vs Similarity Encoding",
   "metadata": {
    "cell_id": "00017-e48ed98c-6d23-410d-94ee-0942d1e7a301",
    "deepnote_cell_type": "markdown"
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.226829Z",
     "start_time": "2021-05-02T23:10:31.223487Z"
    },
    "cell_id": "00018-787bc39f-aa77-4720-a19b-8aeff3d4e6f8",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "382d56a4",
    "execution_start": 1634005726108,
    "execution_millis": 15,
    "deepnote_cell_type": "code"
   },
   "source": "from sklearn.preprocessing import FunctionTransformer, StandardScaler\nfrom sklearn.compose import ColumnTransformer\nfrom sklearn.pipeline import Pipeline\nfrom sklearn.model_selection import KFold, cross_val_score\nfrom sklearn.linear_model import RidgeCV",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.231019Z",
     "start_time": "2021-05-02T23:10:31.227670Z"
    },
    "cell_id": "00019-aa47f5fe-53ce-41da-bdef-1fe4688131df",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "4b3fbead",
    "execution_start": 1634005726644,
    "execution_millis": 3,
    "deepnote_cell_type": "code"
   },
   "source": "y = df[\"Current Annual Salary\"].values",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.234341Z",
     "start_time": "2021-05-02T23:10:31.231985Z"
    },
    "cell_id": "00020-5f0c8f00-27bc-4fc0-b659-37d9d55ae5f8",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "586f264a",
    "execution_start": 1634005727046,
    "execution_millis": 1,
    "deepnote_cell_type": "code"
   },
   "source": "clean_columns = {\n    \"gender\": \"one-hot\",\n    \"department_name\": \"one-hot\",\n    \"assignment_category\": \"one-hot\",\n    \"Year First Hired\": \"numerical\",\n}",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.237897Z",
     "start_time": "2021-05-02T23:10:31.235314Z"
    },
    "cell_id": "00021-a3f08138-60cb-4207-8f1c-f3414526e924",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "5d49104d",
    "execution_start": 1634005727628,
    "execution_millis": 5,
    "deepnote_cell_type": "code"
   },
   "source": "encoding_methods = [\"one-hot\", \"similarity\"]",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.241776Z",
     "start_time": "2021-05-02T23:10:31.238795Z"
    },
    "cell_id": "00022-0c5d72c7-d59d-40e1-bd41-79f63aceb793",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "1774297d",
    "execution_start": 1634005728112,
    "execution_millis": 4,
    "deepnote_cell_type": "code"
   },
   "source": "encoders_dict = {\n    \"one-hot\": OneHotEncoder(handle_unknown=\"ignore\", sparse=False),\n    \"similarity\": SimilarityEncoder(similarity=\"ngram\"),\n    \"numerical\": FunctionTransformer(None),\n}",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:31.246502Z",
     "start_time": "2021-05-02T23:10:31.243055Z"
    },
    "cell_id": "00023-e0933264-619d-4da0-94ca-ca5cd22f272c",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "935bcca0",
    "execution_start": 1634005731789,
    "execution_millis": 5,
    "deepnote_cell_type": "code"
   },
   "source": "def make_pipeline(encoding_method):\n    # static transformers from the other columns\n    transformers = [\n        (enc + \"_\" + col, encoders_dict[enc], [col])\n        for col, enc in clean_columns.items()\n    ]\n    # adding the encoded column\n    transformers += [(encoding_method, encoders_dict[encoding_method], [dirty_column])]\n    pipeline = Pipeline(\n        [\n            # Use ColumnTransformer to combine the features\n            (\"union\", ColumnTransformer(transformers=transformers, remainder=\"drop\")),\n            (\"scaler\", StandardScaler(with_mean=False)),\n            (\"clf\", RidgeCV()),\n        ]\n    )\n    return pipeline",
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-02T23:10:36.330364Z",
     "start_time": "2021-05-02T23:10:31.247314Z"
    },
    "cell_id": "00024-0d7a79e2-3976-4826-adbe-50c6ca92e146",
    "deepnote_to_be_reexecuted": false,
    "source_hash": "6faf09f3",
    "execution_start": 1634005732609,
    "execution_millis": 11677,
    "deepnote_cell_type": "code"
   },
   "source": "all_scores = dict()\n\ncv = KFold(n_splits=5, random_state=12, shuffle=True)\nscoring = \"r2\"\nfor method in encoding_methods:\n    pipeline = make_pipeline(method)\n    scores = cross_val_score(pipeline, df, y, cv=cv, scoring=scoring)\n    print(f\"{method} encoding\")\n    print(f\"{scoring} score:  mean: {np.mean(scores):.3f}; std: {np.std(scores):.3f}\\n\")\n    all_scores[method] = scores",
   "outputs": [
    {
     "name": "stdout",
     "text": "one-hot encoding\nr2 score:  mean: 0.856; std: 0.034\n\nsimilarity encoding\nr2 score:  mean: 0.915; std: 0.012\n\n",
     "output_type": "stream"
    }
   ],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "source": "<a style='text-decoration:none;line-height:16px;display:flex;color:#5B5B62;padding:10px;justify-content:end;' href='https://deepnote.com?utm_source=created-in-deepnote-cell&projectId=87197226-98be-42b2-8527-389082831299' target=\"_blank\">\n<img alt='Created in deepnote.com' style='display:inline;max-height:16px;margin:0px;margin-right:7.5px;' src='data:image/svg+xml;base64,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' > </img>\nCreated in <span style='font-weight:600;margin-left:4px;'>Deepnote</span></a>",
   "metadata": {
    "tags": [],
    "created_in_deepnote_cell": true,
    "deepnote_cell_type": "markdown"
   }
  }
 ],
 "nbformat": 4,
 "nbformat_minor": 5,
 "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": "3.8.1"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "deepnote_notebook_id": "1fe3e4d5-93bb-409e-bbeb-318096ca7076",
  "deepnote": {},
  "deepnote_execution_queue": []
 }
}