From 44051fd217a2413da6bc5dcc249de1f26f43a7f1 Mon Sep 17 00:00:00 2001 From: Alumasa Date: Mon, 21 Dec 2020 21:42:27 +0300 Subject: [PATCH 1/2] US police killings data visualization and analysis --- .gitignore | 1 + ... Police Killings Analysis-checkpoint.ipynb | 5388 +++++++++++++++++ .../US Police Killings Analysis.ipynb | 5388 +++++++++++++++++ 3 files changed, 10777 insertions(+) create mode 100644 .gitignore create mode 100644 Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb create mode 100644 Level 1/Intermediate/US Police Killings Analysis.ipynb diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..1e73c80 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +Level 1/Intermediate/.jovianrc \ No newline at end of file diff --git a/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb b/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb new file mode 100644 index 0000000..aa1d579 --- /dev/null +++ b/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb @@ -0,0 +1,5388 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Import required libraries." + ] + }, + { + "cell_type": "code", + "execution_count": 88, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import warnings\n", + "warnings.filterwarnings('ignore')\n", + "\n", + "import re\n", + "import seaborn as sns\n", + "sns.set_style(\"whitegrid\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Open and read the data sets." + ] + }, + { + "cell_type": "code", + "execution_count": 89, + "metadata": {}, + "outputs": [], + "source": [ + "income = pd.read_csv('datasets/MedianHouseholdIncome2015.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 90, + "metadata": {}, + "outputs": [], + "source": [ + "poverty = pd.read_csv('datasets/PercentagePeopleBelowPovertyLevel.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 91, + "metadata": {}, + "outputs": [], + "source": [ + "education = pd.read_csv('datasets/PercentOver25CompletedHighSchool.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 92, + "metadata": {}, + "outputs": [], + "source": [ + "killings = pd.read_csv('datasets/PoliceKillingsUS.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 93, + "metadata": {}, + "outputs": [], + "source": [ + "city_race = pd.read_csv('datasets/ShareRaceByCity.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 94, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCityMedian Income
0ALAbanda CDP11207
1ALAbbeville city25615
2ALAdamsville city42575
3ALAddison town37083
4ALAkron town21667
\n", + "
" + ], + "text/plain": [ + " Geographic Area City Median Income\n", + "0 AL Abanda CDP 11207\n", + "1 AL Abbeville city 25615\n", + "2 AL Adamsville city 42575\n", + "3 AL Addison town 37083\n", + "4 AL Akron town 21667" + ] + }, + "execution_count": 94, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "income.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 95, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "51" + ] + }, + "execution_count": 95, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "income['Geographic Area'].nunique()" + ] + }, + { + "cell_type": "code", + "execution_count": 96, + "metadata": {}, + "outputs": [], + "source": [ + "city_samp = city_race.sample(frac=0.25, replace=False, random_state=0, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": 97, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 7317 entries, 20414 to 22006\n", + "Data columns (total 7 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic area 7317 non-null object\n", + " 1 City 7317 non-null object\n", + " 2 share_white 7317 non-null object\n", + " 3 share_black 7317 non-null object\n", + " 4 share_native_american 7317 non-null object\n", + " 5 share_asian 7317 non-null object\n", + " 6 share_hispanic 7317 non-null object\n", + "dtypes: object(7)\n", + "memory usage: 457.3+ KB\n" + ] + } + ], + "source": [ + "city_samp.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 98, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0" + ] + }, + "execution_count": 98, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "city_samp.index.duplicated().sum()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 99, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29322 entries, 0 to 29321\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29322 non-null object\n", + " 1 City 29322 non-null object\n", + " 2 Median Income 29271 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.4+ KB\n" + ] + } + ], + "source": [ + "income.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 100, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypoverty_rate
0ALAbanda CDP78.8
1ALAbbeville city29.1
2ALAdamsville city25.5
3ALAddison town30.7
4ALAkron town42
\n", + "
" + ], + "text/plain": [ + " Geographic Area City poverty_rate\n", + "0 AL Abanda CDP 78.8\n", + "1 AL Abbeville city 29.1\n", + "2 AL Adamsville city 25.5\n", + "3 AL Addison town 30.7\n", + "4 AL Akron town 42" + ] + }, + "execution_count": 100, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "poverty.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 101, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29329 non-null object\n", + " 1 City 29329 non-null object\n", + " 2 poverty_rate 29329 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.5+ KB\n" + ] + } + ], + "source": [ + "poverty.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 102, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypercent_completed_hs
0ALAbanda CDP21.2
1ALAbbeville city69.1
2ALAdamsville city78.9
3ALAddison town81.4
4ALAkron town68.6
\n", + "
" + ], + "text/plain": [ + " Geographic Area City percent_completed_hs\n", + "0 AL Abanda CDP 21.2\n", + "1 AL Abbeville city 69.1\n", + "2 AL Adamsville city 78.9\n", + "3 AL Addison town 81.4\n", + "4 AL Akron town 68.6" + ] + }, + "execution_count": 102, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "education.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 103, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29329 non-null object\n", + " 1 City 29329 non-null object\n", + " 2 percent_completed_hs 29329 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.5+ KB\n" + ] + } + ], + "source": [ + "education.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 104, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot02/01/15shotgun53.0MASheltonWATrueattackNot fleeingFalse
14Lewis Lee Lembke02/01/15shotgun47.0MWAlohaORFalseattackNot fleeingFalse
25John Paul Quintero03/01/15shot and Taseredunarmed23.0MHWichitaKSFalseotherNot fleeingFalse
38Matthew Hoffman04/01/15shottoy weapon32.0MWSan FranciscoCATrueattackNot fleeingFalse
49Michael Rodriguez04/01/15shotnail gun39.0MHEvansCOFalseattackNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 02/01/15 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 02/01/15 shot gun 47.0 \n", + "2 5 John Paul Quintero 03/01/15 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 04/01/15 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 04/01/15 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True attack \n", + "1 M W Aloha OR False attack \n", + "2 M H Wichita KS False other \n", + "3 M W San Francisco CA True attack \n", + "4 M H Evans CO False attack \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 104, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 105, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 2535 entries, 0 to 2534\n", + "Data columns (total 14 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 2535 non-null int64 \n", + " 1 name 2535 non-null object \n", + " 2 date 2535 non-null object \n", + " 3 manner_of_death 2535 non-null object \n", + " 4 armed 2526 non-null object \n", + " 5 age 2458 non-null float64\n", + " 6 gender 2535 non-null object \n", + " 7 race 2340 non-null object \n", + " 8 city 2535 non-null object \n", + " 9 state 2535 non-null object \n", + " 10 signs_of_mental_illness 2535 non-null bool \n", + " 11 threat_level 2535 non-null object \n", + " 12 flee 2470 non-null object \n", + " 13 body_camera 2535 non-null bool \n", + "dtypes: bool(2), float64(1), int64(1), object(10)\n", + "memory usage: 242.7+ KB\n" + ] + } + ], + "source": [ + "killings.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 106, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic areaCityshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
0ALAbanda CDP67.230.2001.6
1ALAbbeville city54.441.40.113.1
2ALAdamsville city52.344.90.50.32.3
3ALAddison town99.10.100.10.4
4ALAkron town13.286.5000.3
\n", + "
" + ], + "text/plain": [ + " Geographic area City share_white share_black \\\n", + "0 AL Abanda CDP 67.2 30.2 \n", + "1 AL Abbeville city 54.4 41.4 \n", + "2 AL Adamsville city 52.3 44.9 \n", + "3 AL Addison town 99.1 0.1 \n", + "4 AL Akron town 13.2 86.5 \n", + "\n", + " share_native_american share_asian share_hispanic \n", + "0 0 0 1.6 \n", + "1 0.1 1 3.1 \n", + "2 0.5 0.3 2.3 \n", + "3 0 0.1 0.4 \n", + "4 0 0 0.3 " + ] + }, + "execution_count": 106, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "city_race.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 107, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29268 entries, 0 to 29267\n", + "Data columns (total 7 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic area 29268 non-null object\n", + " 1 City 29268 non-null object\n", + " 2 share_white 29268 non-null object\n", + " 3 share_black 29268 non-null object\n", + " 4 share_native_american 29268 non-null object\n", + " 5 share_asian 29268 non-null object\n", + " 6 share_hispanic 29268 non-null object\n", + "dtypes: object(7)\n", + "memory usage: 1.6+ MB\n" + ] + } + ], + "source": [ + "city_race.info()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will concatenate the `income`, `poverty`, `education` and `city_race` dataframes for compact analysis." + ] + }, + { + "cell_type": "code", + "execution_count": 108, + "metadata": {}, + "outputs": [], + "source": [ + "data = pd.concat([poverty, education, income, city_race], axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": 109, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypoverty_rateGeographic AreaCitypercent_completed_hsGeographic AreaCityMedian IncomeGeographic areaCityshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
26119TXTimpson city42.7TXTimpson city70.8TXTomball city44086TXVenus town79.413.20.51.724.8
20016OHPainesville city23OHPainesville city78.1OHParma city50440OHPleasant Run CDP79.814.50.51.82.2
15984NJEast Rutherford borough10.1NJEast Rutherford borough92.9NJEllisburg CDP61544NJFarmingdale borough89.62.90.53.26.9
14734MTCamas CDP48.6MTCamas CDP87.5MTCharlo CDP44583MTConrad city95.10.21.80.31.5
18451NCMarshville town28.9NCMarshville town71.1NCMaysville town24432NCMorganton city70.112.20.92.416.4
28940WIRichfield village2.8WIRichfield village95.8WIRiver Hills village156250WISpooner city95.10.31.90.71.3
1442ARBlue Eye town74.4ARBlue Eye town16.7ARBlue Eye town(X)ARBooneville city93.510.90.63.2
25763TXPoint Comfort city7.3TXPoint Comfort city89.2TXPortland city62561TXRamos CDP76.7000100
7761INLittle York town20.1INLittle York town77.8INLogansport city32982INLowell town95.90.50.40.36.9
1592ARGreers Ferry city13ARGreers Ferry city81.1ARGreers Ferry city31810ARHackett city920.13.40.60.6
\n", + "
" + ], + "text/plain": [ + " Geographic Area City poverty_rate Geographic Area \\\n", + "26119 TX Timpson city 42.7 TX \n", + "20016 OH Painesville city 23 OH \n", + "15984 NJ East Rutherford borough 10.1 NJ \n", + "14734 MT Camas CDP 48.6 MT \n", + "18451 NC Marshville town 28.9 NC \n", + "28940 WI Richfield village 2.8 WI \n", + "1442 AR Blue Eye town 74.4 AR \n", + "25763 TX Point Comfort city 7.3 TX \n", + "7761 IN Little York town 20.1 IN \n", + "1592 AR Greers Ferry city 13 AR \n", + "\n", + " City percent_completed_hs Geographic Area \\\n", + "26119 Timpson city 70.8 TX \n", + "20016 Painesville city 78.1 OH \n", + "15984 East Rutherford borough 92.9 NJ \n", + "14734 Camas CDP 87.5 MT \n", + "18451 Marshville town 71.1 NC \n", + "28940 Richfield village 95.8 WI \n", + "1442 Blue Eye town 16.7 AR \n", + "25763 Point Comfort city 89.2 TX \n", + "7761 Little York town 77.8 IN \n", + "1592 Greers Ferry city 81.1 AR \n", + "\n", + " City Median Income Geographic area City \\\n", + "26119 Tomball city 44086 TX Venus town \n", + "20016 Parma city 50440 OH Pleasant Run CDP \n", + "15984 Ellisburg CDP 61544 NJ Farmingdale borough \n", + "14734 Charlo CDP 44583 MT Conrad city \n", + "18451 Maysville town 24432 NC Morganton city \n", + "28940 River Hills village 156250 WI Spooner city \n", + "1442 Blue Eye town (X) AR Booneville city \n", + "25763 Portland city 62561 TX Ramos CDP \n", + "7761 Logansport city 32982 IN Lowell town \n", + "1592 Greers Ferry city 31810 AR Hackett city \n", + "\n", + " share_white share_black share_native_american share_asian share_hispanic \n", + "26119 79.4 13.2 0.5 1.7 24.8 \n", + "20016 79.8 14.5 0.5 1.8 2.2 \n", + "15984 89.6 2.9 0.5 3.2 6.9 \n", + "14734 95.1 0.2 1.8 0.3 1.5 \n", + "18451 70.1 12.2 0.9 2.4 16.4 \n", + "28940 95.1 0.3 1.9 0.7 1.3 \n", + "1442 93.5 1 0.9 0.6 3.2 \n", + "25763 76.7 0 0 0 100 \n", + "7761 95.9 0.5 0.4 0.3 6.9 \n", + "1592 92 0.1 3.4 0.6 0.6 " + ] + }, + "execution_count": 109, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.sample(10)" + ] + }, + { + "cell_type": "code", + "execution_count": 110, + "metadata": {}, + "outputs": [], + "source": [ + "#rename the columns\n", + "data.columns = ['state', 'city', 'poverty_rate', 'Geographic_Area_x', 'City_x',\n", + " 'education', 'Geographic_Area_y', 'City_y', 'income',\n", + " 'Geographic_area_z', 'City_z', 'share_white', 'share_black',\n", + " 'share_native_american', 'share_asian', 'share_hispanic']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- The cities are not the same, but the areas(state) are. In this case I will perform my analysis based on geographic area." + ] + }, + { + "cell_type": "code", + "execution_count": 111, + "metadata": {}, + "outputs": [], + "source": [ + "data.drop(['Geographic_Area_x', 'City_x', 'Geographic_Area_y', 'City_y', 'Geographic_area_z', 'City_z'], \n", + " axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 112, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationincomeshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
18204NCDellview town01002361679.317.10.302.1
10302KYWhite Plains city14.482.35733494.11.6000
20058OHPlumwood CDP6.280.35342148.544.90.14.31.8
20849OKMulhall town28.689.4-88.2010.600
23114PAUtica borough20.584.34546998.60.30.10.41
21753PACentre Hall borough694.95375098.20.300.40.9
11271MDNanticoke Acres CDP52.6505766565.3230.45.55.9
12766MNHoffman city21.887.74500094.71.10.50.72.8
21915PAEagles Mere borough2.51005682795.71.20.21.50.8
4443FLHastings town26.381.836196970.70.40.44.3
2124CACamarillo city6.492.28815296.90006.2
14244MOMaplewood city19.691.52531396.60001.7
6547ILGrand Ridge village10.894.83906383.73.30.36.88.8
14122MOIronton city36.380.62375096.90000.6
4150FLArcher city35.590.6211465838100
5427GAMineral Bluff CDP251003309765.632.1003.7
5649GAWarwick city38.564.14300039.157.70.212.5
5573GASocial Circle city12.479.62432154.440.10.223
10038KYIndependence city8.489.82187598.40.300.30.3
28846WINiagara city20.393.28303698.300.401
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education income share_white \\\n", + "18204 NC Dellview town 0 100 23616 79.3 \n", + "10302 KY White Plains city 14.4 82.3 57334 94.1 \n", + "20058 OH Plumwood CDP 6.2 80.3 53421 48.5 \n", + "20849 OK Mulhall town 28.6 89.4 - 88.2 \n", + "23114 PA Utica borough 20.5 84.3 45469 98.6 \n", + "21753 PA Centre Hall borough 6 94.9 53750 98.2 \n", + "11271 MD Nanticoke Acres CDP 52.6 50 57665 65.3 \n", + "12766 MN Hoffman city 21.8 87.7 45000 94.7 \n", + "21915 PA Eagles Mere borough 2.5 100 56827 95.7 \n", + "4443 FL Hastings town 26.3 81.8 36196 97 \n", + "2124 CA Camarillo city 6.4 92.2 88152 96.9 \n", + "14244 MO Maplewood city 19.6 91.5 25313 96.6 \n", + "6547 IL Grand Ridge village 10.8 94.8 39063 83.7 \n", + "14122 MO Ironton city 36.3 80.6 23750 96.9 \n", + "4150 FL Archer city 35.5 90.6 21146 58 \n", + "5427 GA Mineral Bluff CDP 25 100 33097 65.6 \n", + "5649 GA Warwick city 38.5 64.1 43000 39.1 \n", + "5573 GA Social Circle city 12.4 79.6 24321 54.4 \n", + "10038 KY Independence city 8.4 89.8 21875 98.4 \n", + "28846 WI Niagara city 20.3 93.2 83036 98.3 \n", + "\n", + " share_black share_native_american share_asian share_hispanic \n", + "18204 17.1 0.3 0 2.1 \n", + "10302 1.6 0 0 0 \n", + "20058 44.9 0.1 4.3 1.8 \n", + "20849 0 10.6 0 0 \n", + "23114 0.3 0.1 0.4 1 \n", + "21753 0.3 0 0.4 0.9 \n", + "11271 23 0.4 5.5 5.9 \n", + "12766 1.1 0.5 0.7 2.8 \n", + "21915 1.2 0.2 1.5 0.8 \n", + "4443 0.7 0.4 0.4 4.3 \n", + "2124 0 0 0 6.2 \n", + "14244 0 0 0 1.7 \n", + "6547 3.3 0.3 6.8 8.8 \n", + "14122 0 0 0 0.6 \n", + "4150 38 1 0 0 \n", + "5427 32.1 0 0 3.7 \n", + "5649 57.7 0.2 1 2.5 \n", + "5573 40.1 0.2 2 3 \n", + "10038 0.3 0 0.3 0.3 \n", + "28846 0 0.4 0 1 " + ] + }, + "execution_count": 112, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.sample(20)" + ] + }, + { + "cell_type": "code", + "execution_count": 113, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object\n", + " 1 city 29329 non-null object\n", + " 2 poverty_rate 29329 non-null object\n", + " 3 education 29329 non-null object\n", + " 4 income 29271 non-null object\n", + " 5 share_white 29268 non-null object\n", + " 6 share_black 29268 non-null object\n", + " 7 share_native_american 29268 non-null object\n", + " 8 share_asian 29268 non-null object\n", + " 9 share_hispanic 29268 non-null object\n", + "dtypes: object(10)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 114, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot02/01/15shotgun53.0MASheltonWATrueattackNot fleeingFalse
14Lewis Lee Lembke02/01/15shotgun47.0MWAlohaORFalseattackNot fleeingFalse
25John Paul Quintero03/01/15shot and Taseredunarmed23.0MHWichitaKSFalseotherNot fleeingFalse
38Matthew Hoffman04/01/15shottoy weapon32.0MWSan FranciscoCATrueattackNot fleeingFalse
49Michael Rodriguez04/01/15shotnail gun39.0MHEvansCOFalseattackNot fleeingFalse
.............................................
25302822Rodney E. Jacobs28/07/17shotgun31.0MNaNKansas CityMOFalseattackNot fleeingFalse
25312813TK TK28/07/17shotvehicleNaNMNaNAlbuquerqueNMFalseattackCarFalse
25322818Dennis W. Robinson29/07/17shotgun48.0MNaNMelbaIDFalseattackCarFalse
25332817Isaiah Tucker31/07/17shotvehicle28.0MBOshkoshWIFalseattackCarTrue
25342815Dwayne Jeune31/07/17shotknife32.0MBBrooklynNYTrueattackNot fleeingFalse
\n", + "

2535 rows × 14 columns

\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 02/01/15 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 02/01/15 shot gun 47.0 \n", + "2 5 John Paul Quintero 03/01/15 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 04/01/15 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 04/01/15 shot nail gun 39.0 \n", + "... ... ... ... ... ... ... \n", + "2530 2822 Rodney E. Jacobs 28/07/17 shot gun 31.0 \n", + "2531 2813 TK TK 28/07/17 shot vehicle NaN \n", + "2532 2818 Dennis W. Robinson 29/07/17 shot gun 48.0 \n", + "2533 2817 Isaiah Tucker 31/07/17 shot vehicle 28.0 \n", + "2534 2815 Dwayne Jeune 31/07/17 shot knife 32.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True attack \n", + "1 M W Aloha OR False attack \n", + "2 M H Wichita KS False other \n", + "3 M W San Francisco CA True attack \n", + "4 M H Evans CO False attack \n", + "... ... ... ... ... ... ... \n", + "2530 M NaN Kansas City MO False attack \n", + "2531 M NaN Albuquerque NM False attack \n", + "2532 M NaN Melba ID False attack \n", + "2533 M B Oshkosh WI False attack \n", + "2534 M B Brooklyn NY True attack \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False \n", + "... ... ... \n", + "2530 Not fleeing False \n", + "2531 Car False \n", + "2532 Car False \n", + "2533 Car True \n", + "2534 Not fleeing False \n", + "\n", + "[2535 rows x 14 columns]" + ] + }, + "execution_count": 114, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Data Cleaning and Prerocessing\n", + " - filling in missing values and changing data types" + ] + }, + { + "cell_type": "code", + "execution_count": 115, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "state 0\n", + "city 0\n", + "poverty_rate 0\n", + "education 0\n", + "income 58\n", + "share_white 61\n", + "share_black 61\n", + "share_native_american 61\n", + "share_asian 61\n", + "share_hispanic 61\n", + "dtype: int64" + ] + }, + "execution_count": 115, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#check for missing values\n", + "data.isnull().sum()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- These are blocks of code that clean the object columns and convert them to float data types." + ] + }, + { + "cell_type": "code", + "execution_count": 116, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['state', 'city', 'poverty_rate', 'education', 'income', 'share_white',\n", + " 'share_black', 'share_native_american', 'share_asian',\n", + " 'share_hispanic'],\n", + " dtype='object')" + ] + }, + "execution_count": 116, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.columns" + ] + }, + { + "cell_type": "code", + "execution_count": 117, + "metadata": {}, + "outputs": [], + "source": [ + "#income column\n", + "#remove characters that are not digits\n", + "data['income_col'] = data['income'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "#remove existing spaces and join the digits\n", + "data['income_col'] = data['income_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "#fill in entire spaces with zero as a string\n", + "data.loc[data['income_col'] == \"\", 'income_col'] = '0'\n", + "\n", + "#drop the original column\n", + "data.drop('income', axis=1, inplace=True)\n", + "\n", + "#change the type from object to float and rename the column\n", + "data['income'] = data['income_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('income_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 118, + "metadata": {}, + "outputs": [], + "source": [ + "#poverty column\n", + "data['poverty_col'] = data['poverty_rate'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['poverty_col'] = data['poverty_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['poverty_col'] == \"\", 'poverty_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['poverty_rate'] = data['poverty_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('poverty_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 119, + "metadata": {}, + "outputs": [], + "source": [ + "#education column\n", + "data['education_col'] = data['education'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['education_col'] = data['education_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['education_col'] == \"\", 'education_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['education'] = data['education_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('education_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 120, + "metadata": {}, + "outputs": [], + "source": [ + "#share_white column\n", + "data['share_white_col'] = data['share_white'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_white_col'] = data['share_white_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_white_col'] == \"\", 'share_white_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_white'] = data['share_white_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_white_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 121, + "metadata": {}, + "outputs": [], + "source": [ + "#share_black column\n", + "data['share_black_col'] = data['share_black'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_black_col'] = data['share_black_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_black_col'] == \"\", 'share_black_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_black'] = data['share_black_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_black_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 122, + "metadata": {}, + "outputs": [], + "source": [ + "#share_native_american column\n", + "data['share_native_american_col'] = data['share_native_american'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_native_american_col'] = data['share_native_american_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_native_american_col'] == \"\", 'share_native_american_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_native_american'] = data['share_native_american_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_native_american_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 123, + "metadata": {}, + "outputs": [], + "source": [ + "#share_asian column\n", + "data['share_asian_col'] = data['share_asian'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_asian_col'] = data['share_asian_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_asian_col'] == \"\", 'share_asian_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_asian'] = data['share_asian_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_asian_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 124, + "metadata": {}, + "outputs": [], + "source": [ + "#share_hispanic column\n", + "data['share_hispanic_col'] = data['share_hispanic'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_hispanic_col'] = data['share_hispanic_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_hispanic_col'] == \"\", 'share_hispanic_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_hispanic'] = data['share_hispanic_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_hispanic_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 125, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "id 0\n", + "name 0\n", + "date 0\n", + "manner_of_death 0\n", + "armed 9\n", + "age 77\n", + "gender 0\n", + "race 195\n", + "city 0\n", + "state 0\n", + "signs_of_mental_illness 0\n", + "threat_level 0\n", + "flee 65\n", + "body_camera 0\n", + "dtype: int64" + ] + }, + "execution_count": 125, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.isnull().sum()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will fill in the missing values in various columns and also categorize some columns in the killings dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 126, + "metadata": {}, + "outputs": [], + "source": [ + "age_median = killings['age'].median()\n", + "killings['age'].fillna(age_median, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 127, + "metadata": {}, + "outputs": [], + "source": [ + "top_race = killings['race'].describe().top\n", + "killings['race'].fillna(top_race, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 128, + "metadata": {}, + "outputs": [], + "source": [ + "top_flee = killings['flee'].describe().top\n", + "killings['flee'].fillna(top_flee, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 129, + "metadata": {}, + "outputs": [], + "source": [ + "top_armed = killings['armed'].describe().top\n", + "killings['armed'].fillna(top_armed, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 130, + "metadata": {}, + "outputs": [], + "source": [ + "killings.loc[killings['threat_level'] == 'attack', 'threat_level'] = 'high'\n", + "killings.loc[killings['threat_level'] == 'other', 'threat_level'] = 'medium'\n", + "killings.loc[killings['threat_level'] == 'undetermined', 'threat_level'] = 'low'" + ] + }, + { + "cell_type": "code", + "execution_count": 131, + "metadata": {}, + "outputs": [], + "source": [ + "killings['threat_level'] = killings['threat_level'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 132, + "metadata": {}, + "outputs": [], + "source": [ + "killings['manner_of_death'] = killings['manner_of_death'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 133, + "metadata": {}, + "outputs": [], + "source": [ + "killings['gender'] = killings['gender'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 134, + "metadata": {}, + "outputs": [], + "source": [ + "killings['date'] = pd.to_datetime(killings['date'], errors='coerce')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Exploratory Data Analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 135, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
count29329.00000029329.00000029329.00000029329.00000029329.00000029329.00000029329.00000029329.000000
mean146.864400739.397456724.94394662.46315925.73289214.07334082.03614247991.033619
std127.664304291.350599326.169477150.394184118.51309240.698589161.61449527783.222116
min0.0000000.0000000.0000000.0000000.0000000.0000000.0000000.000000
25%48.000000736.000000605.0000001.0000001.0000000.0000008.00000033333.000000
50%121.000000858.000000894.0000007.0000003.0000004.00000025.00000043750.000000
75%212.000000921.000000962.00000035.0000008.00000011.00000071.00000057969.000000
max986.000000999.000000999.000000995.000000997.000000671.000000999.000000250000.000000
\n", + "
" + ], + "text/plain": [ + " poverty_rate education share_white share_black \\\n", + "count 29329.000000 29329.000000 29329.000000 29329.000000 \n", + "mean 146.864400 739.397456 724.943946 62.463159 \n", + "std 127.664304 291.350599 326.169477 150.394184 \n", + "min 0.000000 0.000000 0.000000 0.000000 \n", + "25% 48.000000 736.000000 605.000000 1.000000 \n", + "50% 121.000000 858.000000 894.000000 7.000000 \n", + "75% 212.000000 921.000000 962.000000 35.000000 \n", + "max 986.000000 999.000000 999.000000 995.000000 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "count 29329.000000 29329.000000 29329.000000 29329.000000 \n", + "mean 25.732892 14.073340 82.036142 47991.033619 \n", + "std 118.513092 40.698589 161.614495 27783.222116 \n", + "min 0.000000 0.000000 0.000000 0.000000 \n", + "25% 1.000000 0.000000 8.000000 33333.000000 \n", + "50% 3.000000 4.000000 25.000000 43750.000000 \n", + "75% 8.000000 11.000000 71.000000 57969.000000 \n", + "max 997.000000 671.000000 999.000000 250000.000000 " + ] + }, + "execution_count": 135, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.describe()" + ] + }, + { + "cell_type": "code", + "execution_count": 136, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
count2535.00000025352535253525352535.00000025352535253525352535253525352535
uniqueNaN2481879268NaN261417512342
topNaNTK TK2017-01-24 00:00:00shotgunNaNMWLos AngelesCAFalsehighNot fleeingFalse
freqNaN49823631407NaN24281396394241902161117602264
firstNaNNaN2015-01-03 00:00:00NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
lastNaNNaN2017-12-07 00:00:00NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
mean1445.731755NaNNaNNaNNaN36.526233NaNNaNNaNNaNNaNNaNNaNNaN
std794.259490NaNNaNNaNNaN12.839056NaNNaNNaNNaNNaNNaNNaNNaN
min3.000000NaNNaNNaNNaN6.000000NaNNaNNaNNaNNaNNaNNaNNaN
25%768.500000NaNNaNNaNNaN27.000000NaNNaNNaNNaNNaNNaNNaNNaN
50%1453.000000NaNNaNNaNNaN34.000000NaNNaNNaNNaNNaNNaNNaNNaN
75%2126.500000NaNNaNNaNNaN45.000000NaNNaNNaNNaNNaNNaNNaNNaN
max2822.000000NaNNaNNaNNaN91.000000NaNNaNNaNNaNNaNNaNNaNNaN
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed \\\n", + "count 2535.000000 2535 2535 2535 2535 \n", + "unique NaN 2481 879 2 68 \n", + "top NaN TK TK 2017-01-24 00:00:00 shot gun \n", + "freq NaN 49 8 2363 1407 \n", + "first NaN NaN 2015-01-03 00:00:00 NaN NaN \n", + "last NaN NaN 2017-12-07 00:00:00 NaN NaN \n", + "mean 1445.731755 NaN NaN NaN NaN \n", + "std 794.259490 NaN NaN NaN NaN \n", + "min 3.000000 NaN NaN NaN NaN \n", + "25% 768.500000 NaN NaN NaN NaN \n", + "50% 1453.000000 NaN NaN NaN NaN \n", + "75% 2126.500000 NaN NaN NaN NaN \n", + "max 2822.000000 NaN NaN NaN NaN \n", + "\n", + " age gender race city state signs_of_mental_illness \\\n", + "count 2535.000000 2535 2535 2535 2535 2535 \n", + "unique NaN 2 6 1417 51 2 \n", + "top NaN M W Los Angeles CA False \n", + "freq NaN 2428 1396 39 424 1902 \n", + "first NaN NaN NaN NaN NaN NaN \n", + "last NaN NaN NaN NaN NaN NaN \n", + "mean 36.526233 NaN NaN NaN NaN NaN \n", + "std 12.839056 NaN NaN NaN NaN NaN \n", + "min 6.000000 NaN NaN NaN NaN NaN \n", + "25% 27.000000 NaN NaN NaN NaN NaN \n", + "50% 34.000000 NaN NaN NaN NaN NaN \n", + "75% 45.000000 NaN NaN NaN NaN NaN \n", + "max 91.000000 NaN NaN NaN NaN NaN \n", + "\n", + " threat_level flee body_camera \n", + "count 2535 2535 2535 \n", + "unique 3 4 2 \n", + "top high Not fleeing False \n", + "freq 1611 1760 2264 \n", + "first NaN NaN NaN \n", + "last NaN NaN NaN \n", + "mean NaN NaN NaN \n", + "std NaN NaN NaN \n", + "min NaN NaN NaN \n", + "25% NaN NaN NaN \n", + "50% NaN NaN NaN \n", + "75% NaN NaN NaN \n", + "max NaN NaN NaN " + ] + }, + "execution_count": 136, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.describe(include='all')" + ] + }, + { + "cell_type": "code", + "execution_count": 137, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object \n", + " 1 city 29329 non-null object \n", + " 2 poverty_rate 29329 non-null float64\n", + " 3 education 29329 non-null float64\n", + " 4 share_white 29329 non-null float64\n", + " 5 share_black 29329 non-null float64\n", + " 6 share_native_american 29329 non-null float64\n", + " 7 share_asian 29329 non-null float64\n", + " 8 share_hispanic 29329 non-null float64\n", + " 9 income 29329 non-null float64\n", + "dtypes: float64(8), object(2)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 138, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
0ALAbanda CDP788.0212.0672.0302.00.00.016.011207.0
1ALAbbeville city291.0691.0544.0414.01.01.031.025615.0
2ALAdamsville city255.0789.0523.0449.05.03.023.042575.0
3ALAddison town307.0814.0991.01.00.01.04.037083.0
4ALAkron town42.0686.0132.0865.00.00.03.021667.0
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education share_white share_black \\\n", + "0 AL Abanda CDP 788.0 212.0 672.0 302.0 \n", + "1 AL Abbeville city 291.0 691.0 544.0 414.0 \n", + "2 AL Adamsville city 255.0 789.0 523.0 449.0 \n", + "3 AL Addison town 307.0 814.0 991.0 1.0 \n", + "4 AL Akron town 42.0 686.0 132.0 865.0 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "0 0.0 0.0 16.0 11207.0 \n", + "1 1.0 1.0 31.0 25615.0 \n", + "2 5.0 3.0 23.0 42575.0 \n", + "3 0.0 1.0 4.0 37083.0 \n", + "4 0.0 0.0 3.0 21667.0 " + ] + }, + "execution_count": 138, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 139, + "metadata": {}, + "outputs": [], + "source": [ + "total_state_income = pd.pivot_table(data, values='income', index='state', aggfunc='sum')" + ] + }, + { + "cell_type": "code", + "execution_count": 140, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
income
state
AK14900484.0
AL22155211.0
AR18366199.0
AZ15810888.0
CA84771828.0
\n", + "
" + ], + "text/plain": [ + " income\n", + "state \n", + "AK 14900484.0\n", + "AL 22155211.0\n", + "AR 18366199.0\n", + "AZ 15810888.0\n", + "CA 84771828.0" + ] + }, + "execution_count": 140, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "total_state_income.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 141, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.catplot(x=total_state_income.index, y='income', data=total_state_income, kind='bar', height=5, aspect=14/5)\n", + "plt.xticks(rotation=90)\n", + "\n", + "# Show plot\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 142, + "metadata": {}, + "outputs": [], + "source": [ + "avg_state_poverty = pd.pivot_table(data, values='poverty_rate', index='state', aggfunc='mean') " + ] + }, + { + "cell_type": "code", + "execution_count": 143, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rate
state
AK164.602817
AL187.502564
AR205.609982
AZ221.889135
CA148.695795
\n", + "
" + ], + "text/plain": [ + " poverty_rate\n", + "state \n", + "AK 164.602817\n", + "AL 187.502564\n", + "AR 205.609982\n", + "AZ 221.889135\n", + "CA 148.695795" + ] + }, + "execution_count": 143, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "avg_state_poverty.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 145, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.catplot(x=avg_state_poverty.index, y='poverty_rate', data=avg_state_poverty, kind='bar', height=5, aspect=14/5)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 146, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object \n", + " 1 city 29329 non-null object \n", + " 2 poverty_rate 29329 non-null float64\n", + " 3 education 29329 non-null float64\n", + " 4 share_white 29329 non-null float64\n", + " 5 share_black 29329 non-null float64\n", + " 6 share_native_american 29329 non-null float64\n", + " 7 share_asian 29329 non-null float64\n", + " 8 share_hispanic 29329 non-null float64\n", + " 9 income 29329 non-null float64\n", + "dtypes: float64(8), object(2)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Areas of exploration:\n", + "1. Find the correlation between poverty, education and income in the states\n", + " - which states are poor|rich, more educated, higher income\n", + "2. Top 5 states:\n", + " - with the most killings\n", + " - share of race\n", + " - level of education\n", + " - poverty|income levels\n", + " - e.g. is the race of most killings related to the share of race?\n", + "\n", + "3. Describe the average profile of a person being killed by police:\n", + " - age\n", + " - gender\n", + " - race\n", + " - state\n", + " - poverty|income levels\n", + " - education\n", + " - share of race\n", + "4. Are these killings justified?\n", + " - what is the correlation between manner of death and threat_level|flee\n", + " - did the the threat_level justify the manner of death?" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 1. Exploring the correlation between income, education and poverty rates in the various states." + ] + }, + { + "cell_type": "code", + "execution_count": 147, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
0ALAbanda CDP788.0212.0672.0302.00.00.016.011207.0
1ALAbbeville city291.0691.0544.0414.01.01.031.025615.0
2ALAdamsville city255.0789.0523.0449.05.03.023.042575.0
3ALAddison town307.0814.0991.01.00.01.04.037083.0
4ALAkron town42.0686.0132.0865.00.00.03.021667.0
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education share_white share_black \\\n", + "0 AL Abanda CDP 788.0 212.0 672.0 302.0 \n", + "1 AL Abbeville city 291.0 691.0 544.0 414.0 \n", + "2 AL Adamsville city 255.0 789.0 523.0 449.0 \n", + "3 AL Addison town 307.0 814.0 991.0 1.0 \n", + "4 AL Akron town 42.0 686.0 132.0 865.0 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "0 0.0 0.0 16.0 11207.0 \n", + "1 1.0 1.0 31.0 25615.0 \n", + "2 5.0 3.0 23.0 42575.0 \n", + "3 0.0 1.0 4.0 37083.0 \n", + "4 0.0 0.0 3.0 21667.0 " + ] + }, + "execution_count": 147, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will create pivot tables that aggregate the mean of the values in the columns for each state, then concatenate them into one dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 148, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "state_race = data.pivot_table(index='state', \n", + " values=['share_white', 'share_black', 'share_native_american', 'share_asian', 'share_hispanic'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 149, + "metadata": {}, + "outputs": [], + "source": [ + "state_poverty = data.pivot_table(index='state', \n", + " values=['poverty_rate'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 150, + "metadata": {}, + "outputs": [], + "source": [ + "state_educ = data.pivot_table(index='state', \n", + " values=['education'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 151, + "metadata": {}, + "outputs": [], + "source": [ + "state_income = data.pivot_table(index='state', \n", + " values=['income'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 152, + "metadata": {}, + "outputs": [], + "source": [ + "state_data = pd.concat([state_race, state_poverty, state_income, state_educ], axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": 153, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
share_asianshare_blackshare_hispanicshare_native_americanshare_whitepoverty_rateincomeeducation
state
AK10.6225354.55211322.411268412.892958378.030986164.60281741973.194366634.670423
AL6.104274213.40341926.18461512.299145653.859829187.50256437872.155556724.249573
AR5.005545148.27356741.6303147.114603699.589649205.60998233948.611830727.378928
AZ6.50332610.960089182.332594229.121951557.541020221.88913535057.401330643.731707
CA50.93823924.346912267.30814715.869908634.827201148.69579555697.653088684.221419
\n", + "
" + ], + "text/plain": [ + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AK 10.622535 4.552113 22.411268 412.892958 \n", + "AL 6.104274 213.403419 26.184615 12.299145 \n", + "AR 5.005545 148.273567 41.630314 7.114603 \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 24.346912 267.308147 15.869908 \n", + "\n", + " share_white poverty_rate income education \n", + "state \n", + "AK 378.030986 164.602817 41973.194366 634.670423 \n", + "AL 653.859829 187.502564 37872.155556 724.249573 \n", + "AR 699.589649 205.609982 33948.611830 727.378928 \n", + "AZ 557.541020 221.889135 35057.401330 643.731707 \n", + "CA 634.827201 148.695795 55697.653088 684.221419 " + ] + }, + "execution_count": 153, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "state_data.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Let's look at the correlation between the economic attributes of the states:" + ] + }, + { + "cell_type": "code", + "execution_count": 154, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rateincomeeducation
poverty_rate1.000000-0.626781-0.477182
income-0.6267811.0000000.450140
education-0.4771820.4501401.000000
\n", + "
" + ], + "text/plain": [ + " poverty_rate income education\n", + "poverty_rate 1.000000 -0.626781 -0.477182\n", + "income -0.626781 1.000000 0.450140\n", + "education -0.477182 0.450140 1.000000" + ] + }, + "execution_count": 154, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "state_data[['poverty_rate', 'income', 'education']].corr(method='pearson')" + ] + }, + { + "cell_type": "code", + "execution_count": 155, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "pd.plotting.scatter_matrix(state_data[['poverty_rate', 'income', 'education']], figsize=(10, 10));" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Income and education have a positive but weak correlation. This means that with high levels of education, there is high levels of income.**\n", + "\n", + "**Income and poverty have a worthy negative correlation; meaning that high levels of poverty relate to low levels of income.**\n", + "\n", + "**There is a weak negative correlation between education and poverty. This means that low education levels relate to high poverty levels.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will look at the top and bottom five states in terms of poverty, income and education." + ] + }, + { + "cell_type": "code", + "execution_count": 156, + "metadata": {}, + "outputs": [], + "source": [ + "#poor|rich states\n", + "poor_states = state_data['poverty_rate'].sort_values(ascending=False).head(5)\n", + "rich_states = state_data['poverty_rate'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 157, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " MS 246.044199\n", + " AZ 221.889135\n", + " GA 215.406699\n", + " AR 205.609982\n", + " LA 203.341772\n", + " Name: poverty_rate, dtype: float64,\n", + " state\n", + " DC 18.000000\n", + " NJ 76.143119\n", + " CT 77.375000\n", + " WY 78.549020\n", + " MA 89.170732\n", + " Name: poverty_rate, dtype: float64)" + ] + }, + "execution_count": 157, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "poor_states, rich_states" + ] + }, + { + "cell_type": "code", + "execution_count": 158, + "metadata": {}, + "outputs": [], + "source": [ + "high_inc = state_data['income'].sort_values(ascending=False).head(5)\n", + "low_inc = state_data['income'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 159, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " NJ 78832.957798\n", + " CT 74141.520833\n", + " MD 71692.177606\n", + " MA 69822.195122\n", + " NY 68863.528428\n", + " Name: income, dtype: float64,\n", + " state\n", + " NM 29773.024831\n", + " MS 33512.030387\n", + " DC 33564.000000\n", + " AR 33948.611830\n", + " WV 34913.782716\n", + " Name: income, dtype: float64)" + ] + }, + "execution_count": 159, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "high_inc, low_inc" + ] + }, + { + "cell_type": "code", + "execution_count": 160, + "metadata": {}, + "outputs": [], + "source": [ + "high_educ = state_data['education'].sort_values(ascending=False).head(5)\n", + "low_educ = state_data['education'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 161, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " DC 893.000000\n", + " HI 832.735099\n", + " MA 826.004065\n", + " ME 821.261538\n", + " WI 816.635779\n", + " Name: education, dtype: float64,\n", + " state\n", + " WY 567.490196\n", + " NM 610.611738\n", + " NV 624.503817\n", + " AK 634.670423\n", + " AZ 643.731707\n", + " Name: education, dtype: float64)" + ] + }, + "execution_count": 161, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "high_educ, low_educ" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "*These states standout from the above analysis:*\n", + "\n", + "**MA low poverty, high income, high education**\n", + "\n", + "**DC low poverty, low income and high education**\n", + "\n", + "**NJ, CT low poverty, high income**\n", + "\n", + "**WY low poverty, low education**\n", + "\n", + "**MS, AR high poverty and low income**\n", + "\n", + "**AZ high poverty and low education**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 2. Top5 states" + ] + }, + { + "cell_type": "code", + "execution_count": 162, + "metadata": {}, + "outputs": [], + "source": [ + "top5_states = killings.groupby('state')['state'].count().sort_values(ascending=False).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 163, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/plain": [ + "state\n", + "CA 424\n", + "TX 225\n", + "FL 154\n", + "AZ 118\n", + "OH 79\n", + "Name: state, dtype: int64" + ] + }, + "execution_count": 163, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_states" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**At this point, *AZ* stands out in the top5 states with most killings; and also has high poverty and low education levels.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will select economic data of the top5 states with most killings:" + ] + }, + { + "cell_type": "code", + "execution_count": 164, + "metadata": {}, + "outputs": [], + "source": [ + "top5_data = state_data.query('@state_data.index in @top5_states.index')" + ] + }, + { + "cell_type": "code", + "execution_count": 165, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
share_asianshare_blackshare_hispanicshare_native_americanshare_whitepoverty_rateincomeeducation
state
AZ6.50332610.960089182.332594229.121951557.541020221.88913535057.401330643.731707
CA50.93823924.346912267.30814715.869908634.827201148.69579555697.653088684.221419
FL14.952070123.931373147.2657954.345316709.960784160.60566448552.166667749.193900
OH6.69547336.90946520.6444446.383539815.370370135.60000048856.190123792.947325
TX9.19805451.793360308.7813397.001717718.241557168.23010945645.395535653.486548
\n", + "
" + ], + "text/plain": [ + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 24.346912 267.308147 15.869908 \n", + "FL 14.952070 123.931373 147.265795 4.345316 \n", + "OH 6.695473 36.909465 20.644444 6.383539 \n", + "TX 9.198054 51.793360 308.781339 7.001717 \n", + "\n", + " share_white poverty_rate income education \n", + "state \n", + "AZ 557.541020 221.889135 35057.401330 643.731707 \n", + "CA 634.827201 148.695795 55697.653088 684.221419 \n", + "FL 709.960784 160.605664 48552.166667 749.193900 \n", + "OH 815.370370 135.600000 48856.190123 792.947325 \n", + "TX 718.241557 168.230109 45645.395535 653.486548 " + ] + }, + "execution_count": 165, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_data" + ] + }, + { + "cell_type": "code", + "execution_count": 166, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot2015-02-01shotgun53.0MASheltonWATruehighNot fleeingFalse
14Lewis Lee Lembke2015-02-01shotgun47.0MWAlohaORFalsehighNot fleeingFalse
25John Paul Quintero2015-03-01shot and Taseredunarmed23.0MHWichitaKSFalsemediumNot fleeingFalse
38Matthew Hoffman2015-04-01shottoy weapon32.0MWSan FranciscoCATruehighNot fleeingFalse
49Michael Rodriguez2015-04-01shotnail gun39.0MHEvansCOFalsehighNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 2015-02-01 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 2015-02-01 shot gun 47.0 \n", + "2 5 John Paul Quintero 2015-03-01 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 2015-04-01 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 2015-04-01 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True high \n", + "1 M W Aloha OR False high \n", + "2 M H Wichita KS False medium \n", + "3 M W San Francisco CA True high \n", + "4 M H Evans CO False high \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 166, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 167, + "metadata": {}, + "outputs": [], + "source": [ + "killings['race'] = killings['race'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 168, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 2535 entries, 0 to 2534\n", + "Data columns (total 14 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 2535 non-null int64 \n", + " 1 name 2535 non-null object \n", + " 2 date 2535 non-null datetime64[ns]\n", + " 3 manner_of_death 2535 non-null category \n", + " 4 armed 2535 non-null object \n", + " 5 age 2535 non-null float64 \n", + " 6 gender 2535 non-null category \n", + " 7 race 2535 non-null category \n", + " 8 city 2535 non-null object \n", + " 9 state 2535 non-null object \n", + " 10 signs_of_mental_illness 2535 non-null bool \n", + " 11 threat_level 2535 non-null category \n", + " 12 flee 2535 non-null object \n", + " 13 body_camera 2535 non-null bool \n", + "dtypes: bool(2), category(4), datetime64[ns](1), float64(1), int64(1), object(5)\n", + "memory usage: 173.9+ KB\n" + ] + } + ], + "source": [ + "killings.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 169, + "metadata": {}, + "outputs": [], + "source": [ + "#group state killings by race\n", + "race_count = killings.groupby(['state', 'race'])['race'].count()\n", + "race_count.name = 'race_count'" + ] + }, + { + "cell_type": "code", + "execution_count": 170, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "state_race_count = race_count.reset_index(level='race')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will select the top5 states and their counts of killings." + ] + }, + { + "cell_type": "code", + "execution_count": 171, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race = state_race_count.query('@state_race_count.index in @top5_states.index')" + ] + }, + { + "cell_type": "code", + "execution_count": 172, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
racerace_count
state
AZA0
AZB5
AZH37
AZN8
AZO0
AZW68
CAA15
CAB65
CAH169
CAN1
CAO8
CAW166
FLA1
FLB49
FLH18
FLN0
FLO2
FLW84
OHA2
OHB30
OHH0
OHN0
OHO2
OHW45
TXA2
TXB46
TXH66
TXN1
TXO3
TXW107
\n", + "
" + ], + "text/plain": [ + " race race_count\n", + "state \n", + "AZ A 0\n", + "AZ B 5\n", + "AZ H 37\n", + "AZ N 8\n", + "AZ O 0\n", + "AZ W 68\n", + "CA A 15\n", + "CA B 65\n", + "CA H 169\n", + "CA N 1\n", + "CA O 8\n", + "CA W 166\n", + "FL A 1\n", + "FL B 49\n", + "FL H 18\n", + "FL N 0\n", + "FL O 2\n", + "FL W 84\n", + "OH A 2\n", + "OH B 30\n", + "OH H 0\n", + "OH N 0\n", + "OH O 2\n", + "OH W 45\n", + "TX A 2\n", + "TX B 46\n", + "TX H 66\n", + "TX N 1\n", + "TX O 3\n", + "TX W 107" + ] + }, + "execution_count": 172, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will create a dataframe that contains the percentage share ofeach race killed in each of the top5 states." + ] + }, + { + "cell_type": "code", + "execution_count": 254, + "metadata": {}, + "outputs": [], + "source": [ + "race_pivot = top5_race.pivot_table(index=top5_race.index, values='race_count', columns='race', aggfunc=['sum'])\n", + "race_pivot.columns = ['sum_asian', 'sum_black', 'sum_hispanic', 'sum_natives', 'sum_others', 'sum_whites']\n", + "race_pivot.drop('sum_others', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 255, + "metadata": {}, + "outputs": [], + "source": [ + "race_share = top5_data[['share_asian', 'share_black', 'share_hispanic', 'share_native_american', 'share_white']]" + ] + }, + { + "cell_type": "code", + "execution_count": 256, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
sum_asiansum_blacksum_hispanicsum_nativessum_whitesshare_asianshare_blackshare_hispanicshare_native_americanshare_white
state
AZ05378686.50332610.960089182.332594229.121951557.541020
CA1565169116650.938239240.000000267.30814715.869908634.827201
FL1491808414.952070123.931373147.2657954.345316709.960784
OH23000456.69547336.90946520.6444446.383539815.370370
TX2466611079.19805451.793360308.7813397.001717718.241557
\n", + "
" + ], + "text/plain": [ + " sum_asian sum_black sum_hispanic sum_natives sum_whites \\\n", + "state \n", + "AZ 0 5 37 8 68 \n", + "CA 15 65 169 1 166 \n", + "FL 1 49 18 0 84 \n", + "OH 2 30 0 0 45 \n", + "TX 2 46 66 1 107 \n", + "\n", + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 240.000000 267.308147 15.869908 \n", + "FL 14.952070 123.931373 147.265795 4.345316 \n", + "OH 6.695473 36.909465 20.644444 6.383539 \n", + "TX 9.198054 51.793360 308.781339 7.001717 \n", + "\n", + " share_white \n", + "state \n", + "AZ 557.541020 \n", + "CA 634.827201 \n", + "FL 709.960784 \n", + "OH 815.370370 \n", + "TX 718.241557 " + ] + }, + "execution_count": 256, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race_data = pd.concat([race_pivot, race_share], axis=1)\n", + "top5_race_data['share_black']['CA'] = 240\n", + "top5_race_data" + ] + }, + { + "cell_type": "code", + "execution_count": 257, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race_data['%_asian'] = round((top5_race_data['sum_asian'] / top5_race_data['share_asian']) * 100)\n", + "top5_race_data['%_black'] = round((top5_race_data['sum_black'] / top5_race_data['share_black']) * 100)\n", + "top5_race_data['%_hispanic'] = round((top5_race_data['sum_hispanic'] / top5_race_data['share_hispanic']) * 100)\n", + "top5_race_data['%_natives'] = round((top5_race_data['sum_natives'] / top5_race_data['share_native_american']) * 100)\n", + "top5_race_data['%_whites'] = round((top5_race_data['sum_whites'] / top5_race_data['share_white']) * 100)" + ] + }, + { + "cell_type": "code", + "execution_count": 258, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "top5_race_data.drop(['sum_asian', 'sum_black', 'sum_hispanic', 'sum_natives', 'sum_whites',\n", + " 'share_asian', 'share_black', 'share_hispanic', 'share_native_american',\n", + " 'share_white'], axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 259, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race_pc = top5_race_data.astype(np.int64)" + ] + }, + { + "cell_type": "code", + "execution_count": 260, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
%_asian%_black%_hispanic%_natives%_whites
state
AZ04620312
CA292763626
FL74012012
OH3081006
TX2289211415
\n", + "
" + ], + "text/plain": [ + " %_asian %_black %_hispanic %_natives %_whites\n", + "state \n", + "AZ 0 46 20 3 12\n", + "CA 29 27 63 6 26\n", + "FL 7 40 12 0 12\n", + "OH 30 81 0 0 6\n", + "TX 22 89 21 14 15" + ] + }, + "execution_count": 260, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race_pc" + ] + }, + { + "cell_type": "code", + "execution_count": 261, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['%_asian', '%_black', '%_hispanic', '%_natives', '%_whites'], dtype='object')" + ] + }, + "execution_count": 261, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "races = top5_race_pc.columns\n", + "races" + ] + }, + { + "cell_type": "code", + "execution_count": 262, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#a visualization of each race by state\n", + "for race in races:\n", + " values= top5_race_pc[race]\n", + " labels= top5_race_pc.index\n", + " plt.axis('equal')\n", + " plt.title(race)\n", + " plt.pie(values, labels=labels, radius=2, autopct='%0.0f%%')\n", + " plt.show();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Blacks have a generally higher percentage of killings in each state compared to other races.**\n", + "\n", + "**TX has a large share of killings of Native Americans and Blacks.**\n", + "\n", + "**CA has a large share of killings of Whites, Hisapnics and Asians.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 3. Profile of a person killed by police" + ] + }, + { + "cell_type": "code", + "execution_count": 263, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot2015-02-01shotgun53.0MASheltonWATruehighNot fleeingFalse
14Lewis Lee Lembke2015-02-01shotgun47.0MWAlohaORFalsehighNot fleeingFalse
25John Paul Quintero2015-03-01shot and Taseredunarmed23.0MHWichitaKSFalsemediumNot fleeingFalse
38Matthew Hoffman2015-04-01shottoy weapon32.0MWSan FranciscoCATruehighNot fleeingFalse
49Michael Rodriguez2015-04-01shotnail gun39.0MHEvansCOFalsehighNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 2015-02-01 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 2015-02-01 shot gun 47.0 \n", + "2 5 John Paul Quintero 2015-03-01 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 2015-04-01 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 2015-04-01 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True high \n", + "1 M W Aloha OR False high \n", + "2 M H Wichita KS False medium \n", + "3 M W San Francisco CA True high \n", + "4 M H Evans CO False high \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 263, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 264, + "metadata": {}, + "outputs": [], + "source": [ + "race_avg_age = killings.groupby('race')['age'].mean()" + ] + }, + { + "cell_type": "code", + "execution_count": 265, + "metadata": {}, + "outputs": [], + "source": [ + "race_avg_age.name = 'average_age'" + ] + }, + { + "cell_type": "code", + "execution_count": 266, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "race\n", + "A 36.538462\n", + "B 31.669903\n", + "H 33.018913\n", + "N 30.451613\n", + "O 33.071429\n", + "W 39.942693\n", + "Name: average_age, dtype: float64\n" + ] + } + ], + "source": [ + "print(race_avg_age)" + ] + }, + { + "cell_type": "code", + "execution_count": 267, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "34.11550197750503" + ] + }, + "execution_count": 267, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "race_avg_age.mean()" + ] + }, + { + "cell_type": "code", + "execution_count": 268, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "M 2428\n", + "F 107\n", + "Name: gender, dtype: int64" + ] + }, + "execution_count": 268, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.gender.value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 269, + "metadata": {}, + "outputs": [], + "source": [ + "males = len(killings[killings['gender'] == 'M'])\n", + "females = len(killings[killings['gender'] == 'F'])\n", + "total = killings.shape[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 270, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(2428, 107, 2535)" + ] + }, + "execution_count": 270, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "males, females, total" + ] + }, + { + "cell_type": "code", + "execution_count": 271, + "metadata": {}, + "outputs": [], + "source": [ + "males_pc = round((males / total) * 100)\n", + "females_pc = round((females / total) * 100)" + ] + }, + { + "cell_type": "code", + "execution_count": 272, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(96, 4)" + ] + }, + "execution_count": 272, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "males_pc, females_pc" + ] + }, + { + "cell_type": "code", + "execution_count": 273, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.axis('equal')\n", + "plt.title('Gender Percentage of Killings')\n", + "plt.pie([males_pc, females_pc], labels=['male', 'female'], radius=2, autopct='%0.0f%%')\n", + "plt.show();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**The average profile of a victim killed by police is a Black male, aged 34 years and most likely living in TX.**\n", + "\n", + "**The average profile of a victim killed by police is a Hisapnic male, aged 34 years and most likely living in CA.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 4. Justification of killings" + ] + }, + { + "cell_type": "code", + "execution_count": 274, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.swarmplot(data=killings,\n", + " x='age',\n", + " y='flee',\n", + " hue='manner_of_death')\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Victims of police shootings averagely between the age of 20 and 45 were either not fleeing, or fleeing by car and foot.**\n", + "\n", + "**Victims aged 50 and beyond were mostly shot when not fleeing.**" + ] + }, + { + "cell_type": "code", + "execution_count": 275, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.countplot(data=killings,\n", + " y=\"manner_of_death\",\n", + " hue='threat_level')\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Majority of victims who were shot were a high threat level.**\n", + "\n", + "**Victims who were shot and tasered were both high and medium threat levels.**\n", + "\n", + "*It can be seen that majority of the killings were justified as victims who were high and medium threat levels.*" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/Level 1/Intermediate/US Police Killings Analysis.ipynb b/Level 1/Intermediate/US Police Killings Analysis.ipynb new file mode 100644 index 0000000..aa1d579 --- /dev/null +++ b/Level 1/Intermediate/US Police Killings Analysis.ipynb @@ -0,0 +1,5388 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Import required libraries." + ] + }, + { + "cell_type": "code", + "execution_count": 88, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import warnings\n", + "warnings.filterwarnings('ignore')\n", + "\n", + "import re\n", + "import seaborn as sns\n", + "sns.set_style(\"whitegrid\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Open and read the data sets." + ] + }, + { + "cell_type": "code", + "execution_count": 89, + "metadata": {}, + "outputs": [], + "source": [ + "income = pd.read_csv('datasets/MedianHouseholdIncome2015.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 90, + "metadata": {}, + "outputs": [], + "source": [ + "poverty = pd.read_csv('datasets/PercentagePeopleBelowPovertyLevel.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 91, + "metadata": {}, + "outputs": [], + "source": [ + "education = pd.read_csv('datasets/PercentOver25CompletedHighSchool.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 92, + "metadata": {}, + "outputs": [], + "source": [ + "killings = pd.read_csv('datasets/PoliceKillingsUS.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 93, + "metadata": {}, + "outputs": [], + "source": [ + "city_race = pd.read_csv('datasets/ShareRaceByCity.csv', encoding='windows-1251')" + ] + }, + { + "cell_type": "code", + "execution_count": 94, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCityMedian Income
0ALAbanda CDP11207
1ALAbbeville city25615
2ALAdamsville city42575
3ALAddison town37083
4ALAkron town21667
\n", + "
" + ], + "text/plain": [ + " Geographic Area City Median Income\n", + "0 AL Abanda CDP 11207\n", + "1 AL Abbeville city 25615\n", + "2 AL Adamsville city 42575\n", + "3 AL Addison town 37083\n", + "4 AL Akron town 21667" + ] + }, + "execution_count": 94, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "income.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 95, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "51" + ] + }, + "execution_count": 95, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "income['Geographic Area'].nunique()" + ] + }, + { + "cell_type": "code", + "execution_count": 96, + "metadata": {}, + "outputs": [], + "source": [ + "city_samp = city_race.sample(frac=0.25, replace=False, random_state=0, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": 97, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 7317 entries, 20414 to 22006\n", + "Data columns (total 7 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic area 7317 non-null object\n", + " 1 City 7317 non-null object\n", + " 2 share_white 7317 non-null object\n", + " 3 share_black 7317 non-null object\n", + " 4 share_native_american 7317 non-null object\n", + " 5 share_asian 7317 non-null object\n", + " 6 share_hispanic 7317 non-null object\n", + "dtypes: object(7)\n", + "memory usage: 457.3+ KB\n" + ] + } + ], + "source": [ + "city_samp.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 98, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0" + ] + }, + "execution_count": 98, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "city_samp.index.duplicated().sum()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 99, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29322 entries, 0 to 29321\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29322 non-null object\n", + " 1 City 29322 non-null object\n", + " 2 Median Income 29271 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.4+ KB\n" + ] + } + ], + "source": [ + "income.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 100, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypoverty_rate
0ALAbanda CDP78.8
1ALAbbeville city29.1
2ALAdamsville city25.5
3ALAddison town30.7
4ALAkron town42
\n", + "
" + ], + "text/plain": [ + " Geographic Area City poverty_rate\n", + "0 AL Abanda CDP 78.8\n", + "1 AL Abbeville city 29.1\n", + "2 AL Adamsville city 25.5\n", + "3 AL Addison town 30.7\n", + "4 AL Akron town 42" + ] + }, + "execution_count": 100, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "poverty.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 101, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29329 non-null object\n", + " 1 City 29329 non-null object\n", + " 2 poverty_rate 29329 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.5+ KB\n" + ] + } + ], + "source": [ + "poverty.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 102, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypercent_completed_hs
0ALAbanda CDP21.2
1ALAbbeville city69.1
2ALAdamsville city78.9
3ALAddison town81.4
4ALAkron town68.6
\n", + "
" + ], + "text/plain": [ + " Geographic Area City percent_completed_hs\n", + "0 AL Abanda CDP 21.2\n", + "1 AL Abbeville city 69.1\n", + "2 AL Adamsville city 78.9\n", + "3 AL Addison town 81.4\n", + "4 AL Akron town 68.6" + ] + }, + "execution_count": 102, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "education.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 103, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic Area 29329 non-null object\n", + " 1 City 29329 non-null object\n", + " 2 percent_completed_hs 29329 non-null object\n", + "dtypes: object(3)\n", + "memory usage: 687.5+ KB\n" + ] + } + ], + "source": [ + "education.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 104, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot02/01/15shotgun53.0MASheltonWATrueattackNot fleeingFalse
14Lewis Lee Lembke02/01/15shotgun47.0MWAlohaORFalseattackNot fleeingFalse
25John Paul Quintero03/01/15shot and Taseredunarmed23.0MHWichitaKSFalseotherNot fleeingFalse
38Matthew Hoffman04/01/15shottoy weapon32.0MWSan FranciscoCATrueattackNot fleeingFalse
49Michael Rodriguez04/01/15shotnail gun39.0MHEvansCOFalseattackNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 02/01/15 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 02/01/15 shot gun 47.0 \n", + "2 5 John Paul Quintero 03/01/15 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 04/01/15 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 04/01/15 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True attack \n", + "1 M W Aloha OR False attack \n", + "2 M H Wichita KS False other \n", + "3 M W San Francisco CA True attack \n", + "4 M H Evans CO False attack \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 104, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 105, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 2535 entries, 0 to 2534\n", + "Data columns (total 14 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 2535 non-null int64 \n", + " 1 name 2535 non-null object \n", + " 2 date 2535 non-null object \n", + " 3 manner_of_death 2535 non-null object \n", + " 4 armed 2526 non-null object \n", + " 5 age 2458 non-null float64\n", + " 6 gender 2535 non-null object \n", + " 7 race 2340 non-null object \n", + " 8 city 2535 non-null object \n", + " 9 state 2535 non-null object \n", + " 10 signs_of_mental_illness 2535 non-null bool \n", + " 11 threat_level 2535 non-null object \n", + " 12 flee 2470 non-null object \n", + " 13 body_camera 2535 non-null bool \n", + "dtypes: bool(2), float64(1), int64(1), object(10)\n", + "memory usage: 242.7+ KB\n" + ] + } + ], + "source": [ + "killings.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 106, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic areaCityshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
0ALAbanda CDP67.230.2001.6
1ALAbbeville city54.441.40.113.1
2ALAdamsville city52.344.90.50.32.3
3ALAddison town99.10.100.10.4
4ALAkron town13.286.5000.3
\n", + "
" + ], + "text/plain": [ + " Geographic area City share_white share_black \\\n", + "0 AL Abanda CDP 67.2 30.2 \n", + "1 AL Abbeville city 54.4 41.4 \n", + "2 AL Adamsville city 52.3 44.9 \n", + "3 AL Addison town 99.1 0.1 \n", + "4 AL Akron town 13.2 86.5 \n", + "\n", + " share_native_american share_asian share_hispanic \n", + "0 0 0 1.6 \n", + "1 0.1 1 3.1 \n", + "2 0.5 0.3 2.3 \n", + "3 0 0.1 0.4 \n", + "4 0 0 0.3 " + ] + }, + "execution_count": 106, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "city_race.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 107, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29268 entries, 0 to 29267\n", + "Data columns (total 7 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Geographic area 29268 non-null object\n", + " 1 City 29268 non-null object\n", + " 2 share_white 29268 non-null object\n", + " 3 share_black 29268 non-null object\n", + " 4 share_native_american 29268 non-null object\n", + " 5 share_asian 29268 non-null object\n", + " 6 share_hispanic 29268 non-null object\n", + "dtypes: object(7)\n", + "memory usage: 1.6+ MB\n" + ] + } + ], + "source": [ + "city_race.info()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will concatenate the `income`, `poverty`, `education` and `city_race` dataframes for compact analysis." + ] + }, + { + "cell_type": "code", + "execution_count": 108, + "metadata": {}, + "outputs": [], + "source": [ + "data = pd.concat([poverty, education, income, city_race], axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": 109, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Geographic AreaCitypoverty_rateGeographic AreaCitypercent_completed_hsGeographic AreaCityMedian IncomeGeographic areaCityshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
26119TXTimpson city42.7TXTimpson city70.8TXTomball city44086TXVenus town79.413.20.51.724.8
20016OHPainesville city23OHPainesville city78.1OHParma city50440OHPleasant Run CDP79.814.50.51.82.2
15984NJEast Rutherford borough10.1NJEast Rutherford borough92.9NJEllisburg CDP61544NJFarmingdale borough89.62.90.53.26.9
14734MTCamas CDP48.6MTCamas CDP87.5MTCharlo CDP44583MTConrad city95.10.21.80.31.5
18451NCMarshville town28.9NCMarshville town71.1NCMaysville town24432NCMorganton city70.112.20.92.416.4
28940WIRichfield village2.8WIRichfield village95.8WIRiver Hills village156250WISpooner city95.10.31.90.71.3
1442ARBlue Eye town74.4ARBlue Eye town16.7ARBlue Eye town(X)ARBooneville city93.510.90.63.2
25763TXPoint Comfort city7.3TXPoint Comfort city89.2TXPortland city62561TXRamos CDP76.7000100
7761INLittle York town20.1INLittle York town77.8INLogansport city32982INLowell town95.90.50.40.36.9
1592ARGreers Ferry city13ARGreers Ferry city81.1ARGreers Ferry city31810ARHackett city920.13.40.60.6
\n", + "
" + ], + "text/plain": [ + " Geographic Area City poverty_rate Geographic Area \\\n", + "26119 TX Timpson city 42.7 TX \n", + "20016 OH Painesville city 23 OH \n", + "15984 NJ East Rutherford borough 10.1 NJ \n", + "14734 MT Camas CDP 48.6 MT \n", + "18451 NC Marshville town 28.9 NC \n", + "28940 WI Richfield village 2.8 WI \n", + "1442 AR Blue Eye town 74.4 AR \n", + "25763 TX Point Comfort city 7.3 TX \n", + "7761 IN Little York town 20.1 IN \n", + "1592 AR Greers Ferry city 13 AR \n", + "\n", + " City percent_completed_hs Geographic Area \\\n", + "26119 Timpson city 70.8 TX \n", + "20016 Painesville city 78.1 OH \n", + "15984 East Rutherford borough 92.9 NJ \n", + "14734 Camas CDP 87.5 MT \n", + "18451 Marshville town 71.1 NC \n", + "28940 Richfield village 95.8 WI \n", + "1442 Blue Eye town 16.7 AR \n", + "25763 Point Comfort city 89.2 TX \n", + "7761 Little York town 77.8 IN \n", + "1592 Greers Ferry city 81.1 AR \n", + "\n", + " City Median Income Geographic area City \\\n", + "26119 Tomball city 44086 TX Venus town \n", + "20016 Parma city 50440 OH Pleasant Run CDP \n", + "15984 Ellisburg CDP 61544 NJ Farmingdale borough \n", + "14734 Charlo CDP 44583 MT Conrad city \n", + "18451 Maysville town 24432 NC Morganton city \n", + "28940 River Hills village 156250 WI Spooner city \n", + "1442 Blue Eye town (X) AR Booneville city \n", + "25763 Portland city 62561 TX Ramos CDP \n", + "7761 Logansport city 32982 IN Lowell town \n", + "1592 Greers Ferry city 31810 AR Hackett city \n", + "\n", + " share_white share_black share_native_american share_asian share_hispanic \n", + "26119 79.4 13.2 0.5 1.7 24.8 \n", + "20016 79.8 14.5 0.5 1.8 2.2 \n", + "15984 89.6 2.9 0.5 3.2 6.9 \n", + "14734 95.1 0.2 1.8 0.3 1.5 \n", + "18451 70.1 12.2 0.9 2.4 16.4 \n", + "28940 95.1 0.3 1.9 0.7 1.3 \n", + "1442 93.5 1 0.9 0.6 3.2 \n", + "25763 76.7 0 0 0 100 \n", + "7761 95.9 0.5 0.4 0.3 6.9 \n", + "1592 92 0.1 3.4 0.6 0.6 " + ] + }, + "execution_count": 109, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.sample(10)" + ] + }, + { + "cell_type": "code", + "execution_count": 110, + "metadata": {}, + "outputs": [], + "source": [ + "#rename the columns\n", + "data.columns = ['state', 'city', 'poverty_rate', 'Geographic_Area_x', 'City_x',\n", + " 'education', 'Geographic_Area_y', 'City_y', 'income',\n", + " 'Geographic_area_z', 'City_z', 'share_white', 'share_black',\n", + " 'share_native_american', 'share_asian', 'share_hispanic']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- The cities are not the same, but the areas(state) are. In this case I will perform my analysis based on geographic area." + ] + }, + { + "cell_type": "code", + "execution_count": 111, + "metadata": {}, + "outputs": [], + "source": [ + "data.drop(['Geographic_Area_x', 'City_x', 'Geographic_Area_y', 'City_y', 'Geographic_area_z', 'City_z'], \n", + " axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 112, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationincomeshare_whiteshare_blackshare_native_americanshare_asianshare_hispanic
18204NCDellview town01002361679.317.10.302.1
10302KYWhite Plains city14.482.35733494.11.6000
20058OHPlumwood CDP6.280.35342148.544.90.14.31.8
20849OKMulhall town28.689.4-88.2010.600
23114PAUtica borough20.584.34546998.60.30.10.41
21753PACentre Hall borough694.95375098.20.300.40.9
11271MDNanticoke Acres CDP52.6505766565.3230.45.55.9
12766MNHoffman city21.887.74500094.71.10.50.72.8
21915PAEagles Mere borough2.51005682795.71.20.21.50.8
4443FLHastings town26.381.836196970.70.40.44.3
2124CACamarillo city6.492.28815296.90006.2
14244MOMaplewood city19.691.52531396.60001.7
6547ILGrand Ridge village10.894.83906383.73.30.36.88.8
14122MOIronton city36.380.62375096.90000.6
4150FLArcher city35.590.6211465838100
5427GAMineral Bluff CDP251003309765.632.1003.7
5649GAWarwick city38.564.14300039.157.70.212.5
5573GASocial Circle city12.479.62432154.440.10.223
10038KYIndependence city8.489.82187598.40.300.30.3
28846WINiagara city20.393.28303698.300.401
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education income share_white \\\n", + "18204 NC Dellview town 0 100 23616 79.3 \n", + "10302 KY White Plains city 14.4 82.3 57334 94.1 \n", + "20058 OH Plumwood CDP 6.2 80.3 53421 48.5 \n", + "20849 OK Mulhall town 28.6 89.4 - 88.2 \n", + "23114 PA Utica borough 20.5 84.3 45469 98.6 \n", + "21753 PA Centre Hall borough 6 94.9 53750 98.2 \n", + "11271 MD Nanticoke Acres CDP 52.6 50 57665 65.3 \n", + "12766 MN Hoffman city 21.8 87.7 45000 94.7 \n", + "21915 PA Eagles Mere borough 2.5 100 56827 95.7 \n", + "4443 FL Hastings town 26.3 81.8 36196 97 \n", + "2124 CA Camarillo city 6.4 92.2 88152 96.9 \n", + "14244 MO Maplewood city 19.6 91.5 25313 96.6 \n", + "6547 IL Grand Ridge village 10.8 94.8 39063 83.7 \n", + "14122 MO Ironton city 36.3 80.6 23750 96.9 \n", + "4150 FL Archer city 35.5 90.6 21146 58 \n", + "5427 GA Mineral Bluff CDP 25 100 33097 65.6 \n", + "5649 GA Warwick city 38.5 64.1 43000 39.1 \n", + "5573 GA Social Circle city 12.4 79.6 24321 54.4 \n", + "10038 KY Independence city 8.4 89.8 21875 98.4 \n", + "28846 WI Niagara city 20.3 93.2 83036 98.3 \n", + "\n", + " share_black share_native_american share_asian share_hispanic \n", + "18204 17.1 0.3 0 2.1 \n", + "10302 1.6 0 0 0 \n", + "20058 44.9 0.1 4.3 1.8 \n", + "20849 0 10.6 0 0 \n", + "23114 0.3 0.1 0.4 1 \n", + "21753 0.3 0 0.4 0.9 \n", + "11271 23 0.4 5.5 5.9 \n", + "12766 1.1 0.5 0.7 2.8 \n", + "21915 1.2 0.2 1.5 0.8 \n", + "4443 0.7 0.4 0.4 4.3 \n", + "2124 0 0 0 6.2 \n", + "14244 0 0 0 1.7 \n", + "6547 3.3 0.3 6.8 8.8 \n", + "14122 0 0 0 0.6 \n", + "4150 38 1 0 0 \n", + "5427 32.1 0 0 3.7 \n", + "5649 57.7 0.2 1 2.5 \n", + "5573 40.1 0.2 2 3 \n", + "10038 0.3 0 0.3 0.3 \n", + "28846 0 0.4 0 1 " + ] + }, + "execution_count": 112, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.sample(20)" + ] + }, + { + "cell_type": "code", + "execution_count": 113, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object\n", + " 1 city 29329 non-null object\n", + " 2 poverty_rate 29329 non-null object\n", + " 3 education 29329 non-null object\n", + " 4 income 29271 non-null object\n", + " 5 share_white 29268 non-null object\n", + " 6 share_black 29268 non-null object\n", + " 7 share_native_american 29268 non-null object\n", + " 8 share_asian 29268 non-null object\n", + " 9 share_hispanic 29268 non-null object\n", + "dtypes: object(10)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 114, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot02/01/15shotgun53.0MASheltonWATrueattackNot fleeingFalse
14Lewis Lee Lembke02/01/15shotgun47.0MWAlohaORFalseattackNot fleeingFalse
25John Paul Quintero03/01/15shot and Taseredunarmed23.0MHWichitaKSFalseotherNot fleeingFalse
38Matthew Hoffman04/01/15shottoy weapon32.0MWSan FranciscoCATrueattackNot fleeingFalse
49Michael Rodriguez04/01/15shotnail gun39.0MHEvansCOFalseattackNot fleeingFalse
.............................................
25302822Rodney E. Jacobs28/07/17shotgun31.0MNaNKansas CityMOFalseattackNot fleeingFalse
25312813TK TK28/07/17shotvehicleNaNMNaNAlbuquerqueNMFalseattackCarFalse
25322818Dennis W. Robinson29/07/17shotgun48.0MNaNMelbaIDFalseattackCarFalse
25332817Isaiah Tucker31/07/17shotvehicle28.0MBOshkoshWIFalseattackCarTrue
25342815Dwayne Jeune31/07/17shotknife32.0MBBrooklynNYTrueattackNot fleeingFalse
\n", + "

2535 rows × 14 columns

\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 02/01/15 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 02/01/15 shot gun 47.0 \n", + "2 5 John Paul Quintero 03/01/15 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 04/01/15 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 04/01/15 shot nail gun 39.0 \n", + "... ... ... ... ... ... ... \n", + "2530 2822 Rodney E. Jacobs 28/07/17 shot gun 31.0 \n", + "2531 2813 TK TK 28/07/17 shot vehicle NaN \n", + "2532 2818 Dennis W. Robinson 29/07/17 shot gun 48.0 \n", + "2533 2817 Isaiah Tucker 31/07/17 shot vehicle 28.0 \n", + "2534 2815 Dwayne Jeune 31/07/17 shot knife 32.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True attack \n", + "1 M W Aloha OR False attack \n", + "2 M H Wichita KS False other \n", + "3 M W San Francisco CA True attack \n", + "4 M H Evans CO False attack \n", + "... ... ... ... ... ... ... \n", + "2530 M NaN Kansas City MO False attack \n", + "2531 M NaN Albuquerque NM False attack \n", + "2532 M NaN Melba ID False attack \n", + "2533 M B Oshkosh WI False attack \n", + "2534 M B Brooklyn NY True attack \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False \n", + "... ... ... \n", + "2530 Not fleeing False \n", + "2531 Car False \n", + "2532 Car False \n", + "2533 Car True \n", + "2534 Not fleeing False \n", + "\n", + "[2535 rows x 14 columns]" + ] + }, + "execution_count": 114, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Data Cleaning and Prerocessing\n", + " - filling in missing values and changing data types" + ] + }, + { + "cell_type": "code", + "execution_count": 115, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "state 0\n", + "city 0\n", + "poverty_rate 0\n", + "education 0\n", + "income 58\n", + "share_white 61\n", + "share_black 61\n", + "share_native_american 61\n", + "share_asian 61\n", + "share_hispanic 61\n", + "dtype: int64" + ] + }, + "execution_count": 115, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#check for missing values\n", + "data.isnull().sum()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- These are blocks of code that clean the object columns and convert them to float data types." + ] + }, + { + "cell_type": "code", + "execution_count": 116, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['state', 'city', 'poverty_rate', 'education', 'income', 'share_white',\n", + " 'share_black', 'share_native_american', 'share_asian',\n", + " 'share_hispanic'],\n", + " dtype='object')" + ] + }, + "execution_count": 116, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.columns" + ] + }, + { + "cell_type": "code", + "execution_count": 117, + "metadata": {}, + "outputs": [], + "source": [ + "#income column\n", + "#remove characters that are not digits\n", + "data['income_col'] = data['income'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "#remove existing spaces and join the digits\n", + "data['income_col'] = data['income_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "#fill in entire spaces with zero as a string\n", + "data.loc[data['income_col'] == \"\", 'income_col'] = '0'\n", + "\n", + "#drop the original column\n", + "data.drop('income', axis=1, inplace=True)\n", + "\n", + "#change the type from object to float and rename the column\n", + "data['income'] = data['income_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('income_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 118, + "metadata": {}, + "outputs": [], + "source": [ + "#poverty column\n", + "data['poverty_col'] = data['poverty_rate'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['poverty_col'] = data['poverty_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['poverty_col'] == \"\", 'poverty_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['poverty_rate'] = data['poverty_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('poverty_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 119, + "metadata": {}, + "outputs": [], + "source": [ + "#education column\n", + "data['education_col'] = data['education'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['education_col'] = data['education_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['education_col'] == \"\", 'education_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['education'] = data['education_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('education_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 120, + "metadata": {}, + "outputs": [], + "source": [ + "#share_white column\n", + "data['share_white_col'] = data['share_white'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_white_col'] = data['share_white_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_white_col'] == \"\", 'share_white_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_white'] = data['share_white_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_white_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 121, + "metadata": {}, + "outputs": [], + "source": [ + "#share_black column\n", + "data['share_black_col'] = data['share_black'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_black_col'] = data['share_black_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_black_col'] == \"\", 'share_black_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_black'] = data['share_black_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_black_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 122, + "metadata": {}, + "outputs": [], + "source": [ + "#share_native_american column\n", + "data['share_native_american_col'] = data['share_native_american'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_native_american_col'] = data['share_native_american_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_native_american_col'] == \"\", 'share_native_american_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_native_american'] = data['share_native_american_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_native_american_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 123, + "metadata": {}, + "outputs": [], + "source": [ + "#share_asian column\n", + "data['share_asian_col'] = data['share_asian'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_asian_col'] = data['share_asian_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_asian_col'] == \"\", 'share_asian_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_asian'] = data['share_asian_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_asian_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 124, + "metadata": {}, + "outputs": [], + "source": [ + "#share_hispanic column\n", + "data['share_hispanic_col'] = data['share_hispanic'].apply(lambda x: re.sub(r\"[^0-9]\", \"\", str(x)))\n", + "\n", + "data['share_hispanic_col'] = data['share_hispanic_col'].apply(lambda x: \"\".join(str(x).split()))\n", + "\n", + "data.loc[data['share_hispanic_col'] == \"\", 'share_hispanic_col'] = '0'\n", + "\n", + "#change the type from object to float\n", + "data['share_hispanic'] = data['share_hispanic_col'].astype('float')\n", + "\n", + "#drop the redundant column\n", + "data.drop('share_hispanic_col', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 125, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "id 0\n", + "name 0\n", + "date 0\n", + "manner_of_death 0\n", + "armed 9\n", + "age 77\n", + "gender 0\n", + "race 195\n", + "city 0\n", + "state 0\n", + "signs_of_mental_illness 0\n", + "threat_level 0\n", + "flee 65\n", + "body_camera 0\n", + "dtype: int64" + ] + }, + "execution_count": 125, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.isnull().sum()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will fill in the missing values in various columns and also categorize some columns in the killings dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 126, + "metadata": {}, + "outputs": [], + "source": [ + "age_median = killings['age'].median()\n", + "killings['age'].fillna(age_median, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 127, + "metadata": {}, + "outputs": [], + "source": [ + "top_race = killings['race'].describe().top\n", + "killings['race'].fillna(top_race, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 128, + "metadata": {}, + "outputs": [], + "source": [ + "top_flee = killings['flee'].describe().top\n", + "killings['flee'].fillna(top_flee, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 129, + "metadata": {}, + "outputs": [], + "source": [ + "top_armed = killings['armed'].describe().top\n", + "killings['armed'].fillna(top_armed, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 130, + "metadata": {}, + "outputs": [], + "source": [ + "killings.loc[killings['threat_level'] == 'attack', 'threat_level'] = 'high'\n", + "killings.loc[killings['threat_level'] == 'other', 'threat_level'] = 'medium'\n", + "killings.loc[killings['threat_level'] == 'undetermined', 'threat_level'] = 'low'" + ] + }, + { + "cell_type": "code", + "execution_count": 131, + "metadata": {}, + "outputs": [], + "source": [ + "killings['threat_level'] = killings['threat_level'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 132, + "metadata": {}, + "outputs": [], + "source": [ + "killings['manner_of_death'] = killings['manner_of_death'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 133, + "metadata": {}, + "outputs": [], + "source": [ + "killings['gender'] = killings['gender'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 134, + "metadata": {}, + "outputs": [], + "source": [ + "killings['date'] = pd.to_datetime(killings['date'], errors='coerce')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Exploratory Data Analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 135, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
count29329.00000029329.00000029329.00000029329.00000029329.00000029329.00000029329.00000029329.000000
mean146.864400739.397456724.94394662.46315925.73289214.07334082.03614247991.033619
std127.664304291.350599326.169477150.394184118.51309240.698589161.61449527783.222116
min0.0000000.0000000.0000000.0000000.0000000.0000000.0000000.000000
25%48.000000736.000000605.0000001.0000001.0000000.0000008.00000033333.000000
50%121.000000858.000000894.0000007.0000003.0000004.00000025.00000043750.000000
75%212.000000921.000000962.00000035.0000008.00000011.00000071.00000057969.000000
max986.000000999.000000999.000000995.000000997.000000671.000000999.000000250000.000000
\n", + "
" + ], + "text/plain": [ + " poverty_rate education share_white share_black \\\n", + "count 29329.000000 29329.000000 29329.000000 29329.000000 \n", + "mean 146.864400 739.397456 724.943946 62.463159 \n", + "std 127.664304 291.350599 326.169477 150.394184 \n", + "min 0.000000 0.000000 0.000000 0.000000 \n", + "25% 48.000000 736.000000 605.000000 1.000000 \n", + "50% 121.000000 858.000000 894.000000 7.000000 \n", + "75% 212.000000 921.000000 962.000000 35.000000 \n", + "max 986.000000 999.000000 999.000000 995.000000 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "count 29329.000000 29329.000000 29329.000000 29329.000000 \n", + "mean 25.732892 14.073340 82.036142 47991.033619 \n", + "std 118.513092 40.698589 161.614495 27783.222116 \n", + "min 0.000000 0.000000 0.000000 0.000000 \n", + "25% 1.000000 0.000000 8.000000 33333.000000 \n", + "50% 3.000000 4.000000 25.000000 43750.000000 \n", + "75% 8.000000 11.000000 71.000000 57969.000000 \n", + "max 997.000000 671.000000 999.000000 250000.000000 " + ] + }, + "execution_count": 135, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.describe()" + ] + }, + { + "cell_type": "code", + "execution_count": 136, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
count2535.00000025352535253525352535.00000025352535253525352535253525352535
uniqueNaN2481879268NaN261417512342
topNaNTK TK2017-01-24 00:00:00shotgunNaNMWLos AngelesCAFalsehighNot fleeingFalse
freqNaN49823631407NaN24281396394241902161117602264
firstNaNNaN2015-01-03 00:00:00NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
lastNaNNaN2017-12-07 00:00:00NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
mean1445.731755NaNNaNNaNNaN36.526233NaNNaNNaNNaNNaNNaNNaNNaN
std794.259490NaNNaNNaNNaN12.839056NaNNaNNaNNaNNaNNaNNaNNaN
min3.000000NaNNaNNaNNaN6.000000NaNNaNNaNNaNNaNNaNNaNNaN
25%768.500000NaNNaNNaNNaN27.000000NaNNaNNaNNaNNaNNaNNaNNaN
50%1453.000000NaNNaNNaNNaN34.000000NaNNaNNaNNaNNaNNaNNaNNaN
75%2126.500000NaNNaNNaNNaN45.000000NaNNaNNaNNaNNaNNaNNaNNaN
max2822.000000NaNNaNNaNNaN91.000000NaNNaNNaNNaNNaNNaNNaNNaN
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed \\\n", + "count 2535.000000 2535 2535 2535 2535 \n", + "unique NaN 2481 879 2 68 \n", + "top NaN TK TK 2017-01-24 00:00:00 shot gun \n", + "freq NaN 49 8 2363 1407 \n", + "first NaN NaN 2015-01-03 00:00:00 NaN NaN \n", + "last NaN NaN 2017-12-07 00:00:00 NaN NaN \n", + "mean 1445.731755 NaN NaN NaN NaN \n", + "std 794.259490 NaN NaN NaN NaN \n", + "min 3.000000 NaN NaN NaN NaN \n", + "25% 768.500000 NaN NaN NaN NaN \n", + "50% 1453.000000 NaN NaN NaN NaN \n", + "75% 2126.500000 NaN NaN NaN NaN \n", + "max 2822.000000 NaN NaN NaN NaN \n", + "\n", + " age gender race city state signs_of_mental_illness \\\n", + "count 2535.000000 2535 2535 2535 2535 2535 \n", + "unique NaN 2 6 1417 51 2 \n", + "top NaN M W Los Angeles CA False \n", + "freq NaN 2428 1396 39 424 1902 \n", + "first NaN NaN NaN NaN NaN NaN \n", + "last NaN NaN NaN NaN NaN NaN \n", + "mean 36.526233 NaN NaN NaN NaN NaN \n", + "std 12.839056 NaN NaN NaN NaN NaN \n", + "min 6.000000 NaN NaN NaN NaN NaN \n", + "25% 27.000000 NaN NaN NaN NaN NaN \n", + "50% 34.000000 NaN NaN NaN NaN NaN \n", + "75% 45.000000 NaN NaN NaN NaN NaN \n", + "max 91.000000 NaN NaN NaN NaN NaN \n", + "\n", + " threat_level flee body_camera \n", + "count 2535 2535 2535 \n", + "unique 3 4 2 \n", + "top high Not fleeing False \n", + "freq 1611 1760 2264 \n", + "first NaN NaN NaN \n", + "last NaN NaN NaN \n", + "mean NaN NaN NaN \n", + "std NaN NaN NaN \n", + "min NaN NaN NaN \n", + "25% NaN NaN NaN \n", + "50% NaN NaN NaN \n", + "75% NaN NaN NaN \n", + "max NaN NaN NaN " + ] + }, + "execution_count": 136, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.describe(include='all')" + ] + }, + { + "cell_type": "code", + "execution_count": 137, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object \n", + " 1 city 29329 non-null object \n", + " 2 poverty_rate 29329 non-null float64\n", + " 3 education 29329 non-null float64\n", + " 4 share_white 29329 non-null float64\n", + " 5 share_black 29329 non-null float64\n", + " 6 share_native_american 29329 non-null float64\n", + " 7 share_asian 29329 non-null float64\n", + " 8 share_hispanic 29329 non-null float64\n", + " 9 income 29329 non-null float64\n", + "dtypes: float64(8), object(2)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 138, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
0ALAbanda CDP788.0212.0672.0302.00.00.016.011207.0
1ALAbbeville city291.0691.0544.0414.01.01.031.025615.0
2ALAdamsville city255.0789.0523.0449.05.03.023.042575.0
3ALAddison town307.0814.0991.01.00.01.04.037083.0
4ALAkron town42.0686.0132.0865.00.00.03.021667.0
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education share_white share_black \\\n", + "0 AL Abanda CDP 788.0 212.0 672.0 302.0 \n", + "1 AL Abbeville city 291.0 691.0 544.0 414.0 \n", + "2 AL Adamsville city 255.0 789.0 523.0 449.0 \n", + "3 AL Addison town 307.0 814.0 991.0 1.0 \n", + "4 AL Akron town 42.0 686.0 132.0 865.0 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "0 0.0 0.0 16.0 11207.0 \n", + "1 1.0 1.0 31.0 25615.0 \n", + "2 5.0 3.0 23.0 42575.0 \n", + "3 0.0 1.0 4.0 37083.0 \n", + "4 0.0 0.0 3.0 21667.0 " + ] + }, + "execution_count": 138, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 139, + "metadata": {}, + "outputs": [], + "source": [ + "total_state_income = pd.pivot_table(data, values='income', index='state', aggfunc='sum')" + ] + }, + { + "cell_type": "code", + "execution_count": 140, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
income
state
AK14900484.0
AL22155211.0
AR18366199.0
AZ15810888.0
CA84771828.0
\n", + "
" + ], + "text/plain": [ + " income\n", + "state \n", + "AK 14900484.0\n", + "AL 22155211.0\n", + "AR 18366199.0\n", + "AZ 15810888.0\n", + "CA 84771828.0" + ] + }, + "execution_count": 140, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "total_state_income.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 141, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.catplot(x=total_state_income.index, y='income', data=total_state_income, kind='bar', height=5, aspect=14/5)\n", + "plt.xticks(rotation=90)\n", + "\n", + "# Show plot\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 142, + "metadata": {}, + "outputs": [], + "source": [ + "avg_state_poverty = pd.pivot_table(data, values='poverty_rate', index='state', aggfunc='mean') " + ] + }, + { + "cell_type": "code", + "execution_count": 143, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rate
state
AK164.602817
AL187.502564
AR205.609982
AZ221.889135
CA148.695795
\n", + "
" + ], + "text/plain": [ + " poverty_rate\n", + "state \n", + "AK 164.602817\n", + "AL 187.502564\n", + "AR 205.609982\n", + "AZ 221.889135\n", + "CA 148.695795" + ] + }, + "execution_count": 143, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "avg_state_poverty.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 145, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.catplot(x=avg_state_poverty.index, y='poverty_rate', data=avg_state_poverty, kind='bar', height=5, aspect=14/5)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 146, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 29329 entries, 0 to 29328\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 state 29329 non-null object \n", + " 1 city 29329 non-null object \n", + " 2 poverty_rate 29329 non-null float64\n", + " 3 education 29329 non-null float64\n", + " 4 share_white 29329 non-null float64\n", + " 5 share_black 29329 non-null float64\n", + " 6 share_native_american 29329 non-null float64\n", + " 7 share_asian 29329 non-null float64\n", + " 8 share_hispanic 29329 non-null float64\n", + " 9 income 29329 non-null float64\n", + "dtypes: float64(8), object(2)\n", + "memory usage: 2.2+ MB\n" + ] + } + ], + "source": [ + "data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Areas of exploration:\n", + "1. Find the correlation between poverty, education and income in the states\n", + " - which states are poor|rich, more educated, higher income\n", + "2. Top 5 states:\n", + " - with the most killings\n", + " - share of race\n", + " - level of education\n", + " - poverty|income levels\n", + " - e.g. is the race of most killings related to the share of race?\n", + "\n", + "3. Describe the average profile of a person being killed by police:\n", + " - age\n", + " - gender\n", + " - race\n", + " - state\n", + " - poverty|income levels\n", + " - education\n", + " - share of race\n", + "4. Are these killings justified?\n", + " - what is the correlation between manner of death and threat_level|flee\n", + " - did the the threat_level justify the manner of death?" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 1. Exploring the correlation between income, education and poverty rates in the various states." + ] + }, + { + "cell_type": "code", + "execution_count": 147, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
statecitypoverty_rateeducationshare_whiteshare_blackshare_native_americanshare_asianshare_hispanicincome
0ALAbanda CDP788.0212.0672.0302.00.00.016.011207.0
1ALAbbeville city291.0691.0544.0414.01.01.031.025615.0
2ALAdamsville city255.0789.0523.0449.05.03.023.042575.0
3ALAddison town307.0814.0991.01.00.01.04.037083.0
4ALAkron town42.0686.0132.0865.00.00.03.021667.0
\n", + "
" + ], + "text/plain": [ + " state city poverty_rate education share_white share_black \\\n", + "0 AL Abanda CDP 788.0 212.0 672.0 302.0 \n", + "1 AL Abbeville city 291.0 691.0 544.0 414.0 \n", + "2 AL Adamsville city 255.0 789.0 523.0 449.0 \n", + "3 AL Addison town 307.0 814.0 991.0 1.0 \n", + "4 AL Akron town 42.0 686.0 132.0 865.0 \n", + "\n", + " share_native_american share_asian share_hispanic income \n", + "0 0.0 0.0 16.0 11207.0 \n", + "1 1.0 1.0 31.0 25615.0 \n", + "2 5.0 3.0 23.0 42575.0 \n", + "3 0.0 1.0 4.0 37083.0 \n", + "4 0.0 0.0 3.0 21667.0 " + ] + }, + "execution_count": 147, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will create pivot tables that aggregate the mean of the values in the columns for each state, then concatenate them into one dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 148, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "state_race = data.pivot_table(index='state', \n", + " values=['share_white', 'share_black', 'share_native_american', 'share_asian', 'share_hispanic'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 149, + "metadata": {}, + "outputs": [], + "source": [ + "state_poverty = data.pivot_table(index='state', \n", + " values=['poverty_rate'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 150, + "metadata": {}, + "outputs": [], + "source": [ + "state_educ = data.pivot_table(index='state', \n", + " values=['education'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 151, + "metadata": {}, + "outputs": [], + "source": [ + "state_income = data.pivot_table(index='state', \n", + " values=['income'],\n", + " aggfunc='mean')" + ] + }, + { + "cell_type": "code", + "execution_count": 152, + "metadata": {}, + "outputs": [], + "source": [ + "state_data = pd.concat([state_race, state_poverty, state_income, state_educ], axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": 153, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
share_asianshare_blackshare_hispanicshare_native_americanshare_whitepoverty_rateincomeeducation
state
AK10.6225354.55211322.411268412.892958378.030986164.60281741973.194366634.670423
AL6.104274213.40341926.18461512.299145653.859829187.50256437872.155556724.249573
AR5.005545148.27356741.6303147.114603699.589649205.60998233948.611830727.378928
AZ6.50332610.960089182.332594229.121951557.541020221.88913535057.401330643.731707
CA50.93823924.346912267.30814715.869908634.827201148.69579555697.653088684.221419
\n", + "
" + ], + "text/plain": [ + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AK 10.622535 4.552113 22.411268 412.892958 \n", + "AL 6.104274 213.403419 26.184615 12.299145 \n", + "AR 5.005545 148.273567 41.630314 7.114603 \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 24.346912 267.308147 15.869908 \n", + "\n", + " share_white poverty_rate income education \n", + "state \n", + "AK 378.030986 164.602817 41973.194366 634.670423 \n", + "AL 653.859829 187.502564 37872.155556 724.249573 \n", + "AR 699.589649 205.609982 33948.611830 727.378928 \n", + "AZ 557.541020 221.889135 35057.401330 643.731707 \n", + "CA 634.827201 148.695795 55697.653088 684.221419 " + ] + }, + "execution_count": 153, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "state_data.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- Let's look at the correlation between the economic attributes of the states:" + ] + }, + { + "cell_type": "code", + "execution_count": 154, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
poverty_rateincomeeducation
poverty_rate1.000000-0.626781-0.477182
income-0.6267811.0000000.450140
education-0.4771820.4501401.000000
\n", + "
" + ], + "text/plain": [ + " poverty_rate income education\n", + "poverty_rate 1.000000 -0.626781 -0.477182\n", + "income -0.626781 1.000000 0.450140\n", + "education -0.477182 0.450140 1.000000" + ] + }, + "execution_count": 154, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "state_data[['poverty_rate', 'income', 'education']].corr(method='pearson')" + ] + }, + { + "cell_type": "code", + "execution_count": 155, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "pd.plotting.scatter_matrix(state_data[['poverty_rate', 'income', 'education']], figsize=(10, 10));" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Income and education have a positive but weak correlation. This means that with high levels of education, there is high levels of income.**\n", + "\n", + "**Income and poverty have a worthy negative correlation; meaning that high levels of poverty relate to low levels of income.**\n", + "\n", + "**There is a weak negative correlation between education and poverty. This means that low education levels relate to high poverty levels.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will look at the top and bottom five states in terms of poverty, income and education." + ] + }, + { + "cell_type": "code", + "execution_count": 156, + "metadata": {}, + "outputs": [], + "source": [ + "#poor|rich states\n", + "poor_states = state_data['poverty_rate'].sort_values(ascending=False).head(5)\n", + "rich_states = state_data['poverty_rate'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 157, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " MS 246.044199\n", + " AZ 221.889135\n", + " GA 215.406699\n", + " AR 205.609982\n", + " LA 203.341772\n", + " Name: poverty_rate, dtype: float64,\n", + " state\n", + " DC 18.000000\n", + " NJ 76.143119\n", + " CT 77.375000\n", + " WY 78.549020\n", + " MA 89.170732\n", + " Name: poverty_rate, dtype: float64)" + ] + }, + "execution_count": 157, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "poor_states, rich_states" + ] + }, + { + "cell_type": "code", + "execution_count": 158, + "metadata": {}, + "outputs": [], + "source": [ + "high_inc = state_data['income'].sort_values(ascending=False).head(5)\n", + "low_inc = state_data['income'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 159, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " NJ 78832.957798\n", + " CT 74141.520833\n", + " MD 71692.177606\n", + " MA 69822.195122\n", + " NY 68863.528428\n", + " Name: income, dtype: float64,\n", + " state\n", + " NM 29773.024831\n", + " MS 33512.030387\n", + " DC 33564.000000\n", + " AR 33948.611830\n", + " WV 34913.782716\n", + " Name: income, dtype: float64)" + ] + }, + "execution_count": 159, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "high_inc, low_inc" + ] + }, + { + "cell_type": "code", + "execution_count": 160, + "metadata": {}, + "outputs": [], + "source": [ + "high_educ = state_data['education'].sort_values(ascending=False).head(5)\n", + "low_educ = state_data['education'].sort_values(ascending=True).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 161, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(state\n", + " DC 893.000000\n", + " HI 832.735099\n", + " MA 826.004065\n", + " ME 821.261538\n", + " WI 816.635779\n", + " Name: education, dtype: float64,\n", + " state\n", + " WY 567.490196\n", + " NM 610.611738\n", + " NV 624.503817\n", + " AK 634.670423\n", + " AZ 643.731707\n", + " Name: education, dtype: float64)" + ] + }, + "execution_count": 161, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "high_educ, low_educ" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "*These states standout from the above analysis:*\n", + "\n", + "**MA low poverty, high income, high education**\n", + "\n", + "**DC low poverty, low income and high education**\n", + "\n", + "**NJ, CT low poverty, high income**\n", + "\n", + "**WY low poverty, low education**\n", + "\n", + "**MS, AR high poverty and low income**\n", + "\n", + "**AZ high poverty and low education**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 2. Top5 states" + ] + }, + { + "cell_type": "code", + "execution_count": 162, + "metadata": {}, + "outputs": [], + "source": [ + "top5_states = killings.groupby('state')['state'].count().sort_values(ascending=False).head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 163, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/plain": [ + "state\n", + "CA 424\n", + "TX 225\n", + "FL 154\n", + "AZ 118\n", + "OH 79\n", + "Name: state, dtype: int64" + ] + }, + "execution_count": 163, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_states" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**At this point, *AZ* stands out in the top5 states with most killings; and also has high poverty and low education levels.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will select economic data of the top5 states with most killings:" + ] + }, + { + "cell_type": "code", + "execution_count": 164, + "metadata": {}, + "outputs": [], + "source": [ + "top5_data = state_data.query('@state_data.index in @top5_states.index')" + ] + }, + { + "cell_type": "code", + "execution_count": 165, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
share_asianshare_blackshare_hispanicshare_native_americanshare_whitepoverty_rateincomeeducation
state
AZ6.50332610.960089182.332594229.121951557.541020221.88913535057.401330643.731707
CA50.93823924.346912267.30814715.869908634.827201148.69579555697.653088684.221419
FL14.952070123.931373147.2657954.345316709.960784160.60566448552.166667749.193900
OH6.69547336.90946520.6444446.383539815.370370135.60000048856.190123792.947325
TX9.19805451.793360308.7813397.001717718.241557168.23010945645.395535653.486548
\n", + "
" + ], + "text/plain": [ + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 24.346912 267.308147 15.869908 \n", + "FL 14.952070 123.931373 147.265795 4.345316 \n", + "OH 6.695473 36.909465 20.644444 6.383539 \n", + "TX 9.198054 51.793360 308.781339 7.001717 \n", + "\n", + " share_white poverty_rate income education \n", + "state \n", + "AZ 557.541020 221.889135 35057.401330 643.731707 \n", + "CA 634.827201 148.695795 55697.653088 684.221419 \n", + "FL 709.960784 160.605664 48552.166667 749.193900 \n", + "OH 815.370370 135.600000 48856.190123 792.947325 \n", + "TX 718.241557 168.230109 45645.395535 653.486548 " + ] + }, + "execution_count": 165, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_data" + ] + }, + { + "cell_type": "code", + "execution_count": 166, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot2015-02-01shotgun53.0MASheltonWATruehighNot fleeingFalse
14Lewis Lee Lembke2015-02-01shotgun47.0MWAlohaORFalsehighNot fleeingFalse
25John Paul Quintero2015-03-01shot and Taseredunarmed23.0MHWichitaKSFalsemediumNot fleeingFalse
38Matthew Hoffman2015-04-01shottoy weapon32.0MWSan FranciscoCATruehighNot fleeingFalse
49Michael Rodriguez2015-04-01shotnail gun39.0MHEvansCOFalsehighNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 2015-02-01 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 2015-02-01 shot gun 47.0 \n", + "2 5 John Paul Quintero 2015-03-01 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 2015-04-01 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 2015-04-01 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True high \n", + "1 M W Aloha OR False high \n", + "2 M H Wichita KS False medium \n", + "3 M W San Francisco CA True high \n", + "4 M H Evans CO False high \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 166, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 167, + "metadata": {}, + "outputs": [], + "source": [ + "killings['race'] = killings['race'].astype('category')" + ] + }, + { + "cell_type": "code", + "execution_count": 168, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 2535 entries, 0 to 2534\n", + "Data columns (total 14 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 2535 non-null int64 \n", + " 1 name 2535 non-null object \n", + " 2 date 2535 non-null datetime64[ns]\n", + " 3 manner_of_death 2535 non-null category \n", + " 4 armed 2535 non-null object \n", + " 5 age 2535 non-null float64 \n", + " 6 gender 2535 non-null category \n", + " 7 race 2535 non-null category \n", + " 8 city 2535 non-null object \n", + " 9 state 2535 non-null object \n", + " 10 signs_of_mental_illness 2535 non-null bool \n", + " 11 threat_level 2535 non-null category \n", + " 12 flee 2535 non-null object \n", + " 13 body_camera 2535 non-null bool \n", + "dtypes: bool(2), category(4), datetime64[ns](1), float64(1), int64(1), object(5)\n", + "memory usage: 173.9+ KB\n" + ] + } + ], + "source": [ + "killings.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 169, + "metadata": {}, + "outputs": [], + "source": [ + "#group state killings by race\n", + "race_count = killings.groupby(['state', 'race'])['race'].count()\n", + "race_count.name = 'race_count'" + ] + }, + { + "cell_type": "code", + "execution_count": 170, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "state_race_count = race_count.reset_index(level='race')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will select the top5 states and their counts of killings." + ] + }, + { + "cell_type": "code", + "execution_count": 171, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race = state_race_count.query('@state_race_count.index in @top5_states.index')" + ] + }, + { + "cell_type": "code", + "execution_count": 172, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
racerace_count
state
AZA0
AZB5
AZH37
AZN8
AZO0
AZW68
CAA15
CAB65
CAH169
CAN1
CAO8
CAW166
FLA1
FLB49
FLH18
FLN0
FLO2
FLW84
OHA2
OHB30
OHH0
OHN0
OHO2
OHW45
TXA2
TXB46
TXH66
TXN1
TXO3
TXW107
\n", + "
" + ], + "text/plain": [ + " race race_count\n", + "state \n", + "AZ A 0\n", + "AZ B 5\n", + "AZ H 37\n", + "AZ N 8\n", + "AZ O 0\n", + "AZ W 68\n", + "CA A 15\n", + "CA B 65\n", + "CA H 169\n", + "CA N 1\n", + "CA O 8\n", + "CA W 166\n", + "FL A 1\n", + "FL B 49\n", + "FL H 18\n", + "FL N 0\n", + "FL O 2\n", + "FL W 84\n", + "OH A 2\n", + "OH B 30\n", + "OH H 0\n", + "OH N 0\n", + "OH O 2\n", + "OH W 45\n", + "TX A 2\n", + "TX B 46\n", + "TX H 66\n", + "TX N 1\n", + "TX O 3\n", + "TX W 107" + ] + }, + "execution_count": 172, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- I will create a dataframe that contains the percentage share ofeach race killed in each of the top5 states." + ] + }, + { + "cell_type": "code", + "execution_count": 254, + "metadata": {}, + "outputs": [], + "source": [ + "race_pivot = top5_race.pivot_table(index=top5_race.index, values='race_count', columns='race', aggfunc=['sum'])\n", + "race_pivot.columns = ['sum_asian', 'sum_black', 'sum_hispanic', 'sum_natives', 'sum_others', 'sum_whites']\n", + "race_pivot.drop('sum_others', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 255, + "metadata": {}, + "outputs": [], + "source": [ + "race_share = top5_data[['share_asian', 'share_black', 'share_hispanic', 'share_native_american', 'share_white']]" + ] + }, + { + "cell_type": "code", + "execution_count": 256, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
sum_asiansum_blacksum_hispanicsum_nativessum_whitesshare_asianshare_blackshare_hispanicshare_native_americanshare_white
state
AZ05378686.50332610.960089182.332594229.121951557.541020
CA1565169116650.938239240.000000267.30814715.869908634.827201
FL1491808414.952070123.931373147.2657954.345316709.960784
OH23000456.69547336.90946520.6444446.383539815.370370
TX2466611079.19805451.793360308.7813397.001717718.241557
\n", + "
" + ], + "text/plain": [ + " sum_asian sum_black sum_hispanic sum_natives sum_whites \\\n", + "state \n", + "AZ 0 5 37 8 68 \n", + "CA 15 65 169 1 166 \n", + "FL 1 49 18 0 84 \n", + "OH 2 30 0 0 45 \n", + "TX 2 46 66 1 107 \n", + "\n", + " share_asian share_black share_hispanic share_native_american \\\n", + "state \n", + "AZ 6.503326 10.960089 182.332594 229.121951 \n", + "CA 50.938239 240.000000 267.308147 15.869908 \n", + "FL 14.952070 123.931373 147.265795 4.345316 \n", + "OH 6.695473 36.909465 20.644444 6.383539 \n", + "TX 9.198054 51.793360 308.781339 7.001717 \n", + "\n", + " share_white \n", + "state \n", + "AZ 557.541020 \n", + "CA 634.827201 \n", + "FL 709.960784 \n", + "OH 815.370370 \n", + "TX 718.241557 " + ] + }, + "execution_count": 256, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race_data = pd.concat([race_pivot, race_share], axis=1)\n", + "top5_race_data['share_black']['CA'] = 240\n", + "top5_race_data" + ] + }, + { + "cell_type": "code", + "execution_count": 257, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race_data['%_asian'] = round((top5_race_data['sum_asian'] / top5_race_data['share_asian']) * 100)\n", + "top5_race_data['%_black'] = round((top5_race_data['sum_black'] / top5_race_data['share_black']) * 100)\n", + "top5_race_data['%_hispanic'] = round((top5_race_data['sum_hispanic'] / top5_race_data['share_hispanic']) * 100)\n", + "top5_race_data['%_natives'] = round((top5_race_data['sum_natives'] / top5_race_data['share_native_american']) * 100)\n", + "top5_race_data['%_whites'] = round((top5_race_data['sum_whites'] / top5_race_data['share_white']) * 100)" + ] + }, + { + "cell_type": "code", + "execution_count": 258, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "top5_race_data.drop(['sum_asian', 'sum_black', 'sum_hispanic', 'sum_natives', 'sum_whites',\n", + " 'share_asian', 'share_black', 'share_hispanic', 'share_native_american',\n", + " 'share_white'], axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 259, + "metadata": {}, + "outputs": [], + "source": [ + "top5_race_pc = top5_race_data.astype(np.int64)" + ] + }, + { + "cell_type": "code", + "execution_count": 260, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
%_asian%_black%_hispanic%_natives%_whites
state
AZ04620312
CA292763626
FL74012012
OH3081006
TX2289211415
\n", + "
" + ], + "text/plain": [ + " %_asian %_black %_hispanic %_natives %_whites\n", + "state \n", + "AZ 0 46 20 3 12\n", + "CA 29 27 63 6 26\n", + "FL 7 40 12 0 12\n", + "OH 30 81 0 0 6\n", + "TX 22 89 21 14 15" + ] + }, + "execution_count": 260, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top5_race_pc" + ] + }, + { + "cell_type": "code", + "execution_count": 261, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['%_asian', '%_black', '%_hispanic', '%_natives', '%_whites'], dtype='object')" + ] + }, + "execution_count": 261, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "races = top5_race_pc.columns\n", + "races" + ] + }, + { + "cell_type": "code", + "execution_count": 262, + "metadata": { + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAF8CAYAAAA6gkcXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdd3hUZdrH8e85U5JJJSQhdAhNpCgoxYaIWLEgiyiKKJYV++qqq7LY3dXV1d0VX+u6qyI2LAtiAVdXaYJSpPeaQAglpE1JZuac948BBKWEZGaec2buz3XlUkIy8wOSuXPO8zz3rZmmaSKEEEIchK46gBBCCOuSIiGEEOKQpEgIIYQ4JCkSQgghDkmKhBBCiEOSIiGEEOKQpEgIkaTWrFnDjTfeyMiRIxk6dCjPP/88e3fEf/755/To0YPS0lLFKYVqmpyTECL5VFZWMmLECMaNG0fbtm0Jh8P87ne/49RTT+WKK67g2muvpWvXrqSkpHD77berjisUkisJIZLQ119/Td++fWnbti0ADoeDv/zlLwwdOpSioiIqKioYPXo0kyZNIhgMqg0rlJIiIUQS2r59O61atTrgfenp6bjdbj788EOGDh1KZmYmPXr04KuvvlKUUliBU3UAIUT8NW/enOXLlx/wvqKiIkpKSvj0009p0aIF33zzDRUVFbz99tsMGjRIUVKhmlxJCJGEBgwYwIwZM9i8eTMAwWCQp556ihUrVtCtWzfGjx/P66+/zocffsiuXbtYuXKl4sRCFVm4FiJJLV26lKeffhrTNPF6vQwYMIBly5YxbNgwBg4cuO/jXnvtNTZv3szjjz+uMK1QRYqEEEKIQ5LbTUIIIQ5JioQQQohDkiIhhBDikKRICCGEOCQpEkIIIQ5JioQQQohDkiIhhBDikKRICCGEOCTp3SSSimma1IRrCBqRzqYu3YXb4SZkhAiEAvhCPlIcKTRCRytZBLVeCPqgpiryFvSDwwWuNHCmgNMDzlRwun/+ryM18jHOlMh/HSngaRT5/VAAjBCggcMNrlS1fyFCHIGcuLaB4uJiLr74Yrp27brvfX379mX69Ol88MEHCpNZj2ma+EN+wmaYFEcKAOU15ezw7aDEW8Lmqs1s826j1FfKdt92tvu2U+YvI2SG9j3GqK6juKPZAFyvDYhuOGcKZDSFzALIbAYZBZDdEnLaQlaLyPs9jSMfFwyAGQZXOjjkZzmhjnz12USHDh0YP378vl8XFxczffp0hYnUMkwDf8iPaZqkOlOpqq1iY+VGlu9azqqyVayvWM+Gig1U1lYe9WN7g15Cbg+uaIcO1UD5psjb4ThTIkWjybFQ0A1a9oaCrpCeH7mS0XRwp4OmRTuhEL8iRULYgi/ow8TEoTlYVbaKBdsXsHr3ataXr2dD5Qb8IX9Un8twKrwNFKqBsvWRt5Wf/fx+ZyrkHwNNukCzHtCyF+R1/Pk2ljMl8v9CRJEUCZtYu3YtI0eO3PfrO++8U2Ga2AqGgwTCAVIdqWz3b2dB6QJ+2PYDi3YsYmPFRkxie4fUG/JiONwxfY56CQWgZFHkbdG7P78/rTE06QrNe0DHc6DFiWAaoDnAnaYur0gIUiRs4mC3mxJF2AjvWzBevms5M7fM5KftP7F011K8QW/c83iD3siCs134ymDjjMjb7HGR21D5naHNqdDhbGhzMujOyG0ql0d1WmEzUiSEEr6gD4fmoCxQxv+K/se3xd+yoHQBNeEa1dHwBX2RF1W7Mk3YviLy9uM/I+/L7wztBsCxF0WuNMK1kYJhp2IolLDxd4JYs2YNv/nNb/b9+v7776dPnz4KEx3a3i2mDt3B/NL5fLXpK2ZtmUWpr1R1tF/xBX1odi4SB7NjZeRt7kuRwtCyN3Q4C7oPg7Q80B2RNQ0hfkG2wIqYCRth/CE/hmnwxYYv+GzDZyzesZiwGVYd7bCapDVhyiWf4vlTM9VR4iP/GOh2KfS4Ejw5oLsi5z2EQIqEiDLDMPCFfABM2zSNyesms6B0QcwXm6Mp3ZXOjMun43o8X3WU+GtybOTq4vgrIDVbCoaQIiEazjRNvEEvuqbz9eavmbR2EvNK51n+iuFQdE3np5E/oT2WE7m/n6wKukL3y+D44ZCSETkhbsVdXyKmpEiIeguEAmiaxo/bfmTCignM2TrngJPLdrbgqvm4nusC3h2qo1hD0+MiBaPniMiaRkqm6kQiThJsdU7Eg7fWS024hndWvsNHaz5ip3+n6khRVxOuwZVRIEVir22LI29fPwydzod+v4/cmtLd0jYkwcm/rqiTmlBka+rC7Qt5Y9kbfF/yPYZpKE4VO4FQgIz0PNUxrMcIw8opkbf8znDybdD90sjhPXe66nQiBqRIiMPyBr0EjSDvrXiPiWsmst23XXWkuPCH/JFeSeLQdqyEybfB1DGRnVGn/i5yG8qdIX2lEogUCXFQ3qCX8kA54xaOY+rGqQmz1lBXvpAP0nJVx7CHmkqY+zL88ErkwN5pd0XOYWgO2RmVAKRIiH32ttkuqiriHwv+wYwtM1RHUsYb9EXODIi6M01Y903krVEb6HsznHi1tAOxOSkSgrARJmgEWbJzCeMWjmPh9oWqIylXXVslRaIhyjfB1Pvhu6fgtDuhz+g9xUK61NqNFIkkFjJChIwQM7fM5KVFL7F692rVkSyjKlgdOUwmGiZQDv99BGY/D/3+AL1GRYqFtACxDSkSSWjvlcPsrbP567y/UlRVpDqS5VTWVEKKFImo8ZVFrixmPQdnPBA50S1rFrYgRSKJmKZJIBxg6c6lPPXDU3LlcBgVtRWQ1VJ1jMRTvR2m3AXT/wpnjoWuQyIdd6UbrWVJkUgSvqCP4qpinpj7hKw51IE36CXoTo/+CFMRUbkF/nMzfPcXGPgQHDMo0idKDuZZjvyLJDhf0Ic36OXJH57kq01fqY5jGzGbcy0OtHsjfHgd5HaAsx+HdmfIND2LkSKRoGrDtYSMEOMWjuO9Ve8RMpLrnENDKZ9znWx2rYX3roDWJ8ElL0FGgZzgtghddQARff6Qn2+LvmXQx4N4e8XbUiDqwbJzrhPd5jnwQi/4+jGorYaQ+kmFyU6uJBKIL+ijvKacP878I/NK56mOY2u2m3OdSIxw5AT30g/hgr9htB+IniJXFapIkUgAISNE0Ajyr6X/4vWlr8uVQxTYfs51IvDuxKz1UlJZS9ihkZfuJi1F/k3iTf7Gbc4X9LFk5xIenv0wW6q3qI6TMBJyzrXdtOyF2WUwlz33PaWVNdx0RntuPaMDLoeG0yF3yuNFvgtsqiZUQyAc4KFZD/FN0Teq4yQcb8iLQ4qEOroDc8irvD2vlC3lAQBe+GYtk37awnOX9aBr8yzS3PLvEw9Sjm3IF/Qxa+ssBn08SApEjHiDXpxSJJQx+96Mz53HQ5OXH/D+ojI/w17+nsenLMdXGyJsJO5ME6uQ7wIbCYaD1Bq1PDr7Ub7Y+IXqOAnNH/Lj1F2RuQgy4Te+MpvBgDHc8vbSQ37Iuz8U8f26Xbw6shctG3vkqiKG5ErCJvauPVz8n4ulQMSBYRqEjCCkyXS6eDMv/Bs/bfXy3erDj8XduMvHoOdn8O9ZG/HXhuOULvlIkbC4sBHGH/LzjwX/4Jovr0mayXBWUBOuiRzqEvHT/kzMtqdzzRt1ax0TMkyembqKEf+cw/aqAIGgFItokyJhYb6gjw2VG7js08t4Z+U7quMknUAoADLnOn6cKZiDX+Tv04upDBzdNu4Fm8sZ8My3TFtWiq9WtoBHkxQJi/KH/Hyy5hOGTR7GxsqNquMkJZlzHV/m6X+gLJzK81+vrdfne2vD3PHeQn7/wSKqA0GCYVnUjgZZ7bEYwzAIhAOMmTmGrzd/rTpOUpM513HUuB2cdAvXvbqgwQ/15dJtLNi0m3+N6k27/HRZ1G4guZKwkEAoQIm3hMunXC4FwgJkznX8mINf5Nt1FSwqrojK422vqmHIi7P4dNFWuf3UQFIkLMIX9DFryyyGTB4it5csQuZcx0nXIYSadOPmCT9F9WGDYZP7PlrC41OWy+6nBpDrMAvwh/yMWziOt1e8rTqK2I/MuY6DlEzMC57jkS83EgjFZg3h3R+KWLWtin+N6k16ihOXtPQ4KvK3pVAwHKS8ppzfTvutFAgLisy5zlIdI6GZZz3ClmqYMHdzTJ9nweZyzvnbdNZur5ariqMkRUKRQChAcXUxv5n0GxbtWKQ6jjiIitoKSMlUHSNxNe0Ox1/JyDfj8/W/vaqGwS/M4sulJbJOcRSkSCjgD/pZvHMxl0+5nB3+HarjiEPYO+daxICmYQ55hY8X72TDLl/cnrY2bHDXB4t48vOVckVRR1Ik4swX9DF141RunHZjZB++sKy9c65F9JknjKImoxX3frxEyfOPn7OJa/79A9U1IUzpzXVYUiTiyB/y89ri13hw9oOETfkpxupkznWMpOfBOY9z13/WoLKJ6w8byrj0pdlU+IPSTfYwpEjEiT/k58GZD/LPpf9UHUXUkcy5jg3zvKdZtaOGL5ZsUx2FlduquHDcTHZU1coJ7UOQIhFjhmFQXVvN6K9GM3XTVNVxxFGQOdcx0PpkzGPOY2QdG/jFQ/FuPxc8P4PNZT5qpEHgr0iRiKGwEaaitoIrPruChdut800h6kbmXEeZ7sQc8gr/nLONHdW1qtMcYJe3lsEvzGJ5SSV+2fl0ACkSMRI2wpTXlHPFZ1fICWqbkjnX0WWecgdVjmz+/PlK1VEOqromxPBX5zB3Q5lskd2PFIkYCBkhdgV2Mfyz4Wyp3qI6jqgnmXMdRdkt4fR7Gf3+CtVJDqsmZHDdGz/yxZJtUij2kCIRZcFwkB3+HQyfMpxtXvULc6L+ZM519JgXPc8Pm6v4fl2Z6ihHZJhw98RFjP9+kxQKpEhEVTAcpNRXyhVTrpBDcgnggDnXov46nYvRsi/XvRXdBn6x9uQXK6VQIEUiamrDtWz1buXKz65kV2CX6jgiCmTOdRS4PJgXv8DT3xThteGL7ZNfrOTD+cVJXSikSERBbbiWoqoirvzsSnbX7FYdRxxEaEqI2pdqCU4MYoYjJ2xNv0nwzSBm6NAnbvfOuT5zcj5Ldv361tPcUjcXfl7/wUTHvNuUsprEvVIxz/gjO2pcvDJ9veoo9fbQpGV8tjh5+z1JkWigYDhIqbeUq7+4msraStVxxEEY2w3MKhP3zW60DA1zfaQohL8L4zjFgeY89It0IBSADBlhWi95naDPDYx6W03rjWj6w0eL+Wbl9qQsFLIq1wB7t7le/aUUCCvTHBqEiFxB1AIOMEojhUPvePifk/wh/77bTe+sSWPljy5qw3BtZx+Xtj+w99aGSgePzcvCG9LY4XfQOSfI308tJ8UBi3a6eGJBJv6QhkuHP/So4uSmP58V2OHXufZ/OVzRwc+ITvFreBdL5iUvM3VlGctLqlRHaTDThN+99xOvXHUip3bIw+N2qI4UN3IlUU+maVIdrOaaL69hp3+n6jjiMLRcDa2VRvCfQXCBVqgR/jqMc+CRf0baf851igM+OW8X/xqwm+cWZbCm4sDP/2BdGpcU+vngnDKmXbSD4moH325NIWjArTMacWtXL1MG7eLxPhX8eUEmxp67XKU+B6O+yeHGLt7EKRDHXU4wtxN3vL9YdZSoCRsmN709nx83luFPopPZUiTqyRfycd3U6yiqKlIdRdSB8wwn7tFunIOcGEsNtOYapEDokxDB94IYGw7et2f/OdfDO0RewAvSDE5tVsv32w7s63Rvjyoapxq8tjydR37MYrvfgS+osbrcia7BGS1qAOjWOMSng3ah77nL9dvvcvA4TS5qE4jRnz7OUhvBoGf445T11MZo2pwqIcPkhjfnsbS4gkCSFAopEvXgD/m55b+3sHr3atVRxFEya0yM+QaOUxyEfwyjddBwDnUS/urg3/D7z7nW91u6MExw6gcueP9+djYfrE2jRXqYUZ19dG0cxETDof16F+3qcid7Xz8f612JrsG/V6ZF7c+pknnOE2wqDzFxfrHqKDFRGzYY+a+5rC6toiaU+IVCisRRCoQC3PPdPSzYvkB1FFEP4Rlh9D46mjuyToEOaGAGD77Daf85159siMyW2OrV+X6bm5MLDuw/NLMkhVu7VTNozxXBol0uwia0ywqhAbNKIlcey8qcXPNNzr7bTT3yannqpApeWpbB6nKbLxM2PwG6DWXEm/Y6E3G0AkGDq16fy67q2oRvM27zr8j48of8PDHnCaYXT1cdRdSDudPE3GHiPCvyZe/o6SD0SYjw9DCOUw++EBmZcx0pEjVhGPJlLkEDxvaqojArzHb/z5931/FV3DqjEWlOkwyXSe8mtWyucuB2wLjTdvPnBVk8/ZOGSzcZ16+c/dc+22WFuaVbNfd+n83Ec3Zhy3VRTccc8grvLtzBlt0JcuvsMCr9Ia54bQ5Tbj+NzNTE/XlbM2UsU534gj7eWPYGLy16SXUUEUe39riVm7K6wpsXqo5ieWbfm/D3G0OXJ2aqjhJXfQob8+a1fRJ2x5NcSdRBIBRgTskcKRAJLLw0jDHn17cN3kp9C+28cxmtIJOtZBTAmQ9y+7vWbuAXCz9sKOPBSUt4bHA30tyJ95KaeH+iKAsZIbZWb+W+6fepjiJiyNHNgaPbr38SPK/TeVzdYRj833MKUtmHecGzLN3m4+uV21VHUeLD+VvoVJDJVSe1SbhCkbg30qLEG/Ry41c3Eggn/j1W8Wsy57oOCvtjtjuTa95cpDqJUk9+sZK568sSbmusFInD2LvVtdRXqjqKUETmXB+Bw415yUv836ytlPmsNW0u3kwTbp4wn+Ld/oSaly1F4hD27mRavDNxToyKoydzrg/P7Hc35WYaz06TM0MQ2Rp75WtzqAqESJQ9QVIkDsIf9PPx6o+ZvG6y6ihCMZlzfRg5beGUO7ghCRerD2d7VQ3X/OsHAsHEuJqQIvELteFaVpSt4Ol5T6uOIixA5lwfmnnxC8zaUMn8TdIe/5eWbKngr9NW4quxf9dYKRK/4Av6uPN/d2KYifFTgGiYeMy5XrTTxcivGwOwqcrBFV815sr/NubhH7P2ncp+6IcsLpvWmP9siCyiV9Vq3DM7O6a5DuvYiwg368GNExaqy2Bxr8/cyLxNu23fukOKxH78IT+/+9/vZHCQ2CfWc65fW57O2B+yqNnzOvLkwkzuPK6ad84qwwS+Lk5hd43GzoDOe2eX8dH6SH+nV5anc2MXb8xyHZY7HfPCf/DEtM34auWHqcO5/d2FVAfsfTUhRWKPQCjA60tel55M4gCxnnPdOjPEuH7l+369rMxFnyaRXUKnN6thdqmbFAeEzEhbELduUlTtwB/S6NRIzYuPOfAhSnwab8zeqOT57aTCH+S3b83DX2vfqwkpEoAZDOI2dUYUXkqbzDaq4wgLifWc63Nb1bD/YDyTn+tRutOkqlYnzWlyZosafj+7Ebd1r+bFpelcfYyPJ+Zn8ucFmfhCcRx/2qQL9Lyaq9+SXX91tWBzOeO+WYPXpusTUiQAIxBg3TnnYU6exqTzPmBE5xGqIwkL2TvnOh72/4b0hjSy3JHbOcM7+Hnp9HJME1pnhvm+1E2v/FpOyKtlysY4HfbTNMwhr/DpsjLW7lB0q8umXvpuHUu3VthyvkbSFwnD72frvX8gVFJC6Z+fZMsdv+PeY2/hzbP+hVuXQ1QivnOuu+QEmVsa+bqbXpJCr/zgAb//xqp0Rh3jJRD6eU5FvK4kzB5XUZvdljsnJvfJ6vowTbj57QW2nJGd1EXC8Pup+HQK1d9+u+993pkzWX/BhXTe7mT64Gl0ze2qLqCwhP3nXMfafT2rGLckg8unNSYYhnNb/dwO5rNNqQxoHsDjhPNaB3h9ZTpvrkrn/NZxaBnjyYFzn+SeSWtJ8PEJMVPmrWX0+Pm2W59I6lbhwa1bWXf+IMyaml//pq6TO3o0uTf+ln+ueYsXFr4Q/4DCEj68aCLHzP03zH1ZdRRlzCGvsiZ/IOc8P1d1FNsbe8GxjOjbGo9NGgEm7ZWE4fez5fd3H7xAABgGu156iaLrruP6FpfyyXkfkO5Mj29IYQn7z7lOSq36YB57Ede8IbeZouGZqauo8MfvttOrr77KaaedRs2e17pnn32WkSNH7nvr2bMn77zzziE/PymLhOH3U/6fSfh/OvKIRf/Cn1h3/iAKVu3gu8FTOaX5KXFIKKxk/znXSUd3YA55hTd/LKWkUjohR0NNyOCO9xbij9P6xKeffsqgQYP47LPPALj77rsZP34848eP57LLLqNdu3YMHTr0kJ+fnEXC62X7M8/U/eMrKym+cTRlz/2DF0/7Ow/1fSiG6ezP2GIQfPvABdfwsjDBN39+X+jzEME3goSXRO7PmgGT0CRrLurtP+c62Zgn3YrXlcujny5XHSWh/LChjCmLS2LeVnzu3Lm0bt2a4cOHM2HChAN+b9myZfztb3/jhRdeICUl5ZCPkXRFwvD72XLvHzB9vqP+3PJ33mXT5ZczOPMUpl04hbzU+Cxm2kn4+zDhz8Ow3+u9UWpg/GREDgEAps/E9Jk4r3FiLDL2fZ5+sjW/HCNzrrNUx4i/rOZwxv3c/MFK1UkS0qOfLscX40XsiRMnMmzYMNq1a4fb7WbRosgtw7KyMu666y6efvppmjVrdtjHsOZ3ZYwYgQBV//0a3/ff1/sxalavYf2FF+GZtYhpF37KoMJBUUxof1qOhnPozwtyps8k/L8wjrP3m/rmBMJECokTzHITakFvYs0vx4raCkjJVB0j7swL/86C4mpmrNmpOkpCqq4JcfcHi2K2LbaiooLp06fz1ltvcf3111NdXc3bb79NOBzmrrvuYtSoUfTq1euIj2OP5fUoMWtq2Pb44w1/HL+fkj/cR9YFg/jz449zTquzuXP6XVFIaH96Zz3yog+Yhkno8xDOs5wHfKVpbg29o07oPyEcpzkIzwzjOMVBaFoINHD0d6C543iK+Ai8QS9BdzpJNVWiw0CMNqdx7ZMzVSdJaP9btZ0Zq3cyoHM+buevx+c2xOTJkxk6dCj33RcZvez3+xk4cCBjx46lVatWXHnllXV6HGv+6BYDhs/H1jF/xKisjNpjVn72ORsuHsypoTZMH/yVtPT4BXObCWUQ+jJE6D8hzJ0moa8iPzU5TnDgGhZ52dVyNIyNBnorHb2ljrHMWhvxvUEvIbdHdYz4caZiDn6Rv31bTKXNm9PZwQOfLKEmBiexJ06cyODBg/f92uPx0L9/fz7++GPWrl17wA6nt99++5CPkxTnJIxgEO/s2RSPvik2T+ByUXDvPWRfeinPLHmeCSsnHPlzEphZbhL6TwjXKNdh3wcQ/CiI82InxkIDLVsDE8xKE0ef6P5U1RAXFF7Ag91Hk/6PHqqjxIU58GF2dbuWXn+ZrTpK0hjcozlPDulOWor1bu4kx5VEOMy2Rx+L3eMHg9LSox7Cy8LoHXU0l4beWSc8N0z4hzD6sdb6skyqOde57eGkm7h2/DLVSZLKpJ+2smRrBWELHme31ndjDBh+P2Xj3ya0dWvMn0taekRojbRfXTEc7H2Org4cx0WuGLQsDdfVLlxXu9AyrbMeAck159oc/BLfrK5gydYK1VGSzpiPl1Abtt6NnYQvEmZNDTtfeiluzxfasYNNI67C++8JTBj4b27reVvcnlvERtLMue42lFB+F25+98iHTEX0rdvh5aP5xTE/O3G0ErpIGF4vpU89Va8zEQ17YmnpkUiSYs51ShbmBc/y0BcbbdnOOlE8PXUlobC1/v4TtkiYpkmwpISKSZOVZZCWHokhHnOuVTPPfoyiSoN3f9isOkpSq/SHeHrqKksNKErcIhEIUPLgg5FG7grta+nxt+elpYdNxXrOtXLNjofjLmekTJuzhAlzN7PbV6s6xj4JWSTMYBDvrNn4F1rn3mr5hHfYdPlwaelhQ7Gec62UpmNe8gofLdrBpl1xvi0rDipsmDw8eZllriYSs0iEw2x74gnVMX6lZvVqaelhQ7Gec62SeeK1BDJa8IdPlqqOIvbz9YrtrN/hxQrH2BKuSBiBAOXvf0Bo2zbVUQ5qb0uP7Q89zJ97PczfT/+b6kiiDuI55zpu0vPh7Me486PVMm3Ogsb+Z4kldjolXJHANNn5svUniElLD3uJ55zreDEHPcOK7X6mLi9VHUUcxKLiCmav26V8t1NCFQkjEGD3+x8Q3r1bdZQ6CRYXs+E3QzEnT2PSeR8wovMI1ZHEIcRzznVctDkVs8M5XP2mddbtxK/96bMVhAy1t5wSqkhgmux65RXVKY6OtPSwBV/IB2m5qmNEh8OFOeRlXv1+GzurrbOLRvza+p1eZq/bqbRdR8IUCbtdRfyStPSwtkSac22e8jsqtSye+lKGCdnBs9NWUxtSdzWRMEUCsN9VxC9ISw/rSpg5141aQ7+7+e17Mo7ULpZtrWRRcTmGottOCVEkjECA3e+9b9uriAP8oqXHx+e9T5ozTXWqpJcoc67Ni55nzqZKftiQAN8rSeSZqasIhNTsdEqIIgH2v4r4pb0tPZqu2sn0wdOkpYdiCTHn+pjzMVr05obxi1QnEUdp/qbdrNtereS5bV8kjJoaym28FnE40tLDOmw/59qVhnnxOJ78ejPeGM1UFrH1ly/V9HSyfZEA2PX666ojxJS09FBv75xruzIHjGV7wMk/Z2xQHUXU08y1OympCMT9eW1dJEzDwDtrNqHt21VHiTlp6aGWredc5x8Dva/jmreWqE4iGujpL1dSHeerCXsXiUCAXa++qjpG3EhLD3V8QR+GM1V1jHoxL3mZL1fsYmVpleooooG+WlFKeZw7xNq6SARLSvD/lHwnRqWlR/zZdc61efwVBHM6cPt7slidCEwTnv96bVzXJmxbJMJeLztfTqwdTUdDWnrEly3nXHty4Py/cN+n65Fhc4nj00Vb0ePYtt62RYKwQeWXX6pOoZa09IgbO865Ns/9Mxt2h/hk4RbVUUQU+YNhJi/aErfGf7YsEkYgQNnb4yEYVB3FEqSlR+zZbs51ixOhyyWMlAZ+CenfszYSlCJxGJrG7gnvqE5hKdLSI7ZsNedad2AOeYUJ87ezpTz+WyZF7K3cVvo/Y4IAACAASURBVEVRmT8uz2W7ImEaBtXffUd41y7VUaxHWnrEjJ3mXJt9RuNLacLYSctURxEx9OqM9XFZwLZdkTB8Pna//bbqGJb2y5YeJzc/WXUk27PNnOvMpnDmWG79cJXqJCLGpizeGpcvR9sVCbM2iO/HeapjWN7+LT1eOu0f0tKjgewy59q84DkWbfXy7aodqqOIGAsEDT5ZGPsFbFsVCaM2SPlHH0Y2C4s6kZYe0WP5OdftBmAWnsGotxaqTiLi5I1ZGwmGY/t6aKsiQThExccfq05hO9LSIzosPefamYJ5yYuMm7GFcp808EsWa7ZXs3GXN6bPYasiEdyyhdoNG1XHsCVp6dFwVp5zbfa7h91GGn/77xrVUUSc/Wvmhpj2c7JNkTB8Pna/I9teG0paetSfZedcN24HJ9/G9ROWqk4iFJi6fBsuR+xWsG1TJHA4qPz8C9UpEoK09Kgfq865Ngf/H9PXVbCwqEJ1FKFApT/Eohj+29umSPjmzSNcXq46RuKQlh5HzZJzrrsMJtzkOEZPkJPVyWzi/KKYnZmwRZEIV1dT/u57qmMkJGnpUXeWm3PtzsC88O88OnUDAengl9SmLSvF5YjNy7ktioTmclE9Y4bqGAlLWnrUjdXmXJtnPcJWL4yfs1l1lAbTyjbhmvF/kV/UVOH8/nVc01/A9d3zUL0TAOfCD3B9+3f0zT9GPi7oxzlPDtYCVPiDLNkSm1tOtigSvvkLMGtqVMdIbNLS44gsNee6oBv0uIqRby1WnaTBHKu/wbnwfQhHGnY6l07BaHUiwdNvI9TlfPTq7VDjhZpqgv3vwLHph8jnrfqacKeBKqNbygfzYnPLyfJFIuz1UjnlU9Uxkoa09Dg0y8y51jTMIa8waelO1u+I7R75eDDTcwn1vXbfr7VdG9D85bhmvoSjaAFGXntwOMEIQzgUadnu3YUWrsXMaqYwubVMW7YNZwx2OVm+SGhOJ9Xffqc6RlKRlh4HZ5U512bPq6nJbM3vP7T/VQSA0eJ4TM2x79earwzTnUbwtJsx0xrhWP0NOFMwmnXF+eN4Qp3PxbnyK0Lt++FY9DGOxf+BkNxp2O0LsnxrZdQf1/JFonbTJsJlZapjJKUDW3p8mvQtPSwx5zqtMZz7J+75z1qMRF2rdqdjNI1soDCadkUrL4r8f+EphE6+HjAxM3LRd6zBzGuPmVuIXrRAYWDrmDivGF+UbzlZukgYNTVUTvlMdYyk9nNLj8VJ39LDCnOuzfOfYfWOAFOWlCjNEUtGbiF66QoA9J3rMTObHvD7jrXfEW7fHy0c3NeVVwvXxj2nFU2NwS0nSxcJwmGq/vtf1SmS3gEtPU58KGlbeiifc936JMxjBnH1W4vUZYiDUPeLcWyeh+u759G3ryR8zFn7fk8vXhi5ynC6Cbc4Hseab3Gsm064xfEKE1vHLm8tm6M8jEgzTeu2VA3u2MHafqerjiH242rZkpYvvYg/P5OR39zApqpNqiPFTdfcrrw28EUy/1IY/yfXnZi3z+dfSw0e/2xF/J9f2MYD53fm+n6FOPXoXANY9krCDIepmjpNdQzxC8nc0kPlnGvz5NuoduRIgRBHNH3NDny14ag9nmWLhOHzUf2/b1THEAeTpC09lM25zmoB/e9j9PtSIMSRzdu4m1Sn48gfWEeWLRJ6aiq+hdKPxsqSraWHqjnX5kXP82NRFbPXyVx3cWQ1IYNV26K3FdayRaK2qAjT51MdQxxBMrX0UDLnuuM5GK1P4vrx8gOTqLtpy0upjVI/L0sWCTMcll5NdpIkLT3iPufa5cEc/AJ//V8xVQGZNifqbuaandSEorMuYckiYfh8+L6fozqGOErJ0NIjnnOuzf73s7PWzUvfrovL84nEsXhLRdS6wlqySOgpKfjmz1cdQ9RDorf0iNuc67yO0Hc014xfEvvnEgknbJj8VBSd+TuWLBLBbdswqqpUxxANkKgtPeI159oc/BL/XbWb5SXyfSDqZ9rybQSCDb/lZLkiYRoG1bNmqY4hoiARW3rEY8612X0YobzO3PpeYp+sFrH1/bpdhIyGn5W2XJEwvF58s79XHUNESaK19Ij5nOvUbBj0V8Z+tj5qu1NEclpdWo07CusSlisSWkoKvnnzVMcQUVb52edsGHwJp4baMH3wV7TJbKM6Ur3Ees61efbjbK4I8/684pg9h0gOYcOkqKzhxwgsVySM6mrCu3erjiFiIBFaesR0znXzntB9GFe9KbeZRHQs2Nzw11LLFYmalStVRxCxZPOWHjGbc63pmJe8zPs/7aBod3S7eIrkNX/Tbny1DTtjY6kiYYZCsvU1Sdi1pUdFbQVmDOZcm72uJ5DWjPs/Xhr1xxbJa9nWSsINXLy2VJEw/H4CS5epjiHixI4tPSIjTKM85zqjCZz1CLd/tDq6jyuS3urSKlJdDWv2Z6kioblcBFZIp8ukYrOWHrGYc20OepZlpX7+u2J7VB9XiJqQQUlFoEGPYakigWEQ2i7fKMnILi09oj7num0/zPYDueZNaeAnYmNxA09eW6pI1G5Kniln4tfs0NIjqnOuHS7MS17i5dkl7PLKjGYRGz9u2o2/ASevLVUk/EukT42wdksPb9CLGaU51+Zpv6dSy+Tpqaui8nhCHMzSLRUEG3Aw0zJFwvD5CCyVnR0iwqotPaI2wrRRGzj1Tq5/Z3nDH0uIw1hZUonHXf/Fa8sUCTMUombdetUxhIVYsaVHtIqEOfgFZm+sZN4mOTgqYstbG25QixfLFAnN5SJYVKQ6hrCgys8+Z8MlQyzR0iMqc647X0C42Qn89u2F0QklxBFsq6z/DidLFYnQjh2qYwiLChYVWaKlR4PnXLvTMS96nie/2oyvVhr4ifjY3IAeTpYpEuHdu8FseFtbkcAs0NLj5znX9fvWMQc8SKlf5/VZG6KcTIhDW1NahVnP11fLFIng1hLVEYRNqGzp8fOc63rMlGhyLPQaxdUybU7E2cadvnoPILJMkajZKD9ZibpT2dKjvnOuzSGvMGX5LlaXVscglRCHVrTbRzBs4ysJMxSidu1a1TGE3Shq6VGfOddmjxEEswu5631pAy7ir3i3H4eu1etzLVEkjECA4JatqmMIm4p3S4/InOujKBKeHDjvKe6dvA4ZNidU2LLbT6qrfi/3ligSmCa1xTKJS9RfPFt6ROZcN67zx5vnPcW6slom/SQ/CAk1asMGVYH6zZWwRJHQXC6CW7aojiESQDxaehzVnOuWvTGPvZiRb8htJqHWtnp2g7VGkXC7CZeVqY4hEkSsW3rUec617sAc8grj55VS0oDDTEJEQ9Hu+p2VsESRMAMBOSMhoiqWLT3qOufa7HszXnceD0+W/kxCvV3V9es0bIkiYXi9qiOIBBWLlh51mnOd2QwGjOHWD2Rmu7CGndU19TpQZ4kiEa6sUh1BJLCfW3p8FZWWHnWZc21e+HcWbvHy3ZqdDXouIaKl3B8kGD767XXWKBIVDZucJMQRBYOU/vnPUWnpccQ51+0HYrbtx6g3pYGfsI4Kf7BeB+qsUSR2S7tkER/RaOlx2DnXzhTMwf/H378rprKeWw6FiIVKf4iwYdMiEdq5S3UEkUQa2tLjcHOuzdP/QFk4lee/kQ4CwloqA8F6fZ7yImGaJqGd0iJcxFkDWnoccs5143Zw8i1cN2FZlMMK0XCVfrsWidpawuWyJiHUqE9Lj0PNuTYveYn/ralgUXFFLKIK0SAV/mC9+jepLxKhEOGKStUxRBI72pYeBx1h2nUIofyu3PLOTzFMKkT9VfqDuBxH/5KvvEhgmpi19TvkIUQ01bWlx6+KREom5gXP8fCXGwlIBz9hUVU1IZwOG15JYJpg1G8YhhDRVpeWHr+cc22e9SjF1fDO3M3xjCrEUTFNCNbjhxj1RQIwQ1IkhHUcqaXHAXOumx4Hx1/B1W9KAz9hfYYtT1ybJoSlSAjrOVRLj5/nXEca+H28eCcbdtV/0LwQ8VKPYxIWKBKAKUVCWNTBWnrsnXNtnnoHNRktufdjmVkt7KE+VxLOI39I7EmREJa2p6VH9fTp3Pvcs5zTciC14Vqc/e/nrveXYchatbCJ+jTb1sz6tAWMonBlJcW33Ybvhx9VxhCiTpz5+bR4/h+kHnssQU1nydZq1ZGEqLPjW2XjcjqO6nOscSUhC9fCJkI7drD96adp8cbbLPxyo+o4QhwVrcWR56D8kiWKhFyvCzvJv+8BVs4pYf4Xm1RHEeKo9Dy7NU730V1JWGLhWgi70LOySOnSlUX/LVIdRYijpml2PEwHaCkpqiMIUScF999P6YYKKnb4VUcR4ugdfY2wQpHQ0NMO0ZtfCIvxnHM+87+Uk9XCnux5JaFr6J66tWgWQqVGV11FbS0UrShTHUWIetHs2AVW03W01IMPcBHCShpdfyMLp8litbAnTddwOG1YJNB1ud0kLC/1+ONxNs5h5ffbVEcRol7cqQ7CIRv2btKcTnSPFAlhbU3G/JGVc7YRrJEzPcKe3B4nRtiGXWA1pxM9LV11DCEOSc/OJuXYLrLtVdhaSpoTsx4d/pQXCQA9I0N1BCEOqeD++2Tbq7C9FE/9zk5bo0hkSpEQ1uU5W7a9Cvtzpzntek4CnDmNVUcQ4qByRsq2V5EYUjwum56TAJwFBaojCHFQ2dePZsFU2fYq7C8lzYluyxnXgDNXriSE9aT26IEzpxGr5si2V2F/KWlOHM6jf8m3RJHQs4++fa0QsZb/wBhWfi/bXkVi8GS67Xu7SdM09HRpzSGsQ2/UiNRju/CTbHsVCSKjUf0aqVqiSBg1tTibNFEdQ4h9Cu6LbHut3CnbXkViyMqv36FlSxQJjDDO/HzVKYTYx3P2ecz7UhasReJIt/OVBLouRUJYRs7IkdTWmBSv2K06ihBR4XDpuFKObiLdXpYoEprbLUVCWEb29Tcyf5ocnhOJI7NxKqHa+m3AsESR0N1unM2bq44hBKk9I9teV8u2V5FAMnNT69W3CSxSJABSO3ZUHUEI2fYqElJm41R0R/1e7uvX8SkG3IWFqiMoETJNxmGyHQgBw9DIBV7GxAkUAjegoWsaL5oGG4Hz0RigaXhNk1cxuUuzTK23Nb1RI1I7d+Gnx35QHUWIqMrO9+B027xIOPPywOmEUEh1lLj6DsgE7tJ0Kk2T32OSDfwWjc6axgTTYDpwgmlSDjyFxoOYDEDjI0x+U5+OXeKgCu6/j23rZdurSDw5zdLrdZAOLHS7yaipwd2ypeoYcXcKMGK/F3oHsAvovOcftDMaKzBxA2EgCLiBUtMkALSp5z+8+DXPWeczX7a9igSUnVf/EdGWKRKEDdyFbVWniDuPpuHRNPymydOYXIlGAbDUjCwy/UikGKRqGn3QeBaTy9H4AJOL0HjNNHjdNAiY9VuUEhE5V4+ktsageKVsexWJJz0nAYqElpqKu7Cd6hhK7DBNxmJyBhr9NY079txKetw0yAay9nzcuZrGmD3rD03RWAx0ReNYNKarCp8gsq+Tba8iMaWkO3G66v9Sb5kiobtdpHbtojpG3JWbJo9icjUaZ+25dTQPuB2NBzWdKuD4X6w7TMLkYqCGn/8BA3HMnGhSe/aUba8iYeW1zKz3GQmw0MI1QEqnTqojxN2HmFQDH2DywZ5bRoPReAyTFNOkO9Brv3WHGaZJbzRSNI1TTJO/YqIB98gCdr3lPzCGFbNLZNurSEh5LTNwuOt32hosViTcLVqojhB3N2g6Nxzk/X0O8aLfb7+CkadpPCXFoUEi216PZdFjc1VHESImmnXIxlmPORJ7WeZ2EwAOB468PNUpRBIpuP9+StZVULlTbtiJxJTfOrNBn2+pImHW1uLp3k11DJFEPGedJ+NJRcJyOPV6d3/dy1JFQk9Lw9Ojh+oYIknkXHO1bHsVCS2nWRqhWqNBj2GpIqE5HKSfdJLqGCJJZF97I/PlKkIksPxWmTT0vK2ligTs2eEkp4hFjHlO6ImzUTarZNurSGAFhVm4Uxu2P8lyRQLDwN22jeoUIsHl3x/Z9trQS3EhrKygMOvIH3QElisSJuA57njVMUQC03NySOl8LIu+KVIdRYiY0XWNRgVpDX+cKGSJKkd6Op7evVTHEAlMtr2KZJDfJhMj1PCebpYrEgBpJ0qRELGTdta5LJBuryLBteycg6MBPZv2smSRcLdsgeZ2q44hElDjUdcQCBgUr5JtryKxte2eh6MBJ633smSRMAIBPD1kXUJEX9aoG+QqQiQ83aGR16phJ633PVZUHiXKdI+H9FNPUx1DJBjPiSfibNSIVXNl26tIbE3aZBIORWfnniWLhOZ0knHmANUxRILJv/8B2fYqkkLLzo0bNENif5YsEgApbduip6erjiEShJ6TQ8oxnfnpa9n2KhJf2+Oisx4BFi4SRiBAWp8+qmOIBFFw/wOUrCunapdsexWJTXdq5LXIiN7j1eWDioqKuP322xk5ciTDhw/nkUceobq6mnHjxvHuu+8e8LGXXXYZxcXFDQ+Wnk6m3HISUZJ21jnM/1LGk4rEV9A2i1CU1iOgDkOHAoEAt9xyC0888QTHHx/ZcfTJJ59w9913061b7Np6a7pOxgApEqLhGo8aRcBvsEW2vYok0OrY6K1HQB2uJL799lt69+69r0AADBkyhN27d1NUFNv7u3p6Ou62bWP6HCLxZV97g8yMEEmjU++CqK1HQB2uJIqKimjduvWv3t+yZUtKSkpYtGgRn3/++b73r127NmrhANL79aN248aoPqZIHp4TT8SRnc2quUtURxEi5jJzUxs8ZOiXjlgkCgoKWLx48a/ev3HjRjp06MCgQYO44oor9r3/sssui1o43eMh+4JB7B4/PmqPKZJL/gMPsHyWbHsVyaFdj/yoP+YRr0kGDhzI7NmzDygUEydOpHHjxrRq1SrqgX4p5dhjcTRqFPPnEYlHz8khpVNn6fYqkkbnk5vidDui+phHLBLp6em8/PLLvPjiiwwfPpxhw4axaNEinnvuuagGOaRQiMyzzorPc4mEUvDAA2xdK9teRXLwZLrIKYj+2TLNNM2G95KNMd/ChWy64krVMYTNtJ//E5+/ulx2NYmk0OW05pw2rAOulIZNovslyx6m219ql65yy0kclcbXXivbXkVS6Xxy06gXCLBJkTBDQTLPPlt1DGEjWaOuZ75sexVJwu1x0qR1w0eVHowtioQjPZ3sob9RHUPYhKdXL5zZ2ayWbq8iSbTtnks4HJsdfLYoEgCpXbrILSdRJ/n3y7ZXkVyOOakp7tTo32oCGxUJMxSSW07iiPTcXFI6HSPbXkXSSEl30rxj7H6Atk2RcKSl0WjYMNUxhMUV3H+/bHsVSaVz36aYMbxotk2RAEjp1BFXixaqYwgLSzvzbBlPKpLKcWe2wpUS3QN0+7NVkUDXabRfCxAh9tf4uusi215Xl6uOIkRcNGmbiSfDFdPnsFWR0N1uci4bBs7YLNAIe8u65jrmy1WESCLdz2iJwx3bl3FbFQkAdJ3MM85QnUJYjKe3bHsVycXp1ml/QhN0XYrEARwZGTQedY3qGMJi8u8bE9n2GpRtryI5dDixCaYR+65KtisSAKnduuFs3lx1DGERkW2vnWTbq0gqx5/ZKmZnI/ZnyyKBppFz+eWqUwiLKHjgAbaukW2vInk0KkgjuyAtLs9lyyKhp6SQM/xycMRu25ewj7QBZ0mfJpFUup3eAl3X4vJctiwSADgcZPTvrzqFUKzx9dcT8BtslW2vIkm4Uhx0Oa15VOdYH45ti4QjI4P8225VHUMoli3bXkWS6dqvBRC/MUC2LRIA7rZt8fTooTqGUMTTuzd6VpZsexVJQ3donHhem5jMjTjkc8btmWJAS00l/47bVccQiuTfP4blM7fKtleRNDr2KkB3xmctYi97Fwldx9OzJ+527VRHEXGm5+aS0rEji78pVh1FiLjpc3FhXLa97s/WRQIAl4u8W29RnULE2b5tr2Wy7VUkhzbdcklNj22fpoOxfZHQnU4yBw7EWVCgOoqII9n2KpJNn4vifxUBCVAkANB1cm+4QXUKESeNr78evy8s215F0ihom0VO03Qlz50QRUJ3u2l06VD0rNgMAhfWknXNdXIVIZJK7wsLcbjUvFwnRJHYq/HVV6uOIGLM06cPjqws1swtVR1FiLhoVJBGi06N4nbC+pcSpkjoHg+5112Lo1HsZr0K9fLve0C2vYqkcsrQDugONQUCEqhIAKDr5N12m+oUIkZk26tINnmtMmjZOQfdoe6lOqGKhJ6aSqNLh+Js1kx1FBEDBWPGsGX1btn2KpJGv8s6xq1H06EkVJEA0BwOmtxzt+oYIgbSzhjIgqmbVccQIi6adcgmv3WWsrWIvRKvSLhcZA4ciLt9e9VRRBTl3nBDZNvrGtn2KpLD6Zd3whnj+dV1oT5BDGguFwV/HKM6hoiizKuvlW6vUWSYBv9d8RYT5z/Dhwuepdy/Y9/vTV8zkSVbpu/79TerJvD+/L+wYtscAGpCfqYu/3fcMyeTdj3yyc73oGlqryIgUYuEw0Faz554ekqH2ESQ1rcvjqwsVv8g216jZcPOxQAMO/FeTiq8iBlrP8RXW8WkReNYv+f3APzBany1VVx2wr0sL5kNwLxNX9Kr9TlKcicD3aFx+vBOuBScrj6YhCwSEOkQ23Tsg6pjiCjIu+9+ls/cSli2vUZN+/wenHnMCACqArtIc2URDNfQt+2FdG7ad9/HOXUXhhEmZIRw6C4q/DsJhmvJzWihKnrC69a/Ba5U60zdTNwioWm4C9uSea78xGNnem4uKR06sujrItVREo6uO5i24g2+XfMBHZr0JNuTR9PswgM+xuVIoTDvOL5c/jp9217Ajxs/p0fLAXy3+n2mr5lIMFyjKH1icqc66HtROyU9mg4lYYsEgJ6WRtNHH0VPj8/AcBF9Tfdse63eLS9GsXDOsaO4uu8jfLNywiFf8Lu36MdF3W8GTLI9+RTtXkXzRh1plt2eVaU/xjdwglN9cO5gErpIQOTsRP7v7lQdQ9SHruM5YyDzv5Rtr9G2Yttcftz0JQBOhxtN09CO8HKwoOhrerQaSMioRdd0NJAriSjKb51Jp75Ncbqtc6sJkqRINBo2jJROHVVHEUcp94ZIt9eStbLtNdo65PdgR1URHy54lkmLxtGvwzCcjkPPKlhd+iPtcrvjcrjpkH8CCzZ/xU/F39CxyYlxTJ24NF3j7Ou64FTUxO9wNNM04zdRWxHTMKhZtYoNvxkKif/HTRiF02cz+8ttrJhdojqKEDF1/MBW9L24MK6zq+vKemUrBjRdx926NdmXXKI6iqijtJNOQs/KZPWPsu1VJLb0Rin0vbidJQsEJEmRANDT0ykY8wB6drbqKKIO8v5wP8tmbJFtryLhDbiqM7rTWovV+0uaIgGgud0U3H+f6hjiCBz5eaR06CDdXkXCa9M9l+YdG+FQ2OX1SKybLAb0lBSyzjsPT8+eqqOIwyh4QLa9isTndOsMvPpYXCnW2s30S0lVJCAynKjFP/6OliZnJyxJtr2KJHHSJe0tXyAgCYsEgCMri6Zj/6g6hjiI3Buux18dkm2vIqEVFGbR9bTmljsTcTBJWST01FSyzj+f9H79VEcRv5A1Urq9isTmSnFw/k3dbVEgIEmLBOy57fTXZ2QmtoWknXwyeqZsexWJbcBVnUnxWHO768EkbZEA0Dwemv35z6pjiD3y/nCfbHsVCa3DiU1oe1yeba4iIMmLhO52k35SX7IuuEB1lKTnyM8jpb1sexWJKyMnhQEjO9tisXp/SV0kYE+n2McexdmkieooSa1gzB/Zskq2vYrEpGlw3ujuOCzYm+lI7Jc4BjS3mxbP/wN0+etQQtfx9D+T+VNlwVokphPObUPjZumWPjR3KPZLHAO6y0Vqp07k33WX6ihJKfe3N+CrDlGytkJ1FCGiLr91JicOamu720x7SZHYQ09Lo/FVI8jo3191lKSTddUoFsi2V5GAnG6dQTd3t2QL8Lqyb/IY0D0emj/7V1wtW6qOkjTSTj4ZPUO2vYrEdO4N3UhNd6Fp1m3gdyRSJH5B93ho9dqraCkpqqMkBen2KhJVrwva0uKYHFttdz0YKRK/oDkcuJo1o9kTT6iOkvAi217bs/h/su1VJJY23XI54dw2tl2H2J8UiYPQU1PJHHgm2cOGqY6S0ArGjKVYtr2KBNOoII1zbuiKy+ZXEHtJkTgEPS2NpmMeILVrF9VREpOu4+k/gAWy7VUkEHeqg4t/18P2t5j2J0XiMLTUVFq9+iqOvDzVURJO7m9/K9teRWLR4PybuuPJdKHr9l2o/iUpEoehaRp6Vhat//0vtNRU1XESStZV1zD/C7mKEInj5EvaU1CYjdOVOFcRIEXiiHSXC3erVrT4x98jZ+tFg6Wdcgp6RiZrZNurSBDtT8in+xktE2Kh+pekSNSBnppKeu/eNLnvD6qjJIT8e+9j6fQthEOy7VXYX0FhFgOv6ZKQBQKkSNSZnpZGzmWX0+jKK1RHsTVnkya4ZdurSBCNm6Vz8R09ErZAgBSJo6KneSi4914yzjxTdRTbajJmDMWrduMtl22vwt4yclK45O6eCV0gQIrEUdM9Hlo8+1c8PXqojmI/uo7n9AHSp0nYXmq6i9/ceyIpHidaAu1kOhgpEvWwt3WHu1071VFsJffGG/FVBSlZJ9tehX25UhwMubsnaVludBu2/j5aif8njBE9PZ2270zA1aaN6ii2kXXVNcyXqwhhY7pD46Lbjycr34PDmRwvn8nxp4wBTdfRs7Jo+9670jW2DtJPPRU9PYM1P25XHUWIetE0OO/GbuS1zky4sxCHI0WiATRdx7GnUDibN1cdx9LyZNursLn+I46hZefGCdOTqa6kSDSQ5nDgaNQoUihkTvZBOZs0wd2unWx7FbZ1+hWd6NS7acLvZDoYKRJRoDmdOBo3pu0H8d87xwAAC4RJREFU70ufp4No8scxFK+Uba/ChjQ48+rOdD6pWVIWCJAiETW604kjN5e2772HIydHdRzr0HU8/QYwX7q9CpvRNDj7ui50OLEgaQsESJGIKt3lwlXQhLbvvYujcWPVcSwhd/RofJVBtsm2V2Ejuq5x3o3dKDwuP6kLBEiRiDrN5cLZrBmFH3+Mq4UsZmeNuFquIoSt6A6NQbceR6suuUlfIECKREzobjfO/DzafvQRKZ06qo6jTPppsu1V2IvDqXPRHT1o3rGRFIg9pEjEiOZw4MjOps077+Dp2VN1HCXy7pFtr8I+nG6dwXf1oGlhVtJtcz0cKRIxpGkajowMWr/+TzL691cdJ65+3vZapDqKEEeUkuZkyN0nkN8qM6FGj0aDFIk40NPSaPH3v5E1+GLVUeKmyR//SNHKMrzltaqjCHFY2U08DH+wD7nNM6RAHIQUiTjRPR6aPfIIja+9VnWU2NN1PP3OYMHUzaqTCHFYzTpkc9kDvUnLduNwycvhwcjfShzpHg/5d9xOwSMPgyNxf2LJHT0ar2x7FRZ3TN+mXHRHD9weJ7ouL4WHopmmaaoOkWwMv5/AypUU3XQzRkXivZAWzvyemVO2smrONtVRhDioky5pz3EDEnMmdbRJ+VRA93hI7dqVdpMn4W7fXnWcqNq77XXtPNn2KqzH4dQ5f3Q3KRBHQa4kFDINAzMQYMvdd1P9v29Vx4mKNpM+ZcUGJ3P+s151FCEOkJrh4uLf9aBRQZpscT0KciWhkKbrkZ1Pzz1H7s03qY7TYM6CJrgLC1nyrXR7FdaS1yqD4Q/2oXGzdCkQR0muJCzC8Pnwzp7NlnvuxQwEVMepl+bPP09ZQQ8+f2mJ6ihC7NOtfwtOGdoBp0tH0xJ7HnUsyJWERehpaaSfdhqFn3yMu21b1XGOnq7jOa0/C6RPk7AIV6qD82/qzim/6YDL7ZACUU9SJCxET03F3bo1hR9/RPYll6iOc1Ryb7opsu11faXqKEKQ2yKDKx/uS+uujWWBuoHkdpNFGT4f1TNmUjJmDIbXqzrOERXOnMPMT7ewaq5sexVqde3XnFOHdZTbS1EiVxIWpaelkXFGf9p98Tmp3bqpjnNY6f1OQ09PZ838UtVRRBJzpTg4f3Q3Tr20o9xeiiK5krABw+9n5/+9yK7XXwcL/nO1nvQpK2Xbq1Aot0UGF9x6HJ4Ml/RfijK5krAB3eMh75abaTP+LRy5uarjHMBZUECKbHsVimi6Rq8L2jL0vhPJaJQiBSIGpEjYhJ6Whuf442k/9UuyLrpIdZx9mowdS9EK6fYq4i+naRqXj+3NCee0idxe0uX2UizI7SYbMnw+/EuWsvW++whtU7hQrOu0n7+Qyc8vpnSD7GoS8aFp0OOc1vS+oBCHU0eX4hBTciVhQ3paGp4TT6D955+RM+LKyHeNAnk334y3PCgFQsRN4+bpXP5gH3oPKsTldkiBiAO5krC5sM9H7caNbL37bmo3bIzrcxfOmsPMybLtVcSe7tToc2Ehx5/ZCodTl1tLcSRXEjbnSEsj9ZhjKPz4E/JuuQWczrg8b3q/fmhpabLtVcRcs/bZXPXoSRw3oBVOWXuIO7mSSCCGz0do505Kxo7F98OPMX2uNpM/ZcU6J3MmybZXERsZOSmcPrwTLTvLqWmVpEgkIMPnwzd/Adsee4xgUVHUH99ZUEDhV18z/sHv8VXIriYRXU63Tq9BbTn+zFZoDg2HQ254qCRFIkEZoRCEQpS//wE7xo3DqK6O2mM3f2EcZXnH8fnLS6P2mEKgQafeBfS7rBMOty4tvS1CikSCMwIBzGCQ7c8+S/nEDyEcbtgDOhy0n7eQyc8vkl1NImoKCrMYcFVnsnJTcaXGZ11N1I0UiSRh+HyEdu2i5MGH8M2ZU+/HybvtNpxDR/HOo3OjmE4kq/RGKfS7rCOtu+VKQz6LkiKRZAyfj8DKlWx/5hn8C3866s8vnDWHGZO3sFq2vYoGSMty02tQW449pRmaruFwyrqDVcl1XZLR09Lw9OxJ69dfJ7BqdaRYLFhQp89N798fLS2NtbLtVdRTeiM3vQYV0vmkpqCB0yXrDlYnVxJJzDRNTL+fwOo1kWIxf/5hP77N5CmsWOeQba/iqGXkpNDrgrYc06cpaBpOV3yuHJ566imWLVvGjh07CAQCtGrViqysLFauXMnEiRNp3LgxXq+Xq666iieffJLOnTvHJZedSJEQ+4pFzdq1lD79DP558371Mc6mTSmc9l/Z9iqOSkZOCn0uakfHXk2U3lb6+OOPWb9+Pffccw8AEyZM4Ntvv+XVV1/lzjvvpF+/flx66aVKslmd3G4SaJqGlpZGavfutH71FWrWrWPHuBfwzpixb35FwdixbF6+SwqEqJOsvFT6XFhI+xOboGnWW3MYMWIEs2fPZvTo0eTm5kqBOAwpEmKfvcXC0707Lf72HEZlJTtfeZWKKVNIPfV0Fjy/SHVEYWUatO7SmBPObUNB2yzLL0iPGDGCa6+9lnfffVd1FEuTIiEOypGejiM9nSb33kPBmAcIhjV8lXIVIX4tJc3Jsac0p8fZrXClOHDb4JxDZWUlf/rTn3j00UcZO3YsEydOJD09XXUsS7L+v6ZQyrHnG8cRMrjy4b5s21DBwqmb2byiDGQ1K6nltsig5zmtaN+zCaaJrforPfDAA4wYMYLhw4dTUlLCo48+ytNPP606liVJkRB14txz26DlMY1p0iaLYE2Yxf/f3r3ttHFFYRz/z54Zn8EO4BgIkISmSajai6Sq1PICveUanoFX4hl4gkpV1ZtKLVKF2iYNJG04OAFsDj7gmdkzvXAa5caqGgzGyfeT5tZeGo38afZee/m7HZ7+VKVR7wy4OrkqxnWYf1Tmy29vU6zkcF0HM2SzldbW1jDGsLy8DMDq6iorKyusr6+ztLQ04OquH3U3yXuLAgsO1PaabH6/y9bGAUE7GnRZ0meOA1OflvhscYr5R2WShKFYUpL+UEhIXwTnEcY17D2ts/nDLn9tHhFHerSGWXluhIeLU9z/qoIxDl7aYMxwvTXIxSkkpO+CdoRjHLY3XvPbj/vsPTvW/sWQKFVyPPh6koXFKfy0i+eboVtOkv5SSMilieOYKIhJ4oQXm0ds/3LAyz9qhOcXnEQr/ePAxEyBO59P8OCbSfKlNOaat67K1VJIyJVIkoTg3OL5hqOdBs9+fsWLzSPq+61Bl/bRyeR9ZhfGmH9cZnZhDMcB45m3zQki71JIyEBEoSWJu5vfz389ZHvjgN0ndaIwHnRpHxzHONy8PcKdLyb45HGZ0fEsNopJZbX5LP9NISEDF8cJYaf7lnH8usXukzp7fx6zv3WiMSDvwfUN5bkRKndHmV0YY/peiSROcFNGfwUq/5tCQq6df0PD9RzCjuXV81Ne/l6junXC4U6D2OqRfVfxZpbJu0Wm75eYvldiZDxDFMa4nqNR3HJhCgkZClFgiW2C6xvq1RbVrWMOdxrUqy3q1Sbts3DQJV464zmUyjlKlRzluQIzD8cYv1UAuns+Orsgl0EhIUMr7FhiG+P6LkmccHrY5mi3wcHfZ9SrLWr7Tc5q50PXfpsvpShV8pQqOcZv5ZmYKVAs58jkvW63GAl+ylVrqlwJhYR8cKIoxgYW4xqM69BpRbQbAc3jgLNam7PDc5onAa2TDq3TgOZJh3YjvPQw8dMumYJPtuCTHU2RL6bJl1IUJ7IUxjIUbmQo3EiTxAk2ijGewU9puUgGSyEhH53YJm+6qxJwnO78Ic8QdSzWxtgoxoYJNrREYfesRxRYwsASdSxBp3vOw/MNnu/iprrto65vcL03l2/efm4q4+KnPZKk++Pf/V4wxuClDI7jDPiOiPSmkBARkZ60qCkiIj0pJEREpCeFhIiI9KSQEBGRnhQSIiLSk0JCRER6UkiIiEhPCgkREenpH+z77mfEKmDcAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#a visualization of each race by state\n", + "for race in races:\n", + " values= top5_race_pc[race]\n", + " labels= top5_race_pc.index\n", + " plt.axis('equal')\n", + " plt.title(race)\n", + " plt.pie(values, labels=labels, radius=2, autopct='%0.0f%%')\n", + " plt.show();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Blacks have a generally higher percentage of killings in each state compared to other races.**\n", + "\n", + "**TX has a large share of killings of Native Americans and Blacks.**\n", + "\n", + "**CA has a large share of killings of Whites, Hisapnics and Asians.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 3. Profile of a person killed by police" + ] + }, + { + "cell_type": "code", + "execution_count": 263, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idnamedatemanner_of_deatharmedagegenderracecitystatesigns_of_mental_illnessthreat_levelfleebody_camera
03Tim Elliot2015-02-01shotgun53.0MASheltonWATruehighNot fleeingFalse
14Lewis Lee Lembke2015-02-01shotgun47.0MWAlohaORFalsehighNot fleeingFalse
25John Paul Quintero2015-03-01shot and Taseredunarmed23.0MHWichitaKSFalsemediumNot fleeingFalse
38Matthew Hoffman2015-04-01shottoy weapon32.0MWSan FranciscoCATruehighNot fleeingFalse
49Michael Rodriguez2015-04-01shotnail gun39.0MHEvansCOFalsehighNot fleeingFalse
\n", + "
" + ], + "text/plain": [ + " id name date manner_of_death armed age \\\n", + "0 3 Tim Elliot 2015-02-01 shot gun 53.0 \n", + "1 4 Lewis Lee Lembke 2015-02-01 shot gun 47.0 \n", + "2 5 John Paul Quintero 2015-03-01 shot and Tasered unarmed 23.0 \n", + "3 8 Matthew Hoffman 2015-04-01 shot toy weapon 32.0 \n", + "4 9 Michael Rodriguez 2015-04-01 shot nail gun 39.0 \n", + "\n", + " gender race city state signs_of_mental_illness threat_level \\\n", + "0 M A Shelton WA True high \n", + "1 M W Aloha OR False high \n", + "2 M H Wichita KS False medium \n", + "3 M W San Francisco CA True high \n", + "4 M H Evans CO False high \n", + "\n", + " flee body_camera \n", + "0 Not fleeing False \n", + "1 Not fleeing False \n", + "2 Not fleeing False \n", + "3 Not fleeing False \n", + "4 Not fleeing False " + ] + }, + "execution_count": 263, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 264, + "metadata": {}, + "outputs": [], + "source": [ + "race_avg_age = killings.groupby('race')['age'].mean()" + ] + }, + { + "cell_type": "code", + "execution_count": 265, + "metadata": {}, + "outputs": [], + "source": [ + "race_avg_age.name = 'average_age'" + ] + }, + { + "cell_type": "code", + "execution_count": 266, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "race\n", + "A 36.538462\n", + "B 31.669903\n", + "H 33.018913\n", + "N 30.451613\n", + "O 33.071429\n", + "W 39.942693\n", + "Name: average_age, dtype: float64\n" + ] + } + ], + "source": [ + "print(race_avg_age)" + ] + }, + { + "cell_type": "code", + "execution_count": 267, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "34.11550197750503" + ] + }, + "execution_count": 267, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "race_avg_age.mean()" + ] + }, + { + "cell_type": "code", + "execution_count": 268, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "M 2428\n", + "F 107\n", + "Name: gender, dtype: int64" + ] + }, + "execution_count": 268, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "killings.gender.value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 269, + "metadata": {}, + "outputs": [], + "source": [ + "males = len(killings[killings['gender'] == 'M'])\n", + "females = len(killings[killings['gender'] == 'F'])\n", + "total = killings.shape[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 270, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(2428, 107, 2535)" + ] + }, + "execution_count": 270, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "males, females, total" + ] + }, + { + "cell_type": "code", + "execution_count": 271, + "metadata": {}, + "outputs": [], + "source": [ + "males_pc = round((males / total) * 100)\n", + "females_pc = round((females / total) * 100)" + ] + }, + { + "cell_type": "code", + "execution_count": 272, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(96, 4)" + ] + }, + "execution_count": 272, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "males_pc, females_pc" + ] + }, + { + "cell_type": "code", + "execution_count": 273, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.axis('equal')\n", + "plt.title('Gender Percentage of Killings')\n", + "plt.pie([males_pc, females_pc], labels=['male', 'female'], radius=2, autopct='%0.0f%%')\n", + "plt.show();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**The average profile of a victim killed by police is a Black male, aged 34 years and most likely living in TX.**\n", + "\n", + "**The average profile of a victim killed by police is a Hisapnic male, aged 34 years and most likely living in CA.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 4. Justification of killings" + ] + }, + { + "cell_type": "code", + "execution_count": 274, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.swarmplot(data=killings,\n", + " x='age',\n", + " y='flee',\n", + " hue='manner_of_death')\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Victims of police shootings averagely between the age of 20 and 45 were either not fleeing, or fleeing by car and foot.**\n", + "\n", + "**Victims aged 50 and beyond were mostly shot when not fleeing.**" + ] + }, + { + "cell_type": "code", + "execution_count": 275, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.countplot(data=killings,\n", + " y=\"manner_of_death\",\n", + " hue='threat_level')\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Majority of victims who were shot were a high threat level.**\n", + "\n", + "**Victims who were shot and tasered were both high and medium threat levels.**\n", + "\n", + "*It can be seen that majority of the killings were justified as victims who were high and medium threat levels.*" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} From 74f95d1b6a275fd81cc6806f9f010a7bc050008e Mon Sep 17 00:00:00 2001 From: Alumasa Date: Mon, 21 Dec 2020 21:51:29 +0300 Subject: [PATCH 2/2] Data visualization and analysis --- ... Police Killings Analysis-checkpoint.ipynb | 28 ------------------- .../US Police Killings Analysis.ipynb | 28 ------------------- 2 files changed, 56 deletions(-) diff --git a/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb b/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb index aa1d579..a67c033 100644 --- a/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb +++ b/Level 1/Intermediate/.ipynb_checkpoints/US Police Killings Analysis-checkpoint.ipynb @@ -3152,27 +3152,6 @@ "data.info()" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, { "cell_type": "markdown", "metadata": {}, @@ -3802,13 +3781,6 @@ "**AZ high poverty and low education**" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, { "cell_type": "markdown", "metadata": {}, diff --git a/Level 1/Intermediate/US Police Killings Analysis.ipynb b/Level 1/Intermediate/US Police Killings Analysis.ipynb index aa1d579..a67c033 100644 --- a/Level 1/Intermediate/US Police Killings Analysis.ipynb +++ b/Level 1/Intermediate/US Police Killings Analysis.ipynb @@ -3152,27 +3152,6 @@ "data.info()" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, { "cell_type": "markdown", "metadata": {}, @@ -3802,13 +3781,6 @@ "**AZ high poverty and low education**" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, { "cell_type": "markdown", "metadata": {},