Skip to content

Instantly share code, notes, and snippets.

@parksunwoo

parksunwoo/basic_encoding_code.ipynb

Created August 8, 2018 02:05
Show Gist options
  • Save parksunwoo/9f7c9d7e8c950ece8a6007ab096f2136 to your computer and use it in GitHub Desktop.
Save parksunwoo/9f7c9d7e8c950ece8a6007ab096f2136 to your computer and use it in GitHub Desktop.
Download ZIP
basic encoding code
Display the source blob
Display the rendered blob
Raw
basic_encoding_code.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"#-*- coding: euc-kr -*-\n",
"\n",
"# 공통\n",
"import numpy as np\n",
"import os\n",
"\n",
"\n",
"\n",
"# 데이터셋 저장 경로\n",
"ACCIDENT_PATH = \"../9th_week/교통사망사고정보\"\n",
"SUPPLEMENT_PATH = \"../9th_week/보조데이터/01.서울시 차량 통행 속도\" "
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"\n",
"def load_accident_data(accident_path=ACCIDENT_PATH):\n",
" csv_path = os.path.join(accident_path, \"Kor_Train_교통사망사고정보(12.1~17.6).csv\")\n",
" return pd.read_csv(csv_path, encoding='euc-kr') "
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"train = load_accident_data()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>발생년</th>\n",
" <th>발생년월일시</th>\n",
" <th>발생분</th>\n",
" <th>주야</th>\n",
" <th>요일</th>\n",
" <th>사망자수</th>\n",
" <th>사상자수</th>\n",
" <th>중상자수</th>\n",
" <th>경상자수</th>\n",
" <th>부상신고자수</th>\n",
" <th>...</th>\n",
" <th>도로형태_대분류</th>\n",
" <th>도로형태</th>\n",
" <th>당사자종별_1당_대분류</th>\n",
" <th>당사자종별_1당</th>\n",
" <th>당사자종별_2당_대분류</th>\n",
" <th>당사자종별_2당</th>\n",
" <th>발생위치X_UTMK</th>\n",
" <th>발생위치Y_UTMK</th>\n",
" <th>경도</th>\n",
" <th>위도</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>2017</td>\n",
" <td>2017063023</td>\n",
" <td>14</td>\n",
" <td>야간</td>\n",
" <td>금</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</td>\n",
" <td>단일로</td>\n",
" <td>기타단일로</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>보행자</td>\n",
" <td>보행자</td>\n",
" <td>967633</td>\n",
" <td>1937456</td>\n",
" <td>127.134130</td>\n",
" <td>37.435704</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2017</td>\n",
" <td>2017063021</td>\n",
" <td>45</td>\n",
" <td>야간</td>\n",
" <td>금</td>\n",
" <td>1</td>\n",
" <td>4</td>\n",
" <td>2</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>...</td>\n",
" <td>단일로</td>\n",
" <td>기타단일로</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>966456</td>\n",
" <td>1696405</td>\n",
" <td>127.131220</td>\n",
" <td>35.262534</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2017</td>\n",
" <td>2017063021</td>\n",
" <td>50</td>\n",
" <td>야간</td>\n",
" <td>금</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</td>\n",
" <td>단일로</td>\n",
" <td>기타단일로</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>보행자</td>\n",
" <td>보행자</td>\n",
" <td>900057</td>\n",
" <td>1865697</td>\n",
" <td>126.379922</td>\n",
" <td>36.784142</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2017</td>\n",
" <td>2017063019</td>\n",
" <td>10</td>\n",
" <td>야간</td>\n",
" <td>금</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</td>\n",
" <td>교차로</td>\n",
" <td>교차로부근</td>\n",
" <td>승용차</td>\n",
" <td>승용차</td>\n",
" <td>이륜차</td>\n",
" <td>이륜차</td>\n",
" <td>1095230</td>\n",
" <td>1764891</td>\n",
" <td>128.554945</td>\n",
" <td>35.875973</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2017</td>\n",
" <td>2017063017</td>\n",
" <td>2</td>\n",
" <td>주간</td>\n",
" <td>금</td>\n",
" <td>1</td>\n",
" <td>2</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</td>\n",
" <td>단일로</td>\n",
" <td>교량위</td>\n",
" <td>이륜차</td>\n",
" <td>이륜차</td>\n",
" <td>화물차</td>\n",
" <td>화물차</td>\n",
" <td>947576</td>\n",
" <td>1946355</td>\n",
" <td>126.906786</td>\n",
" <td>37.514993</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 27 columns</p>\n",
"</div>"
],
"text/plain": [
" 발생년 발생년월일시 발생분 주야 요일 사망자수 사상자수 중상자수 경상자수 부상신고자수 ... \\\n",
"0 2017 2017063023 14 야간 금 1 1 0 0 0 ... \n",
"1 2017 2017063021 45 야간 금 1 4 2 1 0 ... \n",
"2 2017 2017063021 50 야간 금 1 1 0 0 0 ... \n",
"3 2017 2017063019 10 야간 금 1 1 0 0 0 ... \n",
"4 2017 2017063017 2 주간 금 1 2 1 0 0 ... \n",
"\n",
" 도로형태_대분류 도로형태 당사자종별_1당_대분류 당사자종별_1당 당사자종별_2당_대분류 당사자종별_2당 발생위치X_UTMK \\\n",
"0 단일로 기타단일로 승용차 승용차 보행자 보행자 967633 \n",
"1 단일로 기타단일로 승용차 승용차 승용차 승용차 966456 \n",
"2 단일로 기타단일로 승용차 승용차 보행자 보행자 900057 \n",
"3 교차로 교차로부근 승용차 승용차 이륜차 이륜차 1095230 \n",
"4 단일로 교량위 이륜차 이륜차 화물차 화물차 947576 \n",
"\n",
" 발생위치Y_UTMK 경도 위도 \n",
"0 1937456 127.134130 37.435704 \n",
"1 1696405 127.131220 35.262534 \n",
"2 1865697 126.379922 36.784142 \n",
"3 1764891 128.554945 35.875973 \n",
"4 1946355 126.906786 37.514993 \n",
"\n",
"[5 rows x 27 columns]"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train.head()"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"def load_supplement_data(accident_path=SUPPLEMENT_PATH):\n",
" csv_path = os.path.join(accident_path, \"2014년 1월 통행속도.csv\")\n",
" return pd.read_csv(csv_path, encoding='euc-kr') "
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>일자</th>\n",
" <th>요일</th>\n",
" <th>도로명</th>\n",
" <th>링크아이디</th>\n",
" <th>시점명</th>\n",
" <th>종점명</th>\n",
" <th>거리</th>\n",
" <th>방향</th>\n",
" <th>01시</th>\n",
" <th>02시</th>\n",
" <th>...</th>\n",
" <th>15시</th>\n",
" <th>16시</th>\n",
" <th>17시</th>\n",
" <th>18시</th>\n",
" <th>19시</th>\n",
" <th>20시</th>\n",
" <th>21시</th>\n",
" <th>22시</th>\n",
" <th>23시</th>\n",
" <th>24시</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>20140101</td>\n",
" <td>수</td>\n",
" <td>4.19로</td>\n",
" <td>1080012200</td>\n",
" <td>통일연수원</td>\n",
" <td>4.19탑</td>\n",
" <td>1179</td>\n",
" <td>상행</td>\n",
" <td>42.77</td>\n",
" <td>45.01</td>\n",
" <td>...</td>\n",
" <td>37.88</td>\n",
" <td>38.75</td>\n",
" <td>39.00</td>\n",
" <td>48.25</td>\n",
" <td>54.50</td>\n",
" <td>64.00</td>\n",
" <td>68.00</td>\n",
" <td>67.00</td>\n",
" <td>58.17</td>\n",
" <td>44.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>20140101</td>\n",
" <td>수</td>\n",
" <td>4.19로</td>\n",
" <td>1080012800</td>\n",
" <td>4.19탑</td>\n",
" <td>국립4.19묘지입구</td>\n",
" <td>352</td>\n",
" <td>상행</td>\n",
" <td>19.36</td>\n",
" <td>21.18</td>\n",
" <td>...</td>\n",
" <td>13.01</td>\n",
" <td>12.61</td>\n",
" <td>10.95</td>\n",
" <td>13.91</td>\n",
" <td>14.27</td>\n",
" <td>18.44</td>\n",
" <td>17.01</td>\n",
" <td>15.87</td>\n",
" <td>18.08</td>\n",
" <td>15.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>20140101</td>\n",
" <td>수</td>\n",
" <td>4.19로</td>\n",
" <td>1080012700</td>\n",
" <td>국립4.19묘지입구</td>\n",
" <td>4.19탑</td>\n",
" <td>354</td>\n",
" <td>하행</td>\n",
" <td>29.08</td>\n",
" <td>31.83</td>\n",
" <td>...</td>\n",
" <td>22.51</td>\n",
" <td>21.07</td>\n",
" <td>18.03</td>\n",
" <td>21.44</td>\n",
" <td>20.63</td>\n",
" <td>23.13</td>\n",
" <td>20.37</td>\n",
" <td>30.03</td>\n",
" <td>23.08</td>\n",
" <td>26.78</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>20140101</td>\n",
" <td>수</td>\n",
" <td>4.19로</td>\n",
" <td>1080012100</td>\n",
" <td>4.19탑</td>\n",
" <td>통일연수원</td>\n",
" <td>1190</td>\n",
" <td>하행</td>\n",
" <td>44.33</td>\n",
" <td>37.25</td>\n",
" <td>...</td>\n",
" <td>29.91</td>\n",
" <td>23.25</td>\n",
" <td>23.67</td>\n",
" <td>23.00</td>\n",
" <td>24.05</td>\n",
" <td>27.25</td>\n",
" <td>35.00</td>\n",
" <td>41.50</td>\n",
" <td>38.42</td>\n",
" <td>25.75</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>20140101</td>\n",
" <td>수</td>\n",
" <td>가락로</td>\n",
" <td>1230024700</td>\n",
" <td>방이초교</td>\n",
" <td>백제고분군</td>\n",
" <td>844</td>\n",
" <td>상행</td>\n",
" <td>24.22</td>\n",
" <td>23.65</td>\n",
" <td>...</td>\n",
" <td>18.63</td>\n",
" <td>20.87</td>\n",
" <td>22.08</td>\n",
" <td>23.78</td>\n",
" <td>18.95</td>\n",
" <td>18.70</td>\n",
" <td>21.07</td>\n",
" <td>18.92</td>\n",
" <td>18.73</td>\n",
" <td>22.75</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 32 columns</p>\n",
"</div>"
],
"text/plain": [
" 일자 요일 도로명 링크아이디 시점명 종점명 거리 방향 01시 \\\n",
"0 20140101 수 4.19로 1080012200 통일연수원 4.19탑 1179 상행 42.77 \n",
"1 20140101 수 4.19로 1080012800 4.19탑 국립4.19묘지입구 352 상행 19.36 \n",
"2 20140101 수 4.19로 1080012700 국립4.19묘지입구 4.19탑 354 하행 29.08 \n",
"3 20140101 수 4.19로 1080012100 4.19탑 통일연수원 1190 하행 44.33 \n",
"4 20140101 수 가락로 1230024700 방이초교 백제고분군 844 상행 24.22 \n",
"\n",
" 02시 ... 15시 16시 17시 18시 19시 20시 21시 22시 \\\n",
"0 45.01 ... 37.88 38.75 39.00 48.25 54.50 64.00 68.00 67.00 \n",
"1 21.18 ... 13.01 12.61 10.95 13.91 14.27 18.44 17.01 15.87 \n",
"2 31.83 ... 22.51 21.07 18.03 21.44 20.63 23.13 20.37 30.03 \n",
"3 37.25 ... 29.91 23.25 23.67 23.00 24.05 27.25 35.00 41.50 \n",
"4 23.65 ... 18.63 20.87 22.08 23.78 18.95 18.70 21.07 18.92 \n",
"\n",
" 23시 24시 \n",
"0 58.17 44.00 \n",
"1 18.08 15.52 \n",
"2 23.08 26.78 \n",
"3 38.42 25.75 \n",
"4 18.73 22.75 \n",
"\n",
"[5 rows x 32 columns]"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train2 = load_supplement_data()\n",
"train2.head()"
]
}
],
"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": "3.6.4"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment