1 Getting & Knowing Data

In [1]:

import pandas as pd

In [2]:

url = 'https://raw.githubusercontent.com/Datamanim/pandas/main/lol.csv'

In [3]:

df=pd.read_csv(url,sep='\t')

In [4]:

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 51490 entries, 0 to 51489
Data columns (total 61 columns):
 #   Column              Non-Null Count  Dtype
---  ------              --------------  -----
 0   gameId              51490 non-null  int64
 1   creationTime        51490 non-null  int64
 2   gameDuration        51490 non-null  int64
 3   seasonId            51490 non-null  int64
 4   winner              51490 non-null  int64
 5   firstBlood          51490 non-null  int64
 6   firstTower          51490 non-null  int64
 7   firstInhibitor      51490 non-null  int64
 8   firstBaron          51490 non-null  int64
 9   firstDragon         51490 non-null  int64
 10  firstRiftHerald     51490 non-null  int64
 11  t1_champ1id         51490 non-null  int64
 12  t1_champ1_sum1      51490 non-null  int64
 13  t1_champ1_sum2      51490 non-null  int64
 14  t1_champ2id         51490 non-null  int64
 15  t1_champ2_sum1      51490 non-null  int64
 16  t1_champ2_sum2      51490 non-null  int64
 17  t1_champ3id         51490 non-null  int64
 18  t1_champ3_sum1      51490 non-null  int64
 19  t1_champ3_sum2      51490 non-null  int64
 20  t1_champ4id         51490 non-null  int64
 21  t1_champ4_sum1      51490 non-null  int64
 22  t1_champ4_sum2      51490 non-null  int64
 23  t1_champ5id         51490 non-null  int64
 24  t1_champ5_sum1      51490 non-null  int64
 25  t1_champ5_sum2      51490 non-null  int64
 26  t1_towerKills       51490 non-null  int64
 27  t1_inhibitorKills   51490 non-null  int64
 28  t1_baronKills       51490 non-null  int64
 29  t1_dragonKills      51490 non-null  int64
 30  t1_riftHeraldKills  51490 non-null  int64
 31  t1_ban1             51490 non-null  int64
 32  t1_ban2             51490 non-null  int64
 33  t1_ban3             51490 non-null  int64
 34  t1_ban4             51490 non-null  int64
 35  t1_ban5             51490 non-null  int64
 36  t2_champ1id         51490 non-null  int64
 37  t2_champ1_sum1      51490 non-null  int64
 38  t2_champ1_sum2      51490 non-null  int64
 39  t2_champ2id         51490 non-null  int64
 40  t2_champ2_sum1      51490 non-null  int64
 41  t2_champ2_sum2      51490 non-null  int64
 42  t2_champ3id         51490 non-null  int64
 43  t2_champ3_sum1      51490 non-null  int64
 44  t2_champ3_sum2      51490 non-null  int64
 45  t2_champ4id         51490 non-null  int64
 46  t2_champ4_sum1      51490 non-null  int64
 47  t2_champ4_sum2      51490 non-null  int64
 48  t2_champ5id         51490 non-null  int64
 49  t2_champ5_sum1      51490 non-null  int64
 50  t2_champ5_sum2      51490 non-null  int64
 51  t2_towerKills       51490 non-null  int64
 52  t2_inhibitorKills   51490 non-null  int64
 53  t2_baronKills       51490 non-null  int64
 54  t2_dragonKills      51490 non-null  int64
 55  t2_riftHeraldKills  51490 non-null  int64
 56  t2_ban1             51490 non-null  int64
 57  t2_ban2             51490 non-null  int64
 58  t2_ban3             51490 non-null  int64
 59  t2_ban4             51490 non-null  int64
 60  t2_ban5             51490 non-null  int64
dtypes: int64(61)
memory usage: 24.0 MB

In [5]:

df.head()

Out[5]:

	gameId	creationTime	gameDuration	seasonId	winner	firstBlood	firstTower	firstInhibitor	firstBaron	firstDragon	...	t2_towerKills	t2_inhibitorKills	t2_baronKills	t2_dragonKills	t2_riftHeraldKills	t2_ban1	t2_ban2	t2_ban3	t2_ban4	t2_ban5
0	3326086514	1504279457970	1949	9	1	2	1	1	1	1	...	5	0	0	1	1	114	67	43	16	51
1	3229566029	1497848803862	1851	9	1	1	1	1	0	1	...	2	0	0	0	0	11	67	238	51	420
2	3327363504	1504360103310	1493	9	1	2	1	1	1	2	...	2	0	0	1	0	157	238	121	57	28
3	3326856598	1504348503996	1758	9	1	1	1	1	1	1	...	0	0	0	0	0	164	18	141	40	51
4	3330080762	1504554410899	2094	9	1	2	1	1	1	1	...	3	0	0	1	0	86	11	201	122	18

5 rows × 61 columns

In [6]:

type(df)

Out[6]:

pandas.core.frame.DataFrame

In [7]:

df.shape()

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-7-0e566b70f572> in <module>
----> 1 df.shape()

TypeError: 'tuple' object is not callable

In [8]:

df.shape

Out[8]:

(51490, 61)

In [10]:

print(df.shape[0])

51490

In [11]:

print(df.shape[1])

61

In [14]:

print(df.columns)

Index(['gameId', 'creationTime', 'gameDuration', 'seasonId', 'winner',
       'firstBlood', 'firstTower', 'firstInhibitor', 'firstBaron',
       'firstDragon', 'firstRiftHerald', 't1_champ1id', 't1_champ1_sum1',
       't1_champ1_sum2', 't1_champ2id', 't1_champ2_sum1', 't1_champ2_sum2',
       't1_champ3id', 't1_champ3_sum1', 't1_champ3_sum2', 't1_champ4id',
       't1_champ4_sum1', 't1_champ4_sum2', 't1_champ5id', 't1_champ5_sum1',
       't1_champ5_sum2', 't1_towerKills', 't1_inhibitorKills', 't1_baronKills',
       't1_dragonKills', 't1_riftHeraldKills', 't1_ban1', 't1_ban2', 't1_ban3',
       't1_ban4', 't1_ban5', 't2_champ1id', 't2_champ1_sum1', 't2_champ1_sum2',
       't2_champ2id', 't2_champ2_sum1', 't2_champ2_sum2', 't2_champ3id',
       't2_champ3_sum1', 't2_champ3_sum2', 't2_champ4id', 't2_champ4_sum1',
       't2_champ4_sum2', 't2_champ5id', 't2_champ5_sum1', 't2_champ5_sum2',
       't2_towerKills', 't2_inhibitorKills', 't2_baronKills', 't2_dragonKills',
       't2_riftHeraldKills', 't2_ban1', 't2_ban2', 't2_ban3', 't2_ban4',
       't2_ban5'],
      dtype='object')

In [15]:

print(df.columns[5])

firstBlood

6번째 컬럼의 데이터 타입을 확인하라

In [20]:

df.iloc[:,5].dtype

Out[20]:

dtype('int64')

In [25]:

df.index

Out[25]:

RangeIndex(start=0, stop=51490, step=1)

6번째 컬럼의 3번째 값은 무엇인가?

In [28]:

df.iloc[2,5]

Out[28]:

2

In [31]:

url2='https://raw.githubusercontent.com/Datamanim/pandas/main/Jeju.csv'

In [32]:

df2=pd.read_csv(url2, encoding='euc-kr')

In [33]:

df2.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9621 entries, 0 to 9620
Data columns (total 13 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   id        9621 non-null   int64  
 1   일자        9621 non-null   object 
 2   시도명       9621 non-null   object 
 3   읍면동명      9621 non-null   object 
 4   거주인구      9621 non-null   float64
 5   근무인구      9621 non-null   float64
 6   방문인구      9621 non-null   float64
 7   총 유동인구    9621 non-null   float64
 8   평균 속도     9621 non-null   float64
 9   평균 소요 시간  9621 non-null   float64
 10  평균 기온     9621 non-null   float64
 11  일강수량      9621 non-null   float64
 12  평균 풍속     9621 non-null   float64
dtypes: float64(9), int64(1), object(3)
memory usage: 977.3+ KB

In [35]:

df2.tail(3)

Out[35]:

	id	일자	시도명	읍면동명	거주인구	근무인구	방문인구	총 유동인구	평균 속도	평균 소요 시간	평균 기온	일강수량	평균 풍속
9618	32066	2020-04-30	제주시	도두동	28397.481	3144.895	84052.697	115595.073	41.053	29.421	20.3	0.0	3.0
9619	32067	2020-04-30	서귀포시	안덕면	348037.846	29106.286	251129.660	628273.792	46.595	49.189	17.6	0.0	3.5
9620	32068	2020-04-30	제주시	연동	1010643.372	65673.477	447622.068	1523938.917	40.863	27.765	14.1	0.0	4.8

수치형 변수를 가진 컬럼을 출력하라

In [38]:

df2.select_dtypes(exclude=object).columns

Out[38]:

Index(['id', '거주인구', '근무인구', '방문인구', '총 유동인구', '평균 속도', '평균 소요 시간', '평균 기온',
       '일강수량', '평균 풍속'],
      dtype='object')

In [39]:

df2.select_dtypes(include=object).columns

Out[39]:

Index(['일자', '시도명', '읍면동명'], dtype='object')

각 컬럼의 결측치 숫자를 파악하라

In [41]:

df2.isnull().sum()

Out[41]:

id          0
일자          0
시도명         0
읍면동명        0
거주인구        0
근무인구        0
방문인구        0
총 유동인구      0
평균 속도       0
평균 소요 시간    0
평균 기온       0
일강수량        0
평균 풍속       0
dtype: int64

In [42]:

df2.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9621 entries, 0 to 9620
Data columns (total 13 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   id        9621 non-null   int64  
 1   일자        9621 non-null   object 
 2   시도명       9621 non-null   object 
 3   읍면동명      9621 non-null   object 
 4   거주인구      9621 non-null   float64
 5   근무인구      9621 non-null   float64
 6   방문인구      9621 non-null   float64
 7   총 유동인구    9621 non-null   float64
 8   평균 속도     9621 non-null   float64
 9   평균 소요 시간  9621 non-null   float64
 10  평균 기온     9621 non-null   float64
 11  일강수량      9621 non-null   float64
 12  평균 풍속     9621 non-null   float64
dtypes: float64(9), int64(1), object(3)
memory usage: 977.3+ KB

각 수치형 변수의 분포(사분위, 평균, 표준편차, 최대 , 최소)를 확인하라

In [43]:

df2.describe()

Out[43]:

	id	거주인구	근무인구	방문인구	총 유동인구	평균 속도	평균 소요 시간	평균 기온	일강수량	평균 풍속
count	9621.000000	9.621000e+03	9621.000000	9621.000000	9.621000e+03	9621.000000	9621.000000	9621.000000	9621.000000	9621.000000
mean	27258.000000	3.174315e+05	35471.201510	195889.561802	5.487922e+05	41.109084	37.215873	13.550828	6.972426	2.753171
std	2777.487804	2.982079e+05	40381.214775	140706.090325	4.608802e+05	8.758631	12.993786	7.745515	27.617260	1.498538
min	22448.000000	9.305552e+03	1407.936000	11538.322000	2.225181e+04	24.333000	12.667000	-9.600000	0.000000	0.000000
25%	24853.000000	9.539939e+04	12074.498000	99632.153000	2.216910e+05	34.250000	27.889000	7.600000	0.000000	1.700000
50%	27258.000000	2.221105e+05	21960.928000	152805.335000	3.866935e+05	39.640000	34.500000	13.400000	0.000000	2.400000
75%	29663.000000	4.106671e+05	40192.032000	236325.109000	6.406918e+05	49.105000	46.176000	19.700000	1.500000	3.400000
max	32068.000000	1.364504e+06	263476.965000	723459.209000	2.066484e+06	103.000000	172.200000	30.400000	587.500000	13.333000

In [44]:

df2["거주인구"]

Out[44]:

0         32249.987
1        213500.997
2       1212382.218
3         33991.653
4        155036.925
           ...     
9616     228260.005
9617     459959.064
9618      28397.481
9619     348037.846
9620    1010643.372
Name: 거주인구, Length: 9621, dtype: float64

In [47]:

df2["평균 속도"].quantile(0.75)-df2["평균 속도"].quantile(0.25)

Out[47]:

14.854999999999997

읍면동명 컬럼의 유일값 갯수를 출력하라

In [48]:

df2["읍면동명"].nunique()

Out[48]:

41

In [49]:

df2["읍면동명"].unique()

Out[49]:

array(['도두동', '외도동', '이도2동', '일도1동', '대천동', '서홍동', '한경면', '송산동', '조천읍',
       '일도2동', '영천동', '예래동', '대륜동', '삼도1동', '이호동', '건입동', '중앙동', '삼양동',
       '삼도2동', '이도1동', '남원읍', '대정읍', '정방동', '효돈동', '아라동', '한림읍', '구좌읍',
       '용담1동', '오라동', '화북동', '연동', '표선면', '중문동', '성산읍', '안덕면', '천지동',
       '노형동', '동홍동', '용담2동', '봉개동', '애월읍'], dtype=object)

In [ ]: