파이썬 (0506) 10주차

!pip install numpy
!pip install pandas

Requirement already satisfied: numpy in /home/koinup4/anaconda3/envs/py37/lib/python3.7/site-packages (1.21.6)
Collecting pandas
  Downloading pandas-1.3.5-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.3 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.3/11.3 MB 89.6 MB/s eta 0:00:00a 0:00:01
Requirement already satisfied: pytz>=2017.3 in /home/koinup4/anaconda3/envs/py37/lib/python3.7/site-packages (from pandas) (2022.7.1)
Requirement already satisfied: numpy>=1.17.3 in /home/koinup4/anaconda3/envs/py37/lib/python3.7/site-packages (from pandas) (1.21.6)
Requirement already satisfied: python-dateutil>=2.7.3 in /home/koinup4/anaconda3/envs/py37/lib/python3.7/site-packages (from pandas) (2.8.2)
Requirement already satisfied: six>=1.5 in /home/koinup4/anaconda3/envs/py37/lib/python3.7/site-packages (from python-dateutil>=2.7.3->pandas) (1.16.0)
Installing collected packages: pandas
Successfully installed pandas-1.3.5

import numpy as np
import pandas as pd

부분 데이터 꺼내기: 판다스를 왜 써야할까?

기본 인덱싱

-예제1: 기본인덱싱

a='asdf'
a[2]

'd'

a[-1]

'f'

- 예제2: 슬라이싱

a='asdf'
a[1:3]

'sd'

a[-2:]

'df'

- 예제3: 스트라이딩

a='afsdf'
a[::2]

'asf'

- 예제4: 불가능한것

a='afsd'
a[[1,2]] # 리스트로 전달해서 뽑는것은 불가능 -> 정수인덱스 리스트화시켜서 인덱싱하는것

TypeError: string indices must be integers

a='afsd'
a[[True,True,False,True]] # 리스트로 전달해서 뽑는것은 불가능 -> 정수인덱스 리스트화시켜서 인덱싱하는것

TypeError: string indices must be integers

팬시인덱싱

- 예제1: 인덱스의 리스트(혹은 ndarray)를 전달

a=np.arange(5)
a[0]

0

a[[0,1]]

array([0, 1])

a[[0,1,-2]]

array([0, 1, 3])

- 예제2: bool로 이루어진 리스트 (혹은 ndarray)를 전달

a=np.arange(55,61)
a

array([55, 56, 57, 58, 59, 60])

a[[True,True,False,True,True,False]]

array([55, 56, 58, 59])

a<58

array([ True,  True,  True, False, False, False])

a[a<58]

array([55, 56, 57])

2차원 자료형의 인덱싱

- 예제1

a=np.arange(4*3).reshape(4,3)
a

array([[ 0,  1,  2],
       [ 3,  4,  5],
       [ 6,  7,  8],
       [ 9, 10, 11]])

a[0:2,1]

array([1, 4])

- 예제2 : 차원을 유지하면서 인덱싱을 하고 싶으면?

a[0:2,[1]]

array([[1],
       [4]])

HASH

- 예제1 : (key, value)

d={'att':67, 'rep':45, 'mid':30, 'fin':100}
d

{'att': 67, 'rep': 45, 'mid': 30, 'fin': 100}

d['att'] # key를 넣으면 value가 리턴

67

- 예제2: numpy비교

np.random.seed(43052)
att = np.random.choice(np.arange(10,21)*5,200)
rep = np.random.choice(np.arange(5,21)*5,200)
mid = np.random.choice(np.arange(0,21)*5,200)
fin = np.random.choice(np.arange(0,21)*5,200)
key = ['202212'+str(s) for s in np.random.choice(np.arange(300,501),200,replace=False)]
test_dic = {key[i] : {'att':att[i], 'rep':rep[i], 'mid':mid[i], 'fin':fin[i]} for i in range(200)}
test_ndarray = np.array([key,att,rep,mid,fin],dtype=np.int64).T

test_dic

{'202212377': {'att': 65, 'rep': 45, 'mid': 0, 'fin': 10},
 '202212473': {'att': 95, 'rep': 30, 'mid': 60, 'fin': 10},
 '202212310': {'att': 65, 'rep': 85, 'mid': 15, 'fin': 20},
 '202212460': {'att': 55, 'rep': 35, 'mid': 35, 'fin': 5},
 '202212320': {'att': 80, 'rep': 60, 'mid': 55, 'fin': 70},
 '202212329': {'att': 75, 'rep': 40, 'mid': 75, 'fin': 85},
 '202212408': {'att': 65, 'rep': 70, 'mid': 60, 'fin': 75},
 '202212319': {'att': 60, 'rep': 25, 'mid': 20, 'fin': 35},
 '202212348': {'att': 95, 'rep': 55, 'mid': 65, 'fin': 90},
 '202212306': {'att': 90, 'rep': 25, 'mid': 95, 'fin': 50},
 '202212308': {'att': 55, 'rep': 45, 'mid': 75, 'fin': 30},
 '202212366': {'att': 95, 'rep': 60, 'mid': 25, 'fin': 55},
 '202212367': {'att': 95, 'rep': 35, 'mid': 0, 'fin': 25},
 '202212461': {'att': 50, 'rep': 55, 'mid': 90, 'fin': 45},
 '202212354': {'att': 50, 'rep': 65, 'mid': 50, 'fin': 70},
 '202212361': {'att': 95, 'rep': 100, 'mid': 25, 'fin': 40},
 '202212400': {'att': 50, 'rep': 65, 'mid': 35, 'fin': 85},
 '202212490': {'att': 65, 'rep': 85, 'mid': 10, 'fin': 5},
 '202212404': {'att': 70, 'rep': 65, 'mid': 65, 'fin': 80},
 '202212326': {'att': 90, 'rep': 70, 'mid': 100, 'fin': 30},
 '202212452': {'att': 80, 'rep': 45, 'mid': 80, 'fin': 85},
 '202212362': {'att': 55, 'rep': 45, 'mid': 85, 'fin': 70},
 '202212396': {'att': 65, 'rep': 35, 'mid': 45, 'fin': 20},
 '202212356': {'att': 70, 'rep': 25, 'mid': 50, 'fin': 70},
 '202212305': {'att': 85, 'rep': 55, 'mid': 30, 'fin': 80},
 '202212398': {'att': 90, 'rep': 30, 'mid': 30, 'fin': 0},
 '202212410': {'att': 100, 'rep': 65, 'mid': 50, 'fin': 70},
 '202212385': {'att': 80, 'rep': 70, 'mid': 50, 'fin': 100},
 '202212430': {'att': 80, 'rep': 35, 'mid': 25, 'fin': 65},
 '202212498': {'att': 55, 'rep': 75, 'mid': 20, 'fin': 25},
 '202212423': {'att': 75, 'rep': 75, 'mid': 85, 'fin': 95},
 '202212327': {'att': 80, 'rep': 95, 'mid': 5, 'fin': 5},
 '202212347': {'att': 95, 'rep': 60, 'mid': 65, 'fin': 10},
 '202212483': {'att': 95, 'rep': 60, 'mid': 90, 'fin': 75},
 '202212447': {'att': 100, 'rep': 75, 'mid': 70, 'fin': 25},
 '202212496': {'att': 100, 'rep': 55, 'mid': 35, 'fin': 85},
 '202212358': {'att': 80, 'rep': 60, 'mid': 65, 'fin': 55},
 '202212399': {'att': 70, 'rep': 80, 'mid': 0, 'fin': 10},
 '202212459': {'att': 85, 'rep': 65, 'mid': 60, 'fin': 60},
 '202212313': {'att': 100, 'rep': 95, 'mid': 0, 'fin': 25},
 '202212304': {'att': 95, 'rep': 60, 'mid': 15, 'fin': 45},
 '202212431': {'att': 75, 'rep': 40, 'mid': 30, 'fin': 10},
 '202212325': {'att': 70, 'rep': 80, 'mid': 50, 'fin': 25},
 '202212471': {'att': 50, 'rep': 45, 'mid': 10, 'fin': 10},
 '202212463': {'att': 100, 'rep': 100, 'mid': 100, 'fin': 50},
 '202212441': {'att': 75, 'rep': 50, 'mid': 60, 'fin': 5},
 '202212445': {'att': 85, 'rep': 50, 'mid': 35, 'fin': 100},
 '202212323': {'att': 80, 'rep': 35, 'mid': 75, 'fin': 80},
 '202212442': {'att': 95, 'rep': 45, 'mid': 35, 'fin': 80},
 '202212346': {'att': 65, 'rep': 85, 'mid': 85, 'fin': 15},
 '202212411': {'att': 90, 'rep': 30, 'mid': 25, 'fin': 5},
 '202212468': {'att': 65, 'rep': 65, 'mid': 35, 'fin': 70},
 '202212331': {'att': 80, 'rep': 65, 'mid': 30, 'fin': 90},
 '202212345': {'att': 95, 'rep': 80, 'mid': 45, 'fin': 35},
 '202212339': {'att': 65, 'rep': 75, 'mid': 50, 'fin': 35},
 '202212383': {'att': 90, 'rep': 55, 'mid': 100, 'fin': 30},
 '202212462': {'att': 95, 'rep': 25, 'mid': 95, 'fin': 90},
 '202212344': {'att': 100, 'rep': 50, 'mid': 80, 'fin': 10},
 '202212472': {'att': 50, 'rep': 55, 'mid': 35, 'fin': 60},
 '202212437': {'att': 90, 'rep': 70, 'mid': 35, 'fin': 25},
 '202212336': {'att': 50, 'rep': 55, 'mid': 15, 'fin': 75},
 '202212438': {'att': 80, 'rep': 50, 'mid': 55, 'fin': 90},
 '202212454': {'att': 50, 'rep': 75, 'mid': 65, 'fin': 90},
 '202212384': {'att': 70, 'rep': 40, 'mid': 90, 'fin': 5},
 '202212402': {'att': 65, 'rep': 85, 'mid': 20, 'fin': 90},
 '202212397': {'att': 60, 'rep': 30, 'mid': 0, 'fin': 50},
 '202212318': {'att': 50, 'rep': 65, 'mid': 15, 'fin': 0},
 '202212371': {'att': 60, 'rep': 95, 'mid': 30, 'fin': 70},
 '202212469': {'att': 70, 'rep': 70, 'mid': 5, 'fin': 0},
 '202212379': {'att': 75, 'rep': 45, 'mid': 15, 'fin': 75},
 '202212364': {'att': 50, 'rep': 60, 'mid': 15, 'fin': 50},
 '202212450': {'att': 85, 'rep': 90, 'mid': 90, 'fin': 90},
 '202212337': {'att': 80, 'rep': 25, 'mid': 85, 'fin': 20},
 '202212458': {'att': 55, 'rep': 75, 'mid': 95, 'fin': 90},
 '202212494': {'att': 85, 'rep': 30, 'mid': 45, 'fin': 15},
 '202212478': {'att': 65, 'rep': 30, 'mid': 45, 'fin': 15},
 '202212373': {'att': 85, 'rep': 95, 'mid': 35, 'fin': 25},
 '202212474': {'att': 60, 'rep': 25, 'mid': 10, 'fin': 50},
 '202212455': {'att': 95, 'rep': 45, 'mid': 90, 'fin': 35},
 '202212317': {'att': 85, 'rep': 50, 'mid': 60, 'fin': 45},
 '202212341': {'att': 60, 'rep': 50, 'mid': 100, 'fin': 70},
 '202212386': {'att': 100, 'rep': 75, 'mid': 60, 'fin': 0},
 '202212328': {'att': 100, 'rep': 90, 'mid': 85, 'fin': 75},
 '202212417': {'att': 55, 'rep': 100, 'mid': 100, 'fin': 60},
 '202212370': {'att': 70, 'rep': 60, 'mid': 30, 'fin': 40},
 '202212486': {'att': 70, 'rep': 90, 'mid': 95, 'fin': 40},
 '202212333': {'att': 55, 'rep': 50, 'mid': 0, 'fin': 5},
 '202212360': {'att': 100, 'rep': 100, 'mid': 45, 'fin': 90},
 '202212350': {'att': 85, 'rep': 70, 'mid': 90, 'fin': 80},
 '202212382': {'att': 100, 'rep': 85, 'mid': 65, 'fin': 85},
 '202212392': {'att': 60, 'rep': 65, 'mid': 35, 'fin': 15},
 '202212449': {'att': 65, 'rep': 75, 'mid': 75, 'fin': 85},
 '202212394': {'att': 65, 'rep': 25, 'mid': 40, 'fin': 0},
 '202212444': {'att': 75, 'rep': 75, 'mid': 50, 'fin': 40},
 '202212487': {'att': 50, 'rep': 55, 'mid': 80, 'fin': 55},
 '202212425': {'att': 75, 'rep': 30, 'mid': 20, 'fin': 50},
 '202212312': {'att': 100, 'rep': 50, 'mid': 25, 'fin': 65},
 '202212448': {'att': 90, 'rep': 30, 'mid': 95, 'fin': 35},
 '202212434': {'att': 55, 'rep': 100, 'mid': 80, 'fin': 0},
 '202212451': {'att': 75, 'rep': 60, 'mid': 15, 'fin': 40},
 '202212433': {'att': 60, 'rep': 25, 'mid': 25, 'fin': 50},
 '202212424': {'att': 85, 'rep': 35, 'mid': 10, 'fin': 60},
 '202212351': {'att': 60, 'rep': 100, 'mid': 55, 'fin': 40},
 '202212324': {'att': 70, 'rep': 55, 'mid': 50, 'fin': 75},
 '202212314': {'att': 80, 'rep': 65, 'mid': 95, 'fin': 85},
 '202212446': {'att': 65, 'rep': 35, 'mid': 15, 'fin': 65},
 '202212401': {'att': 85, 'rep': 70, 'mid': 100, 'fin': 0},
 '202212307': {'att': 100, 'rep': 30, 'mid': 60, 'fin': 65},
 '202212300': {'att': 65, 'rep': 70, 'mid': 55, 'fin': 70},
 '202212342': {'att': 85, 'rep': 55, 'mid': 85, 'fin': 90},
 '202212479': {'att': 85, 'rep': 95, 'mid': 80, 'fin': 10},
 '202212443': {'att': 85, 'rep': 70, 'mid': 75, 'fin': 5},
 '202212387': {'att': 100, 'rep': 35, 'mid': 70, 'fin': 0},
 '202212372': {'att': 95, 'rep': 45, 'mid': 55, 'fin': 65},
 '202212376': {'att': 95, 'rep': 85, 'mid': 40, 'fin': 65},
 '202212466': {'att': 55, 'rep': 50, 'mid': 30, 'fin': 85},
 '202212391': {'att': 85, 'rep': 50, 'mid': 5, 'fin': 65},
 '202212368': {'att': 75, 'rep': 90, 'mid': 85, 'fin': 85},
 '202212427': {'att': 95, 'rep': 70, 'mid': 10, 'fin': 5},
 '202212414': {'att': 85, 'rep': 35, 'mid': 80, 'fin': 95},
 '202212426': {'att': 95, 'rep': 50, 'mid': 80, 'fin': 90},
 '202212316': {'att': 100, 'rep': 65, 'mid': 75, 'fin': 40},
 '202212355': {'att': 95, 'rep': 70, 'mid': 70, 'fin': 0},
 '202212477': {'att': 95, 'rep': 70, 'mid': 20, 'fin': 25},
 '202212484': {'att': 100, 'rep': 60, 'mid': 10, 'fin': 5},
 '202212456': {'att': 55, 'rep': 35, 'mid': 25, 'fin': 10},
 '202212500': {'att': 60, 'rep': 90, 'mid': 40, 'fin': 5},
 '202212381': {'att': 85, 'rep': 90, 'mid': 85, 'fin': 75},
 '202212335': {'att': 75, 'rep': 85, 'mid': 25, 'fin': 35},
 '202212475': {'att': 55, 'rep': 30, 'mid': 50, 'fin': 45},
 '202212343': {'att': 70, 'rep': 60, 'mid': 75, 'fin': 75},
 '202212412': {'att': 80, 'rep': 30, 'mid': 95, 'fin': 5},
 '202212428': {'att': 90, 'rep': 85, 'mid': 80, 'fin': 15},
 '202212330': {'att': 90, 'rep': 25, 'mid': 95, 'fin': 5},
 '202212375': {'att': 60, 'rep': 85, 'mid': 50, 'fin': 20},
 '202212413': {'att': 90, 'rep': 50, 'mid': 95, 'fin': 95},
 '202212303': {'att': 75, 'rep': 95, 'mid': 65, 'fin': 40},
 '202212374': {'att': 60, 'rep': 40, 'mid': 35, 'fin': 0},
 '202212409': {'att': 55, 'rep': 100, 'mid': 15, 'fin': 80},
 '202212440': {'att': 70, 'rep': 75, 'mid': 80, 'fin': 0},
 '202212393': {'att': 75, 'rep': 65, 'mid': 25, 'fin': 20},
 '202212492': {'att': 90, 'rep': 75, 'mid': 80, 'fin': 25},
 '202212357': {'att': 50, 'rep': 75, 'mid': 75, 'fin': 20},
 '202212465': {'att': 55, 'rep': 45, 'mid': 35, 'fin': 45},
 '202212415': {'att': 90, 'rep': 70, 'mid': 90, 'fin': 0},
 '202212405': {'att': 75, 'rep': 30, 'mid': 100, 'fin': 60},
 '202212435': {'att': 90, 'rep': 85, 'mid': 0, 'fin': 40},
 '202212380': {'att': 85, 'rep': 70, 'mid': 35, 'fin': 0},
 '202212369': {'att': 100, 'rep': 75, 'mid': 100, 'fin': 85},
 '202212467': {'att': 55, 'rep': 35, 'mid': 20, 'fin': 10},
 '202212429': {'att': 70, 'rep': 75, 'mid': 90, 'fin': 90},
 '202212495': {'att': 90, 'rep': 90, 'mid': 55, 'fin': 55},
 '202212420': {'att': 55, 'rep': 60, 'mid': 40, 'fin': 0},
 '202212302': {'att': 100, 'rep': 90, 'mid': 5, 'fin': 30},
 '202212481': {'att': 50, 'rep': 55, 'mid': 25, 'fin': 80},
 '202212422': {'att': 100, 'rep': 100, 'mid': 90, 'fin': 55},
 '202212388': {'att': 70, 'rep': 45, 'mid': 70, 'fin': 75},
 '202212480': {'att': 85, 'rep': 95, 'mid': 85, 'fin': 90},
 '202212378': {'att': 55, 'rep': 25, 'mid': 95, 'fin': 45},
 '202212457': {'att': 75, 'rep': 30, 'mid': 10, 'fin': 95},
 '202212419': {'att': 65, 'rep': 85, 'mid': 15, 'fin': 60},
 '202212432': {'att': 70, 'rep': 90, 'mid': 70, 'fin': 0},
 '202212395': {'att': 60, 'rep': 85, 'mid': 70, 'fin': 85},
 '202212464': {'att': 100, 'rep': 25, 'mid': 10, 'fin': 20},
 '202212476': {'att': 75, 'rep': 25, 'mid': 80, 'fin': 25},
 '202212332': {'att': 90, 'rep': 95, 'mid': 40, 'fin': 80},
 '202212301': {'att': 95, 'rep': 90, 'mid': 50, 'fin': 50},
 '202212497': {'att': 90, 'rep': 90, 'mid': 65, 'fin': 85},
 '202212309': {'att': 95, 'rep': 75, 'mid': 50, 'fin': 40},
 '202212493': {'att': 55, 'rep': 60, 'mid': 70, 'fin': 5},
 '202212311': {'att': 95, 'rep': 85, 'mid': 0, 'fin': 15},
 '202212416': {'att': 65, 'rep': 60, 'mid': 35, 'fin': 20},
 '202212489': {'att': 65, 'rep': 50, 'mid': 5, 'fin': 5},
 '202212359': {'att': 90, 'rep': 25, 'mid': 60, 'fin': 25},
 '202212349': {'att': 100, 'rep': 40, 'mid': 40, 'fin': 15},
 '202212403': {'att': 70, 'rep': 25, 'mid': 100, 'fin': 75},
 '202212418': {'att': 100, 'rep': 30, 'mid': 70, 'fin': 70},
 '202212406': {'att': 50, 'rep': 55, 'mid': 55, 'fin': 5},
 '202212485': {'att': 70, 'rep': 35, 'mid': 70, 'fin': 100},
 '202212390': {'att': 70, 'rep': 60, 'mid': 60, 'fin': 80},
 '202212365': {'att': 55, 'rep': 45, 'mid': 90, 'fin': 5},
 '202212338': {'att': 55, 'rep': 55, 'mid': 10, 'fin': 95},
 '202212363': {'att': 65, 'rep': 80, 'mid': 10, 'fin': 30},
 '202212321': {'att': 90, 'rep': 25, 'mid': 35, 'fin': 55},
 '202212499': {'att': 100, 'rep': 30, 'mid': 30, 'fin': 85},
 '202212340': {'att': 70, 'rep': 85, 'mid': 70, 'fin': 65},
 '202212421': {'att': 60, 'rep': 100, 'mid': 45, 'fin': 100},
 '202212407': {'att': 70, 'rep': 25, 'mid': 100, 'fin': 15},
 '202212439': {'att': 70, 'rep': 35, 'mid': 80, 'fin': 25},
 '202212488': {'att': 65, 'rep': 60, 'mid': 30, 'fin': 35},
 '202212453': {'att': 95, 'rep': 35, 'mid': 40, 'fin': 95},
 '202212482': {'att': 50, 'rep': 80, 'mid': 65, 'fin': 90},
 '202212334': {'att': 100, 'rep': 40, 'mid': 80, 'fin': 80},
 '202212322': {'att': 55, 'rep': 30, 'mid': 95, 'fin': 100},
 '202212353': {'att': 65, 'rep': 40, 'mid': 65, 'fin': 70},
 '202212491': {'att': 55, 'rep': 70, 'mid': 40, 'fin': 95},
 '202212352': {'att': 65, 'rep': 85, 'mid': 25, 'fin': 85},
 '202212315': {'att': 85, 'rep': 85, 'mid': 100, 'fin': 10},
 '202212470': {'att': 80, 'rep': 65, 'mid': 35, 'fin': 60},
 '202212436': {'att': 50, 'rep': 95, 'mid': 45, 'fin': 85}}

학번 ’202212460’에 해당하는 학생의 출석점수를 알고 싶다면?

(풀이1)

test_dic['202212460']['att']

55

(풀이2)

test_ndarray

array([['202212377', '202212473', '202212310', '202212460', '202212320',
        '202212329', '202212408', '202212319', '202212348', '202212306',
        '202212308', '202212366', '202212367', '202212461', '202212354',
        '202212361', '202212400', '202212490', '202212404', '202212326',
        '202212452', '202212362', '202212396', '202212356', '202212305',
        '202212398', '202212410', '202212385', '202212430', '202212498',
        '202212423', '202212327', '202212347', '202212483', '202212447',
        '202212496', '202212358', '202212399', '202212459', '202212313',
        '202212304', '202212431', '202212325', '202212471', '202212463',
        '202212441', '202212445', '202212323', '202212442', '202212346',
        '202212411', '202212468', '202212331', '202212345', '202212339',
        '202212383', '202212462', '202212344', '202212472', '202212437',
        '202212336', '202212438', '202212454', '202212384', '202212402',
        '202212397', '202212318', '202212371', '202212469', '202212379',
        '202212364', '202212450', '202212337', '202212458', '202212494',
        '202212478', '202212373', '202212474', '202212455', '202212317',
        '202212341', '202212386', '202212328', '202212417', '202212370',
        '202212486', '202212333', '202212360', '202212350', '202212382',
        '202212392', '202212449', '202212394', '202212444', '202212487',
        '202212425', '202212312', '202212448', '202212434', '202212451',
        '202212433', '202212424', '202212351', '202212324', '202212314',
        '202212446', '202212401', '202212307', '202212300', '202212342',
        '202212479', '202212443', '202212387', '202212372', '202212376',
        '202212466', '202212391', '202212368', '202212427', '202212414',
        '202212426', '202212316', '202212355', '202212477', '202212484',
        '202212456', '202212500', '202212381', '202212335', '202212475',
        '202212343', '202212412', '202212428', '202212330', '202212375',
        '202212413', '202212303', '202212374', '202212409', '202212440',
        '202212393', '202212492', '202212357', '202212465', '202212415',
        '202212405', '202212435', '202212380', '202212369', '202212467',
        '202212429', '202212495', '202212420', '202212302', '202212481',
        '202212422', '202212388', '202212480', '202212378', '202212457',
        '202212419', '202212432', '202212395', '202212464', '202212476',
        '202212332', '202212301', '202212497', '202212309', '202212493',
        '202212311', '202212416', '202212489', '202212359', '202212349',
        '202212403', '202212418', '202212406', '202212485', '202212390',
        '202212365', '202212338', '202212363', '202212321', '202212499',
        '202212340', '202212421', '202212407', '202212439', '202212488',
        '202212453', '202212482', '202212334', '202212322', '202212353',
        '202212491', '202212352', '202212315', '202212470', '202212436'],
       ['65', '95', '65', '55', '80', '75', '65', '60', '95', '90', '55',
        '95', '95', '50', '50', '95', '50', '65', '70', '90', '80', '55',
        '65', '70', '85', '90', '100', '80', '80', '55', '75', '80',
        '95', '95', '100', '100', '80', '70', '85', '100', '95', '75',
        '70', '50', '100', '75', '85', '80', '95', '65', '90', '65',
        '80', '95', '65', '90', '95', '100', '50', '90', '50', '80',
        '50', '70', '65', '60', '50', '60', '70', '75', '50', '85', '80',
        '55', '85', '65', '85', '60', '95', '85', '60', '100', '100',
        '55', '70', '70', '55', '100', '85', '100', '60', '65', '65',
        '75', '50', '75', '100', '90', '55', '75', '60', '85', '60',
        '70', '80', '65', '85', '100', '65', '85', '85', '85', '100',
        '95', '95', '55', '85', '75', '95', '85', '95', '100', '95',
        '95', '100', '55', '60', '85', '75', '55', '70', '80', '90',
        '90', '60', '90', '75', '60', '55', '70', '75', '90', '50', '55',
        '90', '75', '90', '85', '100', '55', '70', '90', '55', '100',
        '50', '100', '70', '85', '55', '75', '65', '70', '60', '100',
        '75', '90', '95', '90', '95', '55', '95', '65', '65', '90',
        '100', '70', '100', '50', '70', '70', '55', '55', '65', '90',
        '100', '70', '60', '70', '70', '65', '95', '50', '100', '55',
        '65', '55', '65', '85', '80', '50'],
       ['45', '30', '85', '35', '60', '40', '70', '25', '55', '25', '45',
        '60', '35', '55', '65', '100', '65', '85', '65', '70', '45',
        '45', '35', '25', '55', '30', '65', '70', '35', '75', '75', '95',
        '60', '60', '75', '55', '60', '80', '65', '95', '60', '40', '80',
        '45', '100', '50', '50', '35', '45', '85', '30', '65', '65',
        '80', '75', '55', '25', '50', '55', '70', '55', '50', '75', '40',
        '85', '30', '65', '95', '70', '45', '60', '90', '25', '75', '30',
        '30', '95', '25', '45', '50', '50', '75', '90', '100', '60',
        '90', '50', '100', '70', '85', '65', '75', '25', '75', '55',
        '30', '50', '30', '100', '60', '25', '35', '100', '55', '65',
        '35', '70', '30', '70', '55', '95', '70', '35', '45', '85', '50',
        '50', '90', '70', '35', '50', '65', '70', '70', '60', '35', '90',
        '90', '85', '30', '60', '30', '85', '25', '85', '50', '95', '40',
        '100', '75', '65', '75', '75', '45', '70', '30', '85', '70',
        '75', '35', '75', '90', '60', '90', '55', '100', '45', '95',
        '25', '30', '85', '90', '85', '25', '25', '95', '90', '90', '75',
        '60', '85', '60', '50', '25', '40', '25', '30', '55', '35', '60',
        '45', '55', '80', '25', '30', '85', '100', '25', '35', '60',
        '35', '80', '40', '30', '40', '70', '85', '85', '65', '95'],
       ['0', '60', '15', '35', '55', '75', '60', '20', '65', '95', '75',
        '25', '0', '90', '50', '25', '35', '10', '65', '100', '80', '85',
        '45', '50', '30', '30', '50', '50', '25', '20', '85', '5', '65',
        '90', '70', '35', '65', '0', '60', '0', '15', '30', '50', '10',
        '100', '60', '35', '75', '35', '85', '25', '35', '30', '45',
        '50', '100', '95', '80', '35', '35', '15', '55', '65', '90',
        '20', '0', '15', '30', '5', '15', '15', '90', '85', '95', '45',
        '45', '35', '10', '90', '60', '100', '60', '85', '100', '30',
        '95', '0', '45', '90', '65', '35', '75', '40', '50', '80', '20',
        '25', '95', '80', '15', '25', '10', '55', '50', '95', '15',
        '100', '60', '55', '85', '80', '75', '70', '55', '40', '30', '5',
        '85', '10', '80', '80', '75', '70', '20', '10', '25', '40', '85',
        '25', '50', '75', '95', '80', '95', '50', '95', '65', '35', '15',
        '80', '25', '80', '75', '35', '90', '100', '0', '35', '100',
        '20', '90', '55', '40', '5', '25', '90', '70', '85', '95', '10',
        '15', '70', '70', '10', '80', '40', '50', '65', '50', '70', '0',
        '35', '5', '60', '40', '100', '70', '55', '70', '60', '90', '10',
        '10', '35', '30', '70', '45', '100', '80', '30', '40', '65',
        '80', '95', '65', '40', '25', '100', '35', '45'],
       ['10', '10', '20', '5', '70', '85', '75', '35', '90', '50', '30',
        '55', '25', '45', '70', '40', '85', '5', '80', '30', '85', '70',
        '20', '70', '80', '0', '70', '100', '65', '25', '95', '5', '10',
        '75', '25', '85', '55', '10', '60', '25', '45', '10', '25', '10',
        '50', '5', '100', '80', '80', '15', '5', '70', '90', '35', '35',
        '30', '90', '10', '60', '25', '75', '90', '90', '5', '90', '50',
        '0', '70', '0', '75', '50', '90', '20', '90', '15', '15', '25',
        '50', '35', '45', '70', '0', '75', '60', '40', '40', '5', '90',
        '80', '85', '15', '85', '0', '40', '55', '50', '65', '35', '0',
        '40', '50', '60', '40', '75', '85', '65', '0', '65', '70', '90',
        '10', '5', '0', '65', '65', '85', '65', '85', '5', '95', '90',
        '40', '0', '25', '5', '10', '5', '75', '35', '45', '75', '5',
        '15', '5', '20', '95', '40', '0', '80', '0', '20', '25', '20',
        '45', '0', '60', '40', '0', '85', '10', '90', '55', '0', '30',
        '80', '55', '75', '90', '45', '95', '60', '0', '85', '20', '25',
        '80', '50', '85', '40', '5', '15', '20', '5', '25', '15', '75',
        '70', '5', '100', '80', '5', '95', '30', '55', '85', '65', '100',
        '15', '25', '35', '95', '90', '80', '100', '70', '95', '85',
        '10', '60', '85']], dtype='<U21')

test_ndarray.T #학번이 string으로 들어가있어서 모든 자료가 string으로 되어있음.. 

array([['202212377', '65', '45', '0', '10'],
       ['202212473', '95', '30', '60', '10'],
       ['202212310', '65', '85', '15', '20'],
       ['202212460', '55', '35', '35', '5'],
       ['202212320', '80', '60', '55', '70'],
       ['202212329', '75', '40', '75', '85'],
       ['202212408', '65', '70', '60', '75'],
       ['202212319', '60', '25', '20', '35'],
       ['202212348', '95', '55', '65', '90'],
       ['202212306', '90', '25', '95', '50'],
       ['202212308', '55', '45', '75', '30'],
       ['202212366', '95', '60', '25', '55'],
       ['202212367', '95', '35', '0', '25'],
       ['202212461', '50', '55', '90', '45'],
       ['202212354', '50', '65', '50', '70'],
       ['202212361', '95', '100', '25', '40'],
       ['202212400', '50', '65', '35', '85'],
       ['202212490', '65', '85', '10', '5'],
       ['202212404', '70', '65', '65', '80'],
       ['202212326', '90', '70', '100', '30'],
       ['202212452', '80', '45', '80', '85'],
       ['202212362', '55', '45', '85', '70'],
       ['202212396', '65', '35', '45', '20'],
       ['202212356', '70', '25', '50', '70'],
       ['202212305', '85', '55', '30', '80'],
       ['202212398', '90', '30', '30', '0'],
       ['202212410', '100', '65', '50', '70'],
       ['202212385', '80', '70', '50', '100'],
       ['202212430', '80', '35', '25', '65'],
       ['202212498', '55', '75', '20', '25'],
       ['202212423', '75', '75', '85', '95'],
       ['202212327', '80', '95', '5', '5'],
       ['202212347', '95', '60', '65', '10'],
       ['202212483', '95', '60', '90', '75'],
       ['202212447', '100', '75', '70', '25'],
       ['202212496', '100', '55', '35', '85'],
       ['202212358', '80', '60', '65', '55'],
       ['202212399', '70', '80', '0', '10'],
       ['202212459', '85', '65', '60', '60'],
       ['202212313', '100', '95', '0', '25'],
       ['202212304', '95', '60', '15', '45'],
       ['202212431', '75', '40', '30', '10'],
       ['202212325', '70', '80', '50', '25'],
       ['202212471', '50', '45', '10', '10'],
       ['202212463', '100', '100', '100', '50'],
       ['202212441', '75', '50', '60', '5'],
       ['202212445', '85', '50', '35', '100'],
       ['202212323', '80', '35', '75', '80'],
       ['202212442', '95', '45', '35', '80'],
       ['202212346', '65', '85', '85', '15'],
       ['202212411', '90', '30', '25', '5'],
       ['202212468', '65', '65', '35', '70'],
       ['202212331', '80', '65', '30', '90'],
       ['202212345', '95', '80', '45', '35'],
       ['202212339', '65', '75', '50', '35'],
       ['202212383', '90', '55', '100', '30'],
       ['202212462', '95', '25', '95', '90'],
       ['202212344', '100', '50', '80', '10'],
       ['202212472', '50', '55', '35', '60'],
       ['202212437', '90', '70', '35', '25'],
       ['202212336', '50', '55', '15', '75'],
       ['202212438', '80', '50', '55', '90'],
       ['202212454', '50', '75', '65', '90'],
       ['202212384', '70', '40', '90', '5'],
       ['202212402', '65', '85', '20', '90'],
       ['202212397', '60', '30', '0', '50'],
       ['202212318', '50', '65', '15', '0'],
       ['202212371', '60', '95', '30', '70'],
       ['202212469', '70', '70', '5', '0'],
       ['202212379', '75', '45', '15', '75'],
       ['202212364', '50', '60', '15', '50'],
       ['202212450', '85', '90', '90', '90'],
       ['202212337', '80', '25', '85', '20'],
       ['202212458', '55', '75', '95', '90'],
       ['202212494', '85', '30', '45', '15'],
       ['202212478', '65', '30', '45', '15'],
       ['202212373', '85', '95', '35', '25'],
       ['202212474', '60', '25', '10', '50'],
       ['202212455', '95', '45', '90', '35'],
       ['202212317', '85', '50', '60', '45'],
       ['202212341', '60', '50', '100', '70'],
       ['202212386', '100', '75', '60', '0'],
       ['202212328', '100', '90', '85', '75'],
       ['202212417', '55', '100', '100', '60'],
       ['202212370', '70', '60', '30', '40'],
       ['202212486', '70', '90', '95', '40'],
       ['202212333', '55', '50', '0', '5'],
       ['202212360', '100', '100', '45', '90'],
       ['202212350', '85', '70', '90', '80'],
       ['202212382', '100', '85', '65', '85'],
       ['202212392', '60', '65', '35', '15'],
       ['202212449', '65', '75', '75', '85'],
       ['202212394', '65', '25', '40', '0'],
       ['202212444', '75', '75', '50', '40'],
       ['202212487', '50', '55', '80', '55'],
       ['202212425', '75', '30', '20', '50'],
       ['202212312', '100', '50', '25', '65'],
       ['202212448', '90', '30', '95', '35'],
       ['202212434', '55', '100', '80', '0'],
       ['202212451', '75', '60', '15', '40'],
       ['202212433', '60', '25', '25', '50'],
       ['202212424', '85', '35', '10', '60'],
       ['202212351', '60', '100', '55', '40'],
       ['202212324', '70', '55', '50', '75'],
       ['202212314', '80', '65', '95', '85'],
       ['202212446', '65', '35', '15', '65'],
       ['202212401', '85', '70', '100', '0'],
       ['202212307', '100', '30', '60', '65'],
       ['202212300', '65', '70', '55', '70'],
       ['202212342', '85', '55', '85', '90'],
       ['202212479', '85', '95', '80', '10'],
       ['202212443', '85', '70', '75', '5'],
       ['202212387', '100', '35', '70', '0'],
       ['202212372', '95', '45', '55', '65'],
       ['202212376', '95', '85', '40', '65'],
       ['202212466', '55', '50', '30', '85'],
       ['202212391', '85', '50', '5', '65'],
       ['202212368', '75', '90', '85', '85'],
       ['202212427', '95', '70', '10', '5'],
       ['202212414', '85', '35', '80', '95'],
       ['202212426', '95', '50', '80', '90'],
       ['202212316', '100', '65', '75', '40'],
       ['202212355', '95', '70', '70', '0'],
       ['202212477', '95', '70', '20', '25'],
       ['202212484', '100', '60', '10', '5'],
       ['202212456', '55', '35', '25', '10'],
       ['202212500', '60', '90', '40', '5'],
       ['202212381', '85', '90', '85', '75'],
       ['202212335', '75', '85', '25', '35'],
       ['202212475', '55', '30', '50', '45'],
       ['202212343', '70', '60', '75', '75'],
       ['202212412', '80', '30', '95', '5'],
       ['202212428', '90', '85', '80', '15'],
       ['202212330', '90', '25', '95', '5'],
       ['202212375', '60', '85', '50', '20'],
       ['202212413', '90', '50', '95', '95'],
       ['202212303', '75', '95', '65', '40'],
       ['202212374', '60', '40', '35', '0'],
       ['202212409', '55', '100', '15', '80'],
       ['202212440', '70', '75', '80', '0'],
       ['202212393', '75', '65', '25', '20'],
       ['202212492', '90', '75', '80', '25'],
       ['202212357', '50', '75', '75', '20'],
       ['202212465', '55', '45', '35', '45'],
       ['202212415', '90', '70', '90', '0'],
       ['202212405', '75', '30', '100', '60'],
       ['202212435', '90', '85', '0', '40'],
       ['202212380', '85', '70', '35', '0'],
       ['202212369', '100', '75', '100', '85'],
       ['202212467', '55', '35', '20', '10'],
       ['202212429', '70', '75', '90', '90'],
       ['202212495', '90', '90', '55', '55'],
       ['202212420', '55', '60', '40', '0'],
       ['202212302', '100', '90', '5', '30'],
       ['202212481', '50', '55', '25', '80'],
       ['202212422', '100', '100', '90', '55'],
       ['202212388', '70', '45', '70', '75'],
       ['202212480', '85', '95', '85', '90'],
       ['202212378', '55', '25', '95', '45'],
       ['202212457', '75', '30', '10', '95'],
       ['202212419', '65', '85', '15', '60'],
       ['202212432', '70', '90', '70', '0'],
       ['202212395', '60', '85', '70', '85'],
       ['202212464', '100', '25', '10', '20'],
       ['202212476', '75', '25', '80', '25'],
       ['202212332', '90', '95', '40', '80'],
       ['202212301', '95', '90', '50', '50'],
       ['202212497', '90', '90', '65', '85'],
       ['202212309', '95', '75', '50', '40'],
       ['202212493', '55', '60', '70', '5'],
       ['202212311', '95', '85', '0', '15'],
       ['202212416', '65', '60', '35', '20'],
       ['202212489', '65', '50', '5', '5'],
       ['202212359', '90', '25', '60', '25'],
       ['202212349', '100', '40', '40', '15'],
       ['202212403', '70', '25', '100', '75'],
       ['202212418', '100', '30', '70', '70'],
       ['202212406', '50', '55', '55', '5'],
       ['202212485', '70', '35', '70', '100'],
       ['202212390', '70', '60', '60', '80'],
       ['202212365', '55', '45', '90', '5'],
       ['202212338', '55', '55', '10', '95'],
       ['202212363', '65', '80', '10', '30'],
       ['202212321', '90', '25', '35', '55'],
       ['202212499', '100', '30', '30', '85'],
       ['202212340', '70', '85', '70', '65'],
       ['202212421', '60', '100', '45', '100'],
       ['202212407', '70', '25', '100', '15'],
       ['202212439', '70', '35', '80', '25'],
       ['202212488', '65', '60', '30', '35'],
       ['202212453', '95', '35', '40', '95'],
       ['202212482', '50', '80', '65', '90'],
       ['202212334', '100', '40', '80', '80'],
       ['202212322', '55', '30', '95', '100'],
       ['202212353', '65', '40', '65', '70'],
       ['202212491', '55', '70', '40', '95'],
       ['202212352', '65', '85', '25', '85'],
       ['202212315', '85', '85', '100', '10'],
       ['202212470', '80', '65', '35', '60'],
       ['202212436', '50', '95', '45', '85']], dtype='<U21')

test_ndarray

array([[202212377,        65,        45,         0,        10],
       [202212473,        95,        30,        60,        10],
       [202212310,        65,        85,        15,        20],
       [202212460,        55,        35,        35,         5],
       [202212320,        80,        60,        55,        70],
       [202212329,        75,        40,        75,        85],
       [202212408,        65,        70,        60,        75],
       [202212319,        60,        25,        20,        35],
       [202212348,        95,        55,        65,        90],
       [202212306,        90,        25,        95,        50],
       [202212308,        55,        45,        75,        30],
       [202212366,        95,        60,        25,        55],
       [202212367,        95,        35,         0,        25],
       [202212461,        50,        55,        90,        45],
       [202212354,        50,        65,        50,        70],
       [202212361,        95,       100,        25,        40],
       [202212400,        50,        65,        35,        85],
       [202212490,        65,        85,        10,         5],
       [202212404,        70,        65,        65,        80],
       [202212326,        90,        70,       100,        30],
       [202212452,        80,        45,        80,        85],
       [202212362,        55,        45,        85,        70],
       [202212396,        65,        35,        45,        20],
       [202212356,        70,        25,        50,        70],
       [202212305,        85,        55,        30,        80],
       [202212398,        90,        30,        30,         0],
       [202212410,       100,        65,        50,        70],
       [202212385,        80,        70,        50,       100],
       [202212430,        80,        35,        25,        65],
       [202212498,        55,        75,        20,        25],
       [202212423,        75,        75,        85,        95],
       [202212327,        80,        95,         5,         5],
       [202212347,        95,        60,        65,        10],
       [202212483,        95,        60,        90,        75],
       [202212447,       100,        75,        70,        25],
       [202212496,       100,        55,        35,        85],
       [202212358,        80,        60,        65,        55],
       [202212399,        70,        80,         0,        10],
       [202212459,        85,        65,        60,        60],
       [202212313,       100,        95,         0,        25],
       [202212304,        95,        60,        15,        45],
       [202212431,        75,        40,        30,        10],
       [202212325,        70,        80,        50,        25],
       [202212471,        50,        45,        10,        10],
       [202212463,       100,       100,       100,        50],
       [202212441,        75,        50,        60,         5],
       [202212445,        85,        50,        35,       100],
       [202212323,        80,        35,        75,        80],
       [202212442,        95,        45,        35,        80],
       [202212346,        65,        85,        85,        15],
       [202212411,        90,        30,        25,         5],
       [202212468,        65,        65,        35,        70],
       [202212331,        80,        65,        30,        90],
       [202212345,        95,        80,        45,        35],
       [202212339,        65,        75,        50,        35],
       [202212383,        90,        55,       100,        30],
       [202212462,        95,        25,        95,        90],
       [202212344,       100,        50,        80,        10],
       [202212472,        50,        55,        35,        60],
       [202212437,        90,        70,        35,        25],
       [202212336,        50,        55,        15,        75],
       [202212438,        80,        50,        55,        90],
       [202212454,        50,        75,        65,        90],
       [202212384,        70,        40,        90,         5],
       [202212402,        65,        85,        20,        90],
       [202212397,        60,        30,         0,        50],
       [202212318,        50,        65,        15,         0],
       [202212371,        60,        95,        30,        70],
       [202212469,        70,        70,         5,         0],
       [202212379,        75,        45,        15,        75],
       [202212364,        50,        60,        15,        50],
       [202212450,        85,        90,        90,        90],
       [202212337,        80,        25,        85,        20],
       [202212458,        55,        75,        95,        90],
       [202212494,        85,        30,        45,        15],
       [202212478,        65,        30,        45,        15],
       [202212373,        85,        95,        35,        25],
       [202212474,        60,        25,        10,        50],
       [202212455,        95,        45,        90,        35],
       [202212317,        85,        50,        60,        45],
       [202212341,        60,        50,       100,        70],
       [202212386,       100,        75,        60,         0],
       [202212328,       100,        90,        85,        75],
       [202212417,        55,       100,       100,        60],
       [202212370,        70,        60,        30,        40],
       [202212486,        70,        90,        95,        40],
       [202212333,        55,        50,         0,         5],
       [202212360,       100,       100,        45,        90],
       [202212350,        85,        70,        90,        80],
       [202212382,       100,        85,        65,        85],
       [202212392,        60,        65,        35,        15],
       [202212449,        65,        75,        75,        85],
       [202212394,        65,        25,        40,         0],
       [202212444,        75,        75,        50,        40],
       [202212487,        50,        55,        80,        55],
       [202212425,        75,        30,        20,        50],
       [202212312,       100,        50,        25,        65],
       [202212448,        90,        30,        95,        35],
       [202212434,        55,       100,        80,         0],
       [202212451,        75,        60,        15,        40],
       [202212433,        60,        25,        25,        50],
       [202212424,        85,        35,        10,        60],
       [202212351,        60,       100,        55,        40],
       [202212324,        70,        55,        50,        75],
       [202212314,        80,        65,        95,        85],
       [202212446,        65,        35,        15,        65],
       [202212401,        85,        70,       100,         0],
       [202212307,       100,        30,        60,        65],
       [202212300,        65,        70,        55,        70],
       [202212342,        85,        55,        85,        90],
       [202212479,        85,        95,        80,        10],
       [202212443,        85,        70,        75,         5],
       [202212387,       100,        35,        70,         0],
       [202212372,        95,        45,        55,        65],
       [202212376,        95,        85,        40,        65],
       [202212466,        55,        50,        30,        85],
       [202212391,        85,        50,         5,        65],
       [202212368,        75,        90,        85,        85],
       [202212427,        95,        70,        10,         5],
       [202212414,        85,        35,        80,        95],
       [202212426,        95,        50,        80,        90],
       [202212316,       100,        65,        75,        40],
       [202212355,        95,        70,        70,         0],
       [202212477,        95,        70,        20,        25],
       [202212484,       100,        60,        10,         5],
       [202212456,        55,        35,        25,        10],
       [202212500,        60,        90,        40,         5],
       [202212381,        85,        90,        85,        75],
       [202212335,        75,        85,        25,        35],
       [202212475,        55,        30,        50,        45],
       [202212343,        70,        60,        75,        75],
       [202212412,        80,        30,        95,         5],
       [202212428,        90,        85,        80,        15],
       [202212330,        90,        25,        95,         5],
       [202212375,        60,        85,        50,        20],
       [202212413,        90,        50,        95,        95],
       [202212303,        75,        95,        65,        40],
       [202212374,        60,        40,        35,         0],
       [202212409,        55,       100,        15,        80],
       [202212440,        70,        75,        80,         0],
       [202212393,        75,        65,        25,        20],
       [202212492,        90,        75,        80,        25],
       [202212357,        50,        75,        75,        20],
       [202212465,        55,        45,        35,        45],
       [202212415,        90,        70,        90,         0],
       [202212405,        75,        30,       100,        60],
       [202212435,        90,        85,         0,        40],
       [202212380,        85,        70,        35,         0],
       [202212369,       100,        75,       100,        85],
       [202212467,        55,        35,        20,        10],
       [202212429,        70,        75,        90,        90],
       [202212495,        90,        90,        55,        55],
       [202212420,        55,        60,        40,         0],
       [202212302,       100,        90,         5,        30],
       [202212481,        50,        55,        25,        80],
       [202212422,       100,       100,        90,        55],
       [202212388,        70,        45,        70,        75],
       [202212480,        85,        95,        85,        90],
       [202212378,        55,        25,        95,        45],
       [202212457,        75,        30,        10,        95],
       [202212419,        65,        85,        15,        60],
       [202212432,        70,        90,        70,         0],
       [202212395,        60,        85,        70,        85],
       [202212464,       100,        25,        10,        20],
       [202212476,        75,        25,        80,        25],
       [202212332,        90,        95,        40,        80],
       [202212301,        95,        90,        50,        50],
       [202212497,        90,        90,        65,        85],
       [202212309,        95,        75,        50,        40],
       [202212493,        55,        60,        70,         5],
       [202212311,        95,        85,         0,        15],
       [202212416,        65,        60,        35,        20],
       [202212489,        65,        50,         5,         5],
       [202212359,        90,        25,        60,        25],
       [202212349,       100,        40,        40,        15],
       [202212403,        70,        25,       100,        75],
       [202212418,       100,        30,        70,        70],
       [202212406,        50,        55,        55,         5],
       [202212485,        70,        35,        70,       100],
       [202212390,        70,        60,        60,        80],
       [202212365,        55,        45,        90,         5],
       [202212338,        55,        55,        10,        95],
       [202212363,        65,        80,        10,        30],
       [202212321,        90,        25,        35,        55],
       [202212499,       100,        30,        30,        85],
       [202212340,        70,        85,        70,        65],
       [202212421,        60,       100,        45,       100],
       [202212407,        70,        25,       100,        15],
       [202212439,        70,        35,        80,        25],
       [202212488,        65,        60,        30,        35],
       [202212453,        95,        35,        40,        95],
       [202212482,        50,        80,        65,        90],
       [202212334,       100,        40,        80,        80],
       [202212322,        55,        30,        95,       100],
       [202212353,        65,        40,        65,        70],
       [202212491,        55,        70,        40,        95],
       [202212352,        65,        85,        25,        85],
       [202212315,        85,        85,       100,        10],
       [202212470,        80,        65,        35,        60],
       [202212436,        50,        95,        45,        85]])

(풀이2)

test_ndarray[test_ndarray[:,0] == 202212460]

array([[202212460,        55,        35,        35,         5]])

test_ndarray[test_ndarray[:,0] == 202212460,1]   # 이게뭐여? 가독성이 떨어짐

array([55])

(풀이2)가 (풀이1)에 비하여 불편한 점 - test_ndarray의 첫칼럼은 student id이고 두번째 칼럼은 att라는 사실을 암기하고 있어야 한다. - student id가 아니고 만약에 학생이름(문자형)을 써서 데이터를 정리한다면 모든 자료형은 문자형이 되어야 한다. - 작성한 코드 가독성이 없다. (위치로 접근하므로)

- 요약: hash 스타일로 정보를 추출하는 것이 유용할 때가 있다. 그리고 보통 hash 스타일로 정보를 뽑는것이 유리함.

numpy는 정보추출을 위해 개발된 자료형이 아니라 행렬 및 벡터의 수학연산을 지원하기 위해 개발된 자료형이다.)

- 소망: 정보를 추출할때는 hash 스타일도 유용하다는 것은 ㅇㅋ $\to$ 하지만 난 가끔 넘파이스타일로 정보를 뽑고 싶엉. 그리고 딕셔너리 형태가 아니고 엑셀처럼(행렬처럼) 데이터를 보고 ㅅ피다!!! $\to$ pandas 개발

pandas 개발동기

엑셀처럼 데이터를 테이블 형태로 정리하고 싶다.

np.random.seed(43052)
att = np.random.choice(np.arange(10,21)*5,20)
rep = np.random.choice(np.arange(5,21)*5,20)
mid = np.random.choice(np.arange(0,21)*5,20)
fin = np.random.choice(np.arange(0,21)*5,20)
key = ['202212'+str(s) for s in np.random.choice(np.arange(300,501),20,replace=False)]
test_dic = {key[i] : {'att':att[i], 'rep':rep[i], 'mid':mid[i], 'fin':fin[i]} for i in range(20)}
test_ndarray = np.array([key,att,rep,mid,fin],dtype=np.int64).T

test_dic

{'202212380': {'att': 65, 'rep': 55, 'mid': 50, 'fin': 40},
 '202212370': {'att': 95, 'rep': 100, 'mid': 50, 'fin': 80},
 '202212363': {'att': 65, 'rep': 90, 'mid': 60, 'fin': 30},
 '202212488': {'att': 55, 'rep': 80, 'mid': 75, 'fin': 80},
 '202212312': {'att': 80, 'rep': 30, 'mid': 30, 'fin': 100},
 '202212377': {'att': 75, 'rep': 40, 'mid': 100, 'fin': 15},
 '202212463': {'att': 65, 'rep': 45, 'mid': 45, 'fin': 90},
 '202212471': {'att': 60, 'rep': 60, 'mid': 25, 'fin': 0},
 '202212400': {'att': 95, 'rep': 65, 'mid': 20, 'fin': 10},
 '202212469': {'att': 90, 'rep': 80, 'mid': 80, 'fin': 20},
 '202212318': {'att': 55, 'rep': 75, 'mid': 35, 'fin': 25},
 '202212432': {'att': 95, 'rep': 95, 'mid': 45, 'fin': 0},
 '202212443': {'att': 95, 'rep': 55, 'mid': 15, 'fin': 35},
 '202212367': {'att': 50, 'rep': 80, 'mid': 40, 'fin': 30},
 '202212458': {'att': 50, 'rep': 55, 'mid': 15, 'fin': 85},
 '202212396': {'att': 95, 'rep': 30, 'mid': 30, 'fin': 95},
 '202212482': {'att': 50, 'rep': 50, 'mid': 45, 'fin': 10},
 '202212452': {'att': 65, 'rep': 55, 'mid': 15, 'fin': 45},
 '202212387': {'att': 70, 'rep': 70, 'mid': 40, 'fin': 35},
 '202212354': {'att': 90, 'rep': 90, 'mid': 80, 'fin': 90}}

• 테이블 형태로 보고 싶다.

(방법1) - 행렬이기는 하지만 방법 2,3,4에 비하여 우리가 원하는 만큼 가독성을 주는 형태는 아님

test_ndarray = np.array([key,att,rep,mid,fin],dtype=np.int64).T
test_ndarray

array([[202212380,        65,        55,        50,        40],
       [202212370,        95,       100,        50,        80],
       [202212363,        65,        90,        60,        30],
       [202212488,        55,        80,        75,        80],
       [202212312,        80,        30,        30,       100],
       [202212377,        75,        40,       100,        15],
       [202212463,        65,        45,        45,        90],
       [202212471,        60,        60,        25,         0],
       [202212400,        95,        65,        20,        10],
       [202212469,        90,        80,        80,        20],
       [202212318,        55,        75,        35,        25],
       [202212432,        95,        95,        45,         0],
       [202212443,        95,        55,        15,        35],
       [202212367,        50,        80,        40,        30],
       [202212458,        50,        55,        15,        85],
       [202212396,        95,        30,        30,        95],
       [202212482,        50,        50,        45,        10],
       [202212452,        65,        55,        15,        45],
       [202212387,        70,        70,        40,        35],
       [202212354,        90,        90,        80,        90]])

(방법2)

pd.DataFrame(test_dic).T

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212400	95	65	20	10
202212469	90	80	80	20
202212318	55	75	35	25
202212432	95	95	45	0
202212443	95	55	15	35
202212367	50	80	40	30
202212458	50	55	15	85
202212396	95	30	30	95
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

(방법3)

test_dic2 = {'att':{key[i]:att[i] for i in range(20)},
             'rep':{key[i]:rep[i] for i in range(20)},
             'mid':{key[i]:mid[i] for i in range(20)},
             'fin':{key[i]:fin[i] for i in range(20)}}

test_dic2

{'att': {'202212380': 65,
  '202212370': 95,
  '202212363': 65,
  '202212488': 55,
  '202212312': 80,
  '202212377': 75,
  '202212463': 65,
  '202212471': 60,
  '202212400': 95,
  '202212469': 90,
  '202212318': 55,
  '202212432': 95,
  '202212443': 95,
  '202212367': 50,
  '202212458': 50,
  '202212396': 95,
  '202212482': 50,
  '202212452': 65,
  '202212387': 70,
  '202212354': 90},
 'rep': {'202212380': 55,
  '202212370': 100,
  '202212363': 90,
  '202212488': 80,
  '202212312': 30,
  '202212377': 40,
  '202212463': 45,
  '202212471': 60,
  '202212400': 65,
  '202212469': 80,
  '202212318': 75,
  '202212432': 95,
  '202212443': 55,
  '202212367': 80,
  '202212458': 55,
  '202212396': 30,
  '202212482': 50,
  '202212452': 55,
  '202212387': 70,
  '202212354': 90},
 'mid': {'202212380': 50,
  '202212370': 50,
  '202212363': 60,
  '202212488': 75,
  '202212312': 30,
  '202212377': 100,
  '202212463': 45,
  '202212471': 25,
  '202212400': 20,
  '202212469': 80,
  '202212318': 35,
  '202212432': 45,
  '202212443': 15,
  '202212367': 40,
  '202212458': 15,
  '202212396': 30,
  '202212482': 45,
  '202212452': 15,
  '202212387': 40,
  '202212354': 80},
 'fin': {'202212380': 40,
  '202212370': 80,
  '202212363': 30,
  '202212488': 80,
  '202212312': 100,
  '202212377': 15,
  '202212463': 90,
  '202212471': 0,
  '202212400': 10,
  '202212469': 20,
  '202212318': 25,
  '202212432': 0,
  '202212443': 35,
  '202212367': 30,
  '202212458': 85,
  '202212396': 95,
  '202212482': 10,
  '202212452': 45,
  '202212387': 35,
  '202212354': 90}}

pd.DataFrame(test_dic2)

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212400	95	65	20	10
202212469	90	80	80	20
202212318	55	75	35	25
202212432	95	95	45	0
202212443	95	55	15	35
202212367	50	80	40	30
202212458	50	55	15	85
202212396	95	30	30	95
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

(방법4)

df = pd.DataFrame({'att':att, 'rep':rep, 'mid':mid, 'fin':fin}, index=key)
df

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212400	95	65	20	10
202212469	90	80	80	20
202212318	55	75	35	25
202212432	95	95	45	0
202212443	95	55	15	35
202212367	50	80	40	30
202212458	50	55	15	85
202212396	95	30	30	95
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

(방법5)

df = pd.DataFrame({'att':att, 'rep':rep, 'mid':mid, 'fin':fin})
df

	att	rep	mid	fin
0	65	55	50	40
1	95	100	50	80
2	65	90	60	30
3	55	80	75	80
4	80	30	30	100
5	75	40	100	15
6	65	45	45	90
7	60	60	25	0
8	95	65	20	10
9	90	80	80	20
10	55	75	35	25
11	95	95	45	0
12	95	55	15	35
13	50	80	40	30
14	50	55	15	85
15	95	30	30	95
16	50	50	45	10
17	65	55	15	45
18	70	70	40	35
19	90	90	80	90

df=df.set_index([key])   #인덱스를 set_index로 설정해줄 수 있음
df

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212400	95	65	20	10
202212469	90	80	80	20
202212318	55	75	35	25
202212432	95	95	45	0
202212443	95	55	15	35
202212367	50	80	40	30
202212458	50	55	15	85
202212396	95	30	30	95
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

해싱으로 원하는 정보를 뽑으면 좋겠다. (마치 딕셔너리처럼)

- 예제1: 출설점수를 출력

test_dic2['att']

{'202212380': 65,
 '202212370': 95,
 '202212363': 65,
 '202212488': 55,
 '202212312': 80,
 '202212377': 75,
 '202212463': 65,
 '202212471': 60,
 '202212400': 95,
 '202212469': 90,
 '202212318': 55,
 '202212432': 95,
 '202212443': 95,
 '202212367': 50,
 '202212458': 50,
 '202212396': 95,
 '202212482': 50,
 '202212452': 65,
 '202212387': 70,
 '202212354': 90}

df['att']

202212380    65
202212370    95
202212363    65
202212488    55
202212312    80
202212377    75
202212463    65
202212471    60
202212400    95
202212469    90
202212318    55
202212432    95
202212443    95
202212367    50
202212458    50
202212396    95
202212482    50
202212452    65
202212387    70
202212354    90
Name: att, dtype: int64

- 예제2 : 학번 202212380의 출석점수 출력

test_dic2['att']['202212380']

65

df['att']['202212380']

65

인덱싱으로 정보를 뽑는 기능도 지원을 하면 좋겠따. (마치 리스트나 넘파이처럼)

- 예제1: 첫번째 학생의 기말고사 성적을 출력하고 싶다.

test_ndarray[0,-1]

40

df.iloc[0,-1]  

40

• 벼락치기: df에서 iloc라는 특수기능을 이용하면 넘파이 인덱싱처럼 원소출력이 가능하다.

-예제2: 홀수번째 학생의 점수를 뽑고 싶다.

test_ndarray[::2]

array([[202212380,        65,        55,        50,        40],
       [202212363,        65,        90,        60,        30],
       [202212312,        80,        30,        30,       100],
       [202212463,        65,        45,        45,        90],
       [202212400,        95,        65,        20,        10],
       [202212318,        55,        75,        35,        25],
       [202212443,        95,        55,        15,        35],
       [202212458,        50,        55,        15,        85],
       [202212482,        50,        50,        45,        10],
       [202212387,        70,        70,        40,        35]])

df.iloc[0::2]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30
202212312	80	30	30	100
202212463	65	45	45	90
202212400	95	65	20	10
202212318	55	75	35	25
202212443	95	55	15	35
202212458	50	55	15	85
202212482	50	50	45	10
202212387	70	70	40	35

- 예제3: 맨 끝에서 3명의 점수를 출력하고 싶다.

test_ndarray[-3:]

array([[202212452,        65,        55,        15,        45],
       [202212387,        70,        70,        40,        35],
       [202212354,        90,        90,        80,        90]])

df.iloc[-3:]

	att	rep	mid	fin
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

- 예제4: 맨 끝에서 3명의 점수를 마지막 2개의 칼럼만 출력하고 싶다.

test_ndarray[-3:,-2:]

array([[15, 45],
       [40, 35],
       [80, 90]])

df.iloc[-3:,-2:]

	mid	fin
202212452	15	45
202212387	40	35
202212354	80	90

궁극: 해싱과 인덱싱을 모두 지원하는 아주 우수한 자료형을 만들고 싶어!

- 예제1: 중간고사 점수가 20점 이상이면서 동시에 출석점수가 60점미만인 학생들의 기말고사 점수를 출력

(방법1) 데이터베이스 스타일

df.query("mid>=20 and att<60")

	att	rep	mid	fin
202212488	55	80	75	80
202212318	55	75	35	25
202212367	50	80	40	30
202212482	50	50	45	10

df.query("mid>=20 and att<60")['fin']

202212488    80
202212318    25
202212367    30
202212482    10
Name: fin, dtype: int64

(방법2) 넘파이 스타일

test_ndarray

array([[202212380,        65,        55,        50,        40],
       [202212370,        95,       100,        50,        80],
       [202212363,        65,        90,        60,        30],
       [202212488,        55,        80,        75,        80],
       [202212312,        80,        30,        30,       100],
       [202212377,        75,        40,       100,        15],
       [202212463,        65,        45,        45,        90],
       [202212471,        60,        60,        25,         0],
       [202212400,        95,        65,        20,        10],
       [202212469,        90,        80,        80,        20],
       [202212318,        55,        75,        35,        25],
       [202212432,        95,        95,        45,         0],
       [202212443,        95,        55,        15,        35],
       [202212367,        50,        80,        40,        30],
       [202212458,        50,        55,        15,        85],
       [202212396,        95,        30,        30,        95],
       [202212482,        50,        50,        45,        10],
       [202212452,        65,        55,        15,        45],
       [202212387,        70,        70,        40,        35],
       [202212354,        90,        90,        80,        90]])

test_ndarray[:,3]>=20 # 중간고사 점수 20점 이상

array([ True,  True,  True,  True,  True,  True,  True,  True,  True,
        True,  True,  True, False,  True, False,  True,  True, False,
        True,  True])

test_ndarray[:,1] < 60 # 출석이 60미만 

array([False, False, False,  True, False, False, False, False, False,
       False,  True, False, False,  True,  True, False,  True, False,
       False, False])

(test_ndarray[:,3]>=20) & (test_ndarray[:,1] < 60)

array([False, False, False,  True, False, False, False, False, False,
       False,  True, False, False,  True, False, False,  True, False,
       False, False])

note: test_ndarray[:,3]>=20 & test_ndarray[:,1] < 60 와 같이 하면 에러가 난다. 가로로 묶어줘야 함

test_ndarray[(test_ndarray[:,3]>=20) & (test_ndarray[:,1] < 60),-1] 

array([80, 25, 30, 10])

• 구현난이도 어려움, 가독성 꽝

- 예제2: ’중간고사점수<기말고사점수’인 학생들의 출석점수 평균을 구하자.

df.query('mid<fin')['att'].mean()

76.66666666666667

pandas 사용법

pandas 공부 1단계

데이터프레임 선언

- 방법1: dictionary에서 만든다.

pd.DataFrame({'att':[30,40,50],'mid':[50,60,70]})  # 리스트

	att	mid
0	30	50
1	40	60
2	50	70

pd.DataFrame({'att':(30,40,50),'mid':(50,60,70)})  # 튜플

	att	mid
0	30	50
1	40	60
2	50	70

pd.DataFrame({'att':np.array([30,40,50]),'mid':np.array([50,60,70])})  # 넘파이어레이

	att	mid
0	30	50
1	40	60
2	50	70

- 방법2: 2차원 ndarray에서 형태변환 통해 만든다.

np.arange(2*3).reshape(2,3)

array([[0, 1, 2],
       [3, 4, 5]])

pd.DataFrame(np.arange(2*3).reshape(2,3))

	0	1	2
0	0	1	2
1	3	4	5

열의 이름 부여

- 방법1: 딕셔너리를 통하여 만들면 딕셔너리의 key가 자동으로 열의 이름이 된다.

pd.DataFrame({'att':np.array([30,40,50]),'mid':np.array([50,60,70])})

	att	mid
0	30	50
1	40	60
2	50	70

- 방법2: pd.DataFrame()의 옵션에 columns를 이용

pd.DataFrame(np.arange(2*3).reshape(2,3),columns=['X1','X2','X3'])

	X1	X2	X3
0	0	1	2
1	3	4	5

- 방법3: df.columns에 원하는 열이름을 덮어씀(1)

df=pd.DataFrame(np.arange(2*3).reshape(2,3))
df

	0	1	2
0	0	1	2
1	3	4	5

df.columns

RangeIndex(start=0, stop=3, step=1)

df.columns=['X1','X2','X3']

df

	X1	X2	X3
0	0	1	2
1	3	4	5

- 방법4: df.columns에 원하는 열이름을 덮어씀(2)

df.columns=pd.Index(['X1','X2','X3'])  # 위와 같은 코드
df

	X1	X2	X3
0	0	1	2
1	3	4	5

방법4가 방법3의 방식보다 컴퓨터가 이해하기 좋다. (=불필요한 에러를 방지할 수 있다.)

df.columns

Index(['X1', 'X2', 'X3'], dtype='object')

type(df.columns)

pandas.core.indexes.base.Index

['X1','X2','X3'], type(['X1','X2','X3'])

(['X1', 'X2', 'X3'], list)

pd.Index(['X1','X2','X3'])

Index(['X1', 'X2', 'X3'], dtype='object')

행의 이름 부여

- 방법1: 중첩 dict이면 nested dic의 key가 알아서 행의 이름으로 된다. (안쪽..)

{'att': {'boram':30, 'iu':40, 'hynn':50}, 'mid':{'boram':5, 'iu':45, 'hynn':90}}

{'att': {'boram': 30, 'iu': 40, 'hynn': 50},
 'mid': {'boram': 5, 'iu': 45, 'hynn': 90}}

pd.DataFrame({'att': {'boram':30, 'iu':40, 'hynn':50}, 'mid':{'boram':5, 'iu':45, 'hynn':90}})

	att	mid
boram	30	5
iu	40	45
hynn	50	90

- 방법2: index옵션 이용

pd.DataFrame({'att': [30, 40,50], 'mid':[5,45, 90]}, index=['boram','iu','hynn'])

	att	mid
boram	30	5
iu	40	45
hynn	50	90

- 방법3: df.index에 덮어씌움

df=pd.DataFrame({'att': [30, 40,50], 'mid':[5,45, 90]})
df

	att	mid
0	30	5
1	40	45
2	50	90

df.index

RangeIndex(start=0, stop=3, step=1)

df.index=['boram','iu','hynn']
df

	att	mid
boram	30	5
iu	40	45
hynn	50	90

df.index= pd.Index(['boram','iu','hynn'])  #이게 컴퓨터가 볼 때 더 안전한 코드!
df

	att	mid
boram	30	5
iu	40	45
hynn	50	90

- 방법4: df.set_index() 를 이용하여 덮어 씌운다.

df=pd.DataFrame({'att': [30, 40,50], 'mid':[5,45, 90]})
df

	att	mid
0	30	5
1	40	45
2	50	90

df.set_index(pd.Index(['boram','iu','hynn']))

	att	mid
boram	30	5
iu	40	45
hynn	50	90

(주의) 아래는 에러가 난다.

df.set_index(['boram','iu','hynn'])

KeyError: "None of ['boram', 'iu', 'hynn'] are in the columns"

df.set_index([['boram','iu','hynn']]) # 꺽쇠를 한번 더 넣어주면 에러를 피할 수 있따.

	att	mid
boram	30	5
iu	40	45
hynn	50	90

자료형, len, shape, for문의 반복변수

df= pd.DataFrame({'att':[30,40,50],'mid':[5,45,90]})
df

	att	mid
0	30	5
1	40	45
2	50	90

- type

type(df)

pandas.core.frame.DataFrame

- len

len(df) #row의 갯수

3

- shape

df.shape

(3, 2)

- for문의 반복변수

for k in df:
    print(k) #딕셔너리와 같다.

att
mid

for k in {'att':[30,40,50],'mid':[5,45,90]}: #딕셔너리
    print(k)

att
mid

pd.Series

- 2차원 ndarray가 pd.DataFrame에 대응한다면 1차원 ndarray는 pd.Series에 대응한다.

a=pd.Series(np.random.randn(10))
a

0    0.106173
1    0.723759
2    0.217990
3    0.194022
4   -0.688990
5   -0.351670
6    0.990933
7    1.212147
8   -0.608965
9    0.032549
dtype: float64

type(a)

pandas.core.series.Series

len(a)

10

a.shape

(10,)

for value in a:
    print(value)   #값들이 반복 넘파이어레이처럼..

0.10617283591748639
0.7237590624253404
0.21798967912700873
0.1940223087322443
-0.6889899757985083
-0.3516696436204985
0.9909329773184973
1.2121468150185186
-0.6089654373693767
0.03254898346416765

pandas공부 2단계

- 데이터

np.random.seed(43052)
att = np.random.choice(np.arange(10,21)*5,20)
rep = np.random.choice(np.arange(5,21)*5,20)
mid = np.random.choice(np.arange(0,21)*5,20)
fin = np.random.choice(np.arange(0,21)*5,20)
key = ['202212'+str(s) for s in np.random.choice(np.arange(300,501),20,replace=False)]

df=pd.DataFrame({'att':att, 'rep':rep, 'mid':mid, 'fin':fin}, index=key)
df

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212400	95	65	20	10
202212469	90	80	80	20
202212318	55	75	35	25
202212432	95	95	45	0
202212443	95	55	15	35
202212367	50	80	40	30
202212458	50	55	15	85
202212396	95	30	30	95
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35
202212354	90	90	80	90

첫번째 칼럼을 선택

- 방법1

df.att 

202212380    65
202212370    95
202212363    65
202212488    55
202212312    80
202212377    75
202212463    65
202212471    60
202212400    95
202212469    90
202212318    55
202212432    95
202212443    95
202212367    50
202212458    50
202212396    95
202212482    50
202212452    65
202212387    70
202212354    90
Name: att, dtype: int64

- 방법2 : dict스타일

df['att']

202212380    65
202212370    95
202212363    65
202212488    55
202212312    80
202212377    75
202212463    65
202212471    60
202212400    95
202212469    90
202212318    55
202212432    95
202212443    95
202212367    50
202212458    50
202212396    95
202212482    50
202212452    65
202212387    70
202212354    90
Name: att, dtype: int64

type(df['att'])

pandas.core.series.Series

- 방법3 : dict스타일

df[['att']]

	att
202212380	65
202212370	95
202212363	65
202212488	55
202212312	80
202212377	75
202212463	65
202212471	60
202212400	95
202212469	90
202212318	55
202212432	95
202212443	95
202212367	50
202212458	50
202212396	95
202212482	50
202212452	65
202212387	70
202212354	90

type(df[['att']])

pandas.core.frame.DataFrame

• df.att 나 df[‘att’]는 series를 리턴하고 df[[‘att’]]는 dataframe을 리턴한다.

- 방법4 : ndarray 스타일

df.iloc[:,0] 

202212380    65
202212370    95
202212363    65
202212488    55
202212312    80
202212377    75
202212463    65
202212471    60
202212400    95
202212469    90
202212318    55
202212432    95
202212443    95
202212367    50
202212458    50
202212396    95
202212482    50
202212452    65
202212387    70
202212354    90
Name: att, dtype: int64

- 방법5: ndarray스타일

df.iloc[:,[0]]

	att
202212380	65
202212370	95
202212363	65
202212488	55
202212312	80
202212377	75
202212463	65
202212471	60
202212400	95
202212469	90
202212318	55
202212432	95
202212443	95
202212367	50
202212458	50
202212396	95
202212482	50
202212452	65
202212387	70
202212354	90

• df.iloc[:,0]은 series를 리턴하고 df.iloc[:,[0]]은 dataframe을 리턴한다.

- 방법6: ndarray 스타일과 dict 스타일의 혼합

df.loc[:,'att'] 

202212380    65
202212370    95
202212363    65
202212488    55
202212312    80
202212377    75
202212463    65
202212471    60
202212400    95
202212469    90
202212318    55
202212432    95
202212443    95
202212367    50
202212458    50
202212396    95
202212482    50
202212452    65
202212387    70
202212354    90
Name: att, dtype: int64

- 방법7: ndarray 스타일과 dict 스타일의 혼합

df.loc[:,['att']] 

	att
202212380	65
202212370	95
202212363	65
202212488	55
202212312	80
202212377	75
202212463	65
202212471	60
202212400	95
202212469	90
202212318	55
202212432	95
202212443	95
202212367	50
202212458	50
202212396	95
202212482	50
202212452	65
202212387	70
202212354	90

• df.loc[:,‘att’]은 series를 리턴하고 df.loc[:,[‘att’]]은 dataframe을 리턴한다.

- 방법7: nparray 스타일 + bool 인덱싱

df.iloc[:,[True,False,False,False]]

	att
202212380	65
202212370	95
202212363	65
202212488	55
202212312	80
202212377	75
202212463	65
202212471	60
202212400	95
202212469	90
202212318	55
202212432	95
202212443	95
202212367	50
202212458	50
202212396	95
202212482	50
202212452	65
202212387	70
202212354	90

- 방법8: ndarray와 dict의 홉합형 + bool 인덱싱

df.loc[:,[True,False,False,False]]

	att
202212380	65
202212370	95
202212363	65
202212488	55
202212312	80
202212377	75
202212463	65
202212471	60
202212400	95
202212469	90
202212318	55
202212432	95
202212443	95
202212367	50
202212458	50
202212396	95
202212482	50
202212452	65
202212387	70
202212354	90

여러개의 칼럼을 선택

- 방법1: dict스타일

df[['att','fin']]

	att	fin
202212380	65	40
202212370	95	80
202212363	65	30
202212488	55	80
202212312	80	100
202212377	75	15
202212463	65	90
202212471	60	0
202212400	95	10
202212469	90	20
202212318	55	25
202212432	95	0
202212443	95	35
202212367	50	30
202212458	50	85
202212396	95	95
202212482	50	10
202212452	65	45
202212387	70	35
202212354	90	90

- 방법2: ndarray 스타일 (정수리스트로 인덱싱, 슬라이싱, 스트라이딩)

df.iloc[:,[0,1]] #정수의 리스트를 전달하여 칼럼추출

	att	rep
202212380	65	55
202212370	95	100
202212363	65	90
202212488	55	80
202212312	80	30
202212377	75	40
202212463	65	45
202212471	60	60
202212400	95	65
202212469	90	80
202212318	55	75
202212432	95	95
202212443	95	55
202212367	50	80
202212458	50	55
202212396	95	30
202212482	50	50
202212452	65	55
202212387	70	70
202212354	90	90

df.iloc[:,0:2]  #슬라이싱, 0,1,2에서 마지막 2는 제외되고 0,1에 해당하는 것만 추출

	att	rep
202212380	65	55
202212370	95	100
202212363	65	90
202212488	55	80
202212312	80	30
202212377	75	40
202212463	65	45
202212471	60	60
202212400	95	65
202212469	90	80
202212318	55	75
202212432	95	95
202212443	95	55
202212367	50	80
202212458	50	55
202212396	95	30
202212482	50	50
202212452	65	55
202212387	70	70
202212354	90	90

df.iloc[:,2:]  #슬라이싱

	mid	fin
202212380	50	40
202212370	50	80
202212363	60	30
202212488	75	80
202212312	30	100
202212377	100	15
202212463	45	90
202212471	25	0
202212400	20	10
202212469	80	20
202212318	35	25
202212432	45	0
202212443	15	35
202212367	40	30
202212458	15	85
202212396	30	95
202212482	45	10
202212452	15	45
202212387	40	35
202212354	80	90

df.iloc[:,::2]  #슬라이싱, 스트라이딩

	att	mid
202212380	65	50
202212370	95	50
202212363	65	60
202212488	55	75
202212312	80	30
202212377	75	100
202212463	65	45
202212471	60	25
202212400	95	20
202212469	90	80
202212318	55	35
202212432	95	45
202212443	95	15
202212367	50	40
202212458	50	15
202212396	95	30
202212482	50	45
202212452	65	15
202212387	70	40
202212354	90	80

- 방법3: ndarray와 dict의 혼합형

df.loc[:,['att','mid']]

	att	mid
202212380	65	50
202212370	95	50
202212363	65	60
202212488	55	75
202212312	80	30
202212377	75	100
202212463	65	45
202212471	60	25
202212400	95	20
202212469	90	80
202212318	55	35
202212432	95	45
202212443	95	15
202212367	50	40
202212458	50	15
202212396	95	30
202212482	50	45
202212452	65	15
202212387	70	40
202212354	90	80

df.loc[:,'att':'rep']  # key로 하는 슬라이싱은 마지막 'rep'까지 표시되어 나온다

	att	rep
202212380	65	55
202212370	95	100
202212363	65	90
202212488	55	80
202212312	80	30
202212377	75	40
202212463	65	45
202212471	60	60
202212400	95	65
202212469	90	80
202212318	55	75
202212432	95	95
202212443	95	55
202212367	50	80
202212458	50	55
202212396	95	30
202212482	50	50
202212452	65	55
202212387	70	70
202212354	90	90

df.loc[:,:'rep']

	att	rep
202212380	65	55
202212370	95	100
202212363	65	90
202212488	55	80
202212312	80	30
202212377	75	40
202212463	65	45
202212471	60	60
202212400	95	65
202212469	90	80
202212318	55	75
202212432	95	95
202212443	95	55
202212367	50	80
202212458	50	55
202212396	95	30
202212482	50	50
202212452	65	55
202212387	70	70
202212354	90	90

df.loc[:,'rep':]

	rep	mid	fin
202212380	55	50	40
202212370	100	50	80
202212363	90	60	30
202212488	80	75	80
202212312	30	30	100
202212377	40	100	15
202212463	45	45	90
202212471	60	25	0
202212400	65	20	10
202212469	80	80	20
202212318	75	35	25
202212432	95	45	0
202212443	55	15	35
202212367	80	40	30
202212458	55	15	85
202212396	30	30	95
202212482	50	45	10
202212452	55	15	45
202212387	70	40	35
202212354	90	80	90

- 방법4: bool을 이용한 인덱싱

df.iloc[:,[True,False,True,False]]

	att	mid
202212380	65	50
202212370	95	50
202212363	65	60
202212488	55	75
202212312	80	30
202212377	75	100
202212463	65	45
202212471	60	25
202212400	95	20
202212469	90	80
202212318	55	35
202212432	95	45
202212443	95	15
202212367	50	40
202212458	50	15
202212396	95	30
202212482	50	45
202212452	65	15
202212387	70	40
202212354	90	80

df.loc[:,[True,False,True,False]]

	att	mid
202212380	65	50
202212370	95	50
202212363	65	60
202212488	55	75
202212312	80	30
202212377	75	100
202212463	65	45
202212471	60	25
202212400	95	20
202212469	90	80
202212318	55	35
202212432	95	45
202212443	95	15
202212367	50	40
202212458	50	15
202212396	95	30
202212482	50	45
202212452	65	15
202212387	70	40
202212354	90	80

test_ndarray[:,range(2)]

array([[202212380,        65],
       [202212370,        95],
       [202212363,        65],
       [202212488,        55],
       [202212312,        80],
       [202212377,        75],
       [202212463,        65],
       [202212471,        60],
       [202212400,        95],
       [202212469,        90],
       [202212318,        55],
       [202212432,        95],
       [202212443,        95],
       [202212367,        50],
       [202212458,        50],
       [202212396,        95],
       [202212482,        50],
       [202212452,        65],
       [202212387,        70],
       [202212354,        90]])

df.iloc[:,range(2)]

	att	rep
202212380	65	55
202212370	95	100
202212363	65	90
202212488	55	80
202212312	80	30
202212377	75	40
202212463	65	45
202212471	60	60
202212400	95	65
202212469	90	80
202212318	55	75
202212432	95	95
202212443	95	55
202212367	50	80
202212458	50	55
202212396	95	30
202212482	50	50
202212452	65	55
202212387	70	70
202212354	90	90

첫번째 행을 선택

- 방법1

test_ndarray[0,:]

array([202212380,        65,        55,        50,        40])

test_ndarray[0]

array([202212380,        65,        55,        50,        40])

df.iloc[0]

att    65
rep    55
mid    50
fin    40
Name: 202212380, dtype: int64

- 방법2

df.iloc[[0]]  # 데이터프레임처럼

	att	rep	mid	fin
202212380	65	55	50	40

- 방법3

df.iloc[0,:]

att    65
rep    55
mid    50
fin    40
Name: 202212380, dtype: int64

- 방법4

df.iloc[[0],:]

	att	rep	mid	fin
202212380	65	55	50	40

- 방법5

df.loc['202212380']

att    65
rep    55
mid    50
fin    40
Name: 202212380, dtype: int64

- 방법6

df.loc[['202212380']]  # 데이터프레임처럼

	att	rep	mid	fin
202212380	65	55	50	40

- 방법7

df.loc['202212380',:]

att    65
rep    55
mid    50
fin    40
Name: 202212380, dtype: int64

- 방법8

df.loc[['202212380'],:]

	att	rep	mid	fin
202212380	65	55	50	40

- 방법9

len(df)

20

[True]+[False]*19

[True,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False,
 False]

_lst = [True]+[False]*19

df.iloc[_lst]

	att	rep	mid	fin
202212380	65	55	50	40

df.iloc[_lst,:]

	att	rep	mid	fin
202212380	65	55	50	40

df.loc[_lst]

	att	rep	mid	fin
202212380	65	55	50	40

df.loc[_lst,:]

	att	rep	mid	fin
202212380	65	55	50	40

여러개의 행을 선택

- 방법1

df.iloc[[0,2]]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30

df.iloc[[0,2],:]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30

- 방법2

df.loc[['202212380','202212363']]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30

df.loc[['202212380','202212363'],:]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30

- 그 밖의 방법들

df.iloc[::4]  # 스트라이딩

	att	rep	mid	fin
202212380	65	55	50	40
202212312	80	30	30	100
202212400	95	65	20	10
202212443	95	55	15	35
202212482	50	50	45	10

df.iloc[:5]

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100

df.loc[:'202212312']

	att	rep	mid	fin
202212380	65	55	50	40
202212370	95	100	50	80
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100

df.att < 80

202212380     True
202212370    False
202212363     True
202212488     True
202212312    False
202212377     True
202212463     True
202212471     True
202212400    False
202212469    False
202212318     True
202212432    False
202212443    False
202212367     True
202212458     True
202212396    False
202212482     True
202212452     True
202212387     True
202212354    False
Name: att, dtype: bool

df.loc[df.att<80]

	att	rep	mid	fin
202212380	65	55	50	40
202212363	65	90	60	30
202212488	55	80	75	80
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212318	55	75	35	25
202212367	50	80	40	30
202212458	50	55	15	85
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35

df.loc[list(df.att<80),'rep':]  # 리스트로 바꿔주는게 컴퓨터에게 좀 더 명확한 전달

	rep	mid	fin
202212380	55	50	40
202212363	90	60	30
202212488	80	75	80
202212377	40	100	15
202212463	45	45	90
202212471	60	25	0
202212318	75	35	25
202212367	80	40	30
202212458	55	15	85
202212482	50	45	10
202212452	55	15	45
202212387	70	40	35

df.loc[df.att<80,'rep':] # 하지만 리스트화 안해도 되긴 한다..

	rep	mid	fin
202212380	55	50	40
202212363	90	60	30
202212488	80	75	80
202212377	40	100	15
202212463	45	45	90
202212471	60	25	0
202212318	75	35	25
202212367	80	40	30
202212458	55	15	85
202212482	50	45	10
202212452	55	15	45
202212387	70	40	35

df.iloc[list(df.att<80),1:]

	rep	mid	fin
202212380	55	50	40
202212363	90	60	30
202212488	80	75	80
202212377	40	100	15
202212463	45	45	90
202212471	60	25	0
202212318	75	35	25
202212367	80	40	30
202212458	55	15	85
202212482	50	45	10
202212452	55	15	45
202212387	70	40	35

- 아래는 에러가 난다.

df.iloc[df.att<80,1:]

ValueError: Location based indexing can only have [integer, integer slice (START point is INCLUDED, END point is EXCLUDED), listlike of integers, boolean array] types

query (중요!!!)

- 예제1

df.query('att==90 and mid>30')

	att	rep	mid	fin
202212469	90	80	80	20
202212354	90	90	80	90

- 예제2

df.query('att<rep and mid<fin')

	att	rep	mid	fin
202212370	95	100	50	80
202212488	55	80	75	80
202212458	50	55	15	85

- 예제3

df.query('att<rep<80')

	att	rep	mid	fin
202212318	55	75	35	25
202212458	50	55	15	85

- 예제4

df.query('50<att<=90 and mid<fin')

	att	rep	mid	fin
202212488	55	80	75	80
202212312	80	30	30	100
202212463	65	45	45	90
202212452	65	55	15	45
202212354	90	90	80	90

- 예제5

df.query('(mid+fin)/2 >= 60')

	att	rep	mid	fin
202212370	95	100	50	80
202212488	55	80	75	80
202212312	80	30	30	100
202212463	65	45	45	90
202212396	95	30	30	95
202212354	90	90	80	90

- 예제6

_mean = df.att.mean()
_mean

73.0

df.query('att>=73')

	att	rep	mid	fin
202212370	95	100	50	80
202212312	80	30	30	100
202212377	75	40	100	15
202212400	95	65	20	10
202212469	90	80	80	20
202212432	95	95	45	0
202212443	95	55	15	35
202212396	95	30	30	95
202212354	90	90	80	90

df.query('att>=_mean')  # keyError 가 난다!

UndefinedVariableError: name '_mean' is not defined

df.query('att>=@_mean')   # 앞에 @ 골뱅이를 붙여주면 에러 안난다.

	att	rep	mid	fin
202212370	95	100	50	80
202212312	80	30	30	100
202212377	75	40	100	15
202212400	95	65	20	10
202212469	90	80	80	20
202212432	95	95	45	0
202212443	95	55	15	35
202212396	95	30	30	95
202212354	90	90	80	90

- 예제7

df.query('index <= "202212354"')

	att	rep	mid	fin
202212312	80	30	30	100
202212318	55	75	35	25
202212354	90	90	80	90

df.query('index <= "202212354" or index=="202212387"')  # 밖에를 큰따옴표 하고 안쪽을 작은따옴표 해도 된다.

	att	rep	mid	fin
202212312	80	30	30	100
202212318	55	75	35	25
202212387	70	70	40	35
202212354	90	90	80	90

사실 이 기능은 시계열자료에서 꽃핀다.

- 예제8

pd.date_range('20230101',periods=10)

DatetimeIndex(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04',
               '2023-01-05', '2023-01-06', '2023-01-07', '2023-01-08',
               '2023-01-09', '2023-01-10'],
              dtype='datetime64[ns]', freq='D')

_df=pd.DataFrame(np.random.normal(size=(10,4)),columns=list('ABCD'), index=pd.date_range('20230101',periods=10))
_df

	A	B	C	D
2023-01-01	-0.259429	0.369731	-0.279944	0.099409
2023-01-02	-0.932515	-0.311629	0.828348	-0.225257
2023-01-03	-0.011607	0.927334	-0.753145	1.013249
2023-01-04	-1.050379	-0.323094	0.813898	1.035724
2023-01-05	-0.921175	0.513109	-0.905361	0.893707
2023-01-06	-1.521594	0.856883	-0.401441	-1.111551
2023-01-07	0.958028	-0.015302	0.891259	-0.826834
2023-01-08	1.822226	-1.258543	-0.705506	-0.519831
2023-01-09	-0.593394	-1.399224	-1.616172	-0.626952
2023-01-10	-0.083539	0.528519	0.051522	0.126757

_df.query("'2023-01-02' < index <= '2023-01-09'")

	A	B	C	D
2023-01-03	-0.011607	0.927334	-0.753145	1.013249
2023-01-04	-1.050379	-0.323094	0.813898	1.035724
2023-01-05	-0.921175	0.513109	-0.905361	0.893707
2023-01-06	-1.521594	0.856883	-0.401441	-1.111551
2023-01-07	0.958028	-0.015302	0.891259	-0.826834
2023-01-08	1.822226	-1.258543	-0.705506	-0.519831
2023-01-09	-0.593394	-1.399224	-1.616172	-0.626952

_df.query("'2023-01-02' < index <= '2023-01-09' and A+B<C")

	A	B	C	D
2023-01-04	-1.050379	-0.323094	0.813898	1.035724
2023-01-06	-1.521594	0.856883	-0.401441	-1.111551
2023-01-09	-0.593394	-1.399224	-1.616172	-0.626952

- query가 만능은 아니다.

df.columns = pd.Index(['att score', 'rep score', 'mid score','fin score'])

df.query("att score < 90")  # 변수이름에 띄어쓰기가 들어가면 에러가 난다.

SyntaxError: invalid syntax (<unknown>, line 1)

df.att socre

SyntaxError: invalid syntax (<ipython-input-285-4116dfe6888b>, line 1)

df.loc[df["att score"] < 90, :] # 이렇게 하면 됨

	att score	rep score	mid score	fin score
202212380	65	55	50	40
202212363	65	90	60	30
202212488	55	80	75	80
202212312	80	30	30	100
202212377	75	40	100	15
202212463	65	45	45	90
202212471	60	60	25	0
202212318	55	75	35	25
202212367	50	80	40	30
202212458	50	55	15	85
202212482	50	50	45	10
202212452	65	55	15	45
202212387	70	70	40	35

pandas 공부 3단계

전치

ndarray = np.arange(2*3).reshape(2,3)
df=pd.DataFrame(ndarray)
df

	0	1	2
0	0	1	2
1	3	4	5

ndarray.T

array([[0, 3],
       [1, 4],
       [2, 5]])

df.T

	0	1
0	0	3
1	1	4
2	2	5

합

ndarray.sum(axis=0)

array([3, 5, 7])

df.sum(axis=0)

0    3
1    5
2    7
dtype: int64

ndarray.sum(axis=1)

array([ 3, 12])

df.sum(axis=1)

0     3
1    12
dtype: int64

cumsum

df

	0	1	2
0	0	1	2
1	3	4	5

ndarray.cumsum(axis=0) #누적해서 더해짐

array([[0, 1, 2],
       [3, 5, 7]])

df.cumsum(axis=0)

	0	1	2
0	0	1	2
1	3	5	7

ndarray.cumsum(axis=1)

array([[ 0,  1,  3],
       [ 3,  7, 12]])

df.cumsum(axis=1)

	0	1	2
0	0	1	3
1	3	7	12

형태변환

ndarray.tolist()

[[0, 1, 2], [3, 4, 5]]

df.to_numpy()

array([[0, 1, 2],
       [3, 4, 5]])

df.to_numpy().tolist()

[[0, 1, 2], [3, 4, 5]]

df.to_dict()

{0: {0: 0, 1: 3}, 1: {0: 1, 1: 4}, 2: {0: 2, 1: 5}}

pandas 공부 4단계 (생략)

숙제

- 아래의 DF에서 1,3번째 열을 추출하라.

df= pd.DataFrame({'att':[90,90,95],'rep':[80,90,90],'mid':[50,60,70], 'fin':[70,80,50]})
df

	att	rep	mid	fin
0	90	80	50	70
1	90	90	60	80
2	95	90	70	50

df[['att','mid']]

	att	mid
0	90	50
1	90	60
2	95	70

df.iloc[:,[0,2]]

	att	mid
0	90	50
1	90	60
2	95	70

df.iloc[:,::2]

	att	mid
0	90	50
1	90	60
2	95	70

df.loc[:,['att','mid']]

	att	mid
0	90	50
1	90	60
2	95	70