꼰대코더

As you can see the most concentrated cluster of racist searches are in the Deep South but rather along a spine of the appalachin running from Georgia and into Southern VermontNow the eastern part of Kentucky shown a tremendous amount of racist beliefs on GoogleAnd that's a trend we're going to see moving forward a lot as we Explore More of AppalachiaMany people in Kentucky aren't even casually r..

Things are so bad in California now that California farmers are moving to MexicoApparently the Hispanics aren't coming up here as much so they thought let's take the farm to themI'm not saying I want to work in the fields cuz I don'tBut farming how we made our countryEven our food supply is getting outsourced now what the hell AmericaWell in the US they pay field workers $10 an hour in Mexico th..

We're in McDow County West Virginia and this is cold country mister or it was coal countryThere probably isn't a place in America that's seen a bigger boom then bust where we are right nowMcDow county is now the poorest place in the countryYou could say McDow county is West Virginia's West VirginiaDriving around this part of appalache land it's kind of like exploring a real live ghost town in pr..

our tiny little foal Prince is finally big enough to go out to pasture with the herd the herd is very excited to meet the new foal  and right here they're surrounding him smelling him but a new mama will very quickly put everybody in their place tell them to back off and they know they better listen to her or else Prince's mama shab will keep him by her side very closely for the first few days a..

세금전 가격 price before taxplastic 발음은 l sound 가 확실히 들리도록... ae sound가 뒤에 있기 때문에 강하게 발음

CSV 읽어 들일때 index 지정df = pd.read_csv(‘xxx.csv’, index_col=’Country’)      Id 를 index 로 변경다시 되돌리기 df.set_index(‘Id’, inplace=True)df.reset_index(inplace=True)

URL Sales data 로딩Data Processing1. missing data 찾기2. missing data를 로우(axis=0) 단위로 자체(inplace=True)에서 삭제하고 확인하기3.  불순 데이터(엉터리 데이터 = 같은 형타입이 아닌) 삭제4. 중복된 data 보기Feature Engineering1. 주문당 구매액( Total  ) 생성2. Order Date 를 월일시로 분리3. Purchse Address 를 주소, 시, 주, 우편번호로 분리4. State 와 ZIP 으로 재 분리5. 불필요한 칼럼들 삭제6. Order Date 순으로 정렬Sales 분석1. 매출액이 가장 큰 달은?2. 각 달의 요일별 주문수는?3. 시간대별 주문수는?4. 가장 많이 팔린 제품은?5. 각 도시별 ..

population_dict = { 'California': 39538223,                               'Texas': 29145505,                                'Florida': 21538187,                                'New York': 20201249,                                'Pennsylvania': 13002700 }population1 = pd.Series(population_dict)population_list = [39538223, 29145505, 21538187, 20201249, 13002700]population2 = pd.Series( population..

1. Population Series population_dict = { 'California': 39538223,                                'Texas': 29145505,                                 'Florida': 21538187,                                 'New York': 20201249,                                 'Pennsylvania': 13002700 } population = pd.Series(population_dict)  혹은 데이터값이 리스트로 있을 경우population_list = [39538223, 29145505, 21538187, 20201249..

ROC ( Receiver Operating Characteristic) AUC( Area Under the Curve) score 는 binary classifiers의 퍼포먼스 측정에 유명한 메트릭스이다. 평가를 위해선 ROC curve의 밑쪽 면적을 측정하면 다양한 결정 쓰레시홀드에서의 classifier의 평가를 보여준다. ROC curve는 Y축엔 True Positive rate(TPR=Recall), X축엔 False Positive rate(FPR) 를 플롯- TPR : 스팸메일에 대해 스팸이라 판단하는 비율- FPR : 정상메일에 대해 스팸이라 판단하는 비율 binary classifier는 0.5 를 기준으로 판단을 한다. 만약 이 기준(결정 쓰레시홀드)을 임으로 0.1, 0.3, 0...