인프런 커뮤니티 질문&답변

import pandas as pd

train = pd.read_csv("data/customer_train.csv")

test = pd.read_csv("data/customer_test.csv")

# 사용자 코딩

# print(train.shape, test.shape)

# print(train.info())

# print(test.isnull().sum())

y_train = train.pop('총구매액')

# print(train.shape, test.shape)

m = train['환불금액'].mean()

train['환불금액'] = train['환불금액'].fillna(m)

test['환불금액'] = test['환불금액'].fillna(m)

# 원핫, 라벨링인코더

# print(train.info())

# print(train.shape, test.shape)

#라벨인코더

# cols = ['주구매상품', '주구매지점']

# from sklearn.preprocessing import LabelEncoder

# le = LabelEncoder()

# for col in cols:

# train[col] = le.fit_transform(train[col])

# test[col] = le.transform(test[col])

#원핫

print(train.shape, test.shape)

data = pd.concat([train,test], axis = 0)

data_oh = pd.get_dummies(data)

train = data_oh.iloc[:len(train)].copy()

test = data_oh.iloc[len(train):].copy()

print(train.shape, test.shape)

# print(train.head())

#스플릿

from sklearn.model_selection import train_test_split

X_tr, X_var, y_tr, y_var = train_test_split(train,

y_train,

test_size = 0.15,

random_state = 0)

#랜포

from sklearn.ensemble import RandomForestRegressor

rf = RandomForestRegressor(random_state=0)

rf.fit(X_tr, y_tr)

pred = rf.predict(X_var)

import lightgbm as lgb

lgbmr = lgb.LGBMRegressor(random_state=0)

lgbmr.fit(X_tr, y_tr)

pred1 = lgbmr.predict(X_var)

from sklearn.metrics import root_mean_squared_error

rmse = root_mean_squared_error(y_var, pred1)

print(rmse)

#라벨 rf 716.2595627489613

#라벨 lgb 667.9290102574973

#원핫 rf - 762.9476701424611

#원핫 lgb - 652.1802049238468

#제출

pred = lgbmr.predict(test)

submit = pd.DataFrame({'pred' : pred})

submit.to_csv('result.csv', index=False)

print(pd.read_csv('result.csv'))

저렇게 4개 비교해서 원핫 lgb로 제출했는데요 옳게 했는지 코드 점검 한번만 부탁드립니다

 그리고 질문이 있는데요 실제 시험 들어가면 import pands as pd랑 트레인 테스트 저거 주나요? 외워야하나요?

저 화면 처럼 실제 시험도 저렇게 구성되어있나요?