인프런 커뮤니티 질문&답변

모의고사 1회 입니다

아래 코드에 틀린건 없을까요?

최종적으로 제출할때 print는 여기에 1개만 있어야하는거죠?

print(roc_auc_score(y_val, pred[:,1])) 평가지표에 print 하면 안되는거맞죵? 확인부탁드립니다.

#기출1회

import pandas as pd

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

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

#***********************데이터확인

# print(train.shape, test.shape)

# print(train.head()) #target=성별

# print(test.head())

#문자형2개

# print(train.info())

#결측치 있음

# print(train.isnull().sum())

# 환불금액       2295

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

# 환불금액       1611

#***********************전처리 *결합it인

#결측치제거/있음

train['환불금액']=train['환불금액'].fillna(0)

test['환불금액']=test['환불금액'].fillna(0)

#train합치기/없음

# pd.concat([X_train, y_train['성별']],axis=1)

#id없애기/있음

train= train.drop('회원ID',axis=1)

test_id= test.pop('회원ID')

#t타켓

target=train.pop('성별')

#인코딩

from sklearn.preprocessing import LabelEncoder

# from sklearn import preprocessing

# print(dir(preprocessing))

# print(help(preprocessing.LabelEncoder))

cols= train.select_dtypes(include='object').columns

for col in cols :

      le= LabelEncoder()

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

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

#***********************분리

from sklearn.model_selection import train_test_split

from sklearn import model_selection

# print(dir(model_selection))

# print(help(model_selection.train_test_split))

X_tr, X_val, y_tr, y_val = train_test_split(

                                                                                                        train,

                                                                                                                             target,

                                                                                                                                      test_size=0.2,

                                                                                                                                      random_state=2022

)

#***********************모델

from sklearn.ensemble import RandomForestClassifier

# model= RandomForestClassifier(random_state=0)

model= RandomForestClassifier(random_state=0, max_depth=7, n_estimators=1000)

model.fit(X_tr, y_tr)

pred= model.predict_proba(X_val)

#***********************평가

from sklearn.metrics import roc_auc_score

# from sklearn import metrics

# print(dir(metrics))

# print(help(metrics.roc_auc_score))

print(roc_auc_score(y_val, pred[:,1]))

# 0.6186558526810393 (random_state=0)

# 0.6641618297401879 (random_state=0, max_depth=7, n_estimators=1000)

#***********************예측

pred= model.predict_proba(test)[:,1]

result= pd.DataFrame({

                                                                   'pred':pred

})

#***********************저장

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

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