day7

1.又前面直接导入数据预处理和特征工程的数据

from sklearn.model_selection import GridSearchCV, KFold, train_test_split
import pandas as pd
import numpy as np
from sklearn.metrics import precision_score,roc_auc_score
from sklearn.metrics import accuracy_score,make_scorer
from sklearn.metrics import precision_score#这个就可以用评分函数了
from sklearn.metrics import recall_score
from sklearn.metrics import f1_score
from sklearn.metrics import roc_curve, auc
from sklearn.preprocessing import StandardScaler
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
data_all=pd.read_csv(r'C:\Users\lxy\Desktop\data_all.csv',encoding='gbk')
x_feature = list(data_all.columns)
x_feature.remove('status')
x_val = data_all[x_feature]
y_val = data_all['status']

2.划分数据集并作归一化

x_train,x_test,y_train,y_test=train_test_split(x_final,y_val,test_size=0.3,random_state=2018)
scaler = StandardScaler()
scaler.fit(x_train)
x_train= scaler.transform(x_train)
x_test= scaler.transform(x_test)

3.发现融合模型后AUC下降飞快，不知道是不是过拟合，尝试了多种方式，还是下降了很多，这点没搞明白

from sklearn import model_selection
from sklearn import datasets  
from sklearn.linear_model import LogisticRegression
from sklearn.naive_bayes import GaussianNB 
from xgboost.sklearn import XGBClassifier
from sklearn.ensemble import RandomForestClassifier
import lightgbm as lgb
from mlxtend.classifier import StackingClassifier
from sklearn.svm import LinearSVC
lr = LogisticRegression(random_state=2018,C=0.1)
lgb_model = lgb.LGBMClassifier(boosting_type='GBDT',random_state=2018,silent=0)
gbdt=GradientBoostingClassifier(random_state=2018,max_depth=3,n_estimators=50)
xgbc = XGBClassifier(random_state=2018,lamba=2,min_child_weight=1,silent=1,max_depth=3,eta=0.1,subsample=0.6)
rf = RandomForestClassifier(n_estimators=500,oob_score=True, random_state=2018)
svm=LinearSVC(random_state=2018,tol=0.01)
sclf = StackingClassifier(classifiers=[lr, gbdt, xgbc,rf,svm], meta_classifier=lgb_model)
sclf1 = StackingClassifier(classifiers=[gbdt, xgbc,svm], meta_classifier=lgb_model)
sclf2 = StackingClassifier(classifiers=[gbdt, xgbc,svm], meta_classifier=lr)
sclf3 = StackingClassifier(classifiers=[svm], meta_classifier=lr)
#sclf.fit(x_train,y_train)
#sclf_predict = sclf.predict(x_test)
#sclf_predict_proba = sclf.predict_proba(X_test)[:, 1]
#scores(y_test, sclf_predict, sclf_predict_proba)
scores(sclf,x_train,y_train,x_test,y_test)
scores(sclf1,x_train,y_train,x_test,y_test)
scores(sclf2,x_train,y_train,x_test,y_test)
scores(sclf3,x_train,y_train,x_test,y_test)

4.结果

#sclf
准确率 0.7883672039243167
精确率 0.6815286624203821
召回率 0.298050139275766
F1-score 0.4147286821705426
AUC 0.6715126287127163
#sclf1
准确率 0.7897687456201822
精确率 0.6398104265402843
召回率 0.37604456824512533
F1-score 0.47368421052631576
AUC 0.6740438483928515
#sclf2
准确率 0.7897687456201822
精确率 0.6398104265402843
召回率 0.37604456824512533
F1-score 0.47368421052631576
AUC 0.6740907952802729
#sclf3
准确率 0.7820602662929222
精确率 0.6643835616438356
召回率 0.27019498607242337
F1-score 0.3841584158415841
AUC 0.6121574181298446