LogisticRegression逻辑回归模型.zip

# -*- coding: utf-8 -*- """ Created on Thu Sep 10 17:07:47 2020 @author: 谢树鹏 """ import pandas as pd import matplotlib.pyplot as plt #%matplotlib inline import seaborn as sns import warnings import matplotlib warnings.filterwarnings('ignore') #警告过滤器 pd.set_option('display.max_columns',None) train_data= pd.read_csv(r'E:\文档\我的文档\大数据\题目和模板\Train.csv') train_data.rename(columns={'PAY_0':'PAY_1'},inplace= True) matplotlib.rcParams['font.sans-serif'] = ['KaiTi'] matplotlib.rcParams['font.serif'] = ['KaiTi'] sns.set_style("darkgrid",{"font.sans-serif":['KaiTi', 'Arial']}) matplotlib.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题,或者转换负号为字符串 #对LIMIT_BAL（信用额度）进行Bining化后再进行Factorize化 bins = [1,5,20,35,50,100]#如何分级比较好？ train_data['LIMIT_BAL_bins'] = pd.cut(train_data['LIMIT_BAL'] / 10000,bins) by_bins = train_data.groupby('LIMIT_BAL_bins')['y'].mean() #print(by_bins) train_data['LIMIT_BAL_bins_id'] = pd.factorize(train_data['LIMIT_BAL_bins'])[0] train_data['LIMIT_BAL_bins_id'].unique() #对AGE进行scaling from sklearn import preprocessing scaler = preprocessing.StandardScaler() train_data['AGE'] = scaler.fit_transform(train_data['AGE'].values.reshape(-1,1)) # train_data #对BILL_AMT1到BILL_AMT6,PAY_AMT1到PAY_AMT6进行Scaling train_data['BILL_AMT1'] = scaler.fit_transform(train_data['BILL_AMT1'].values.reshape(-1,1)) train_data['BILL_AMT2'] = scaler.fit_transform(train_data['BILL_AMT2'].values.reshape(-1,1)) train_data['BILL_AMT3'] = scaler.fit_transform(train_data['BILL_AMT3'].values.reshape(-1,1)) train_data['BILL_AMT4'] = scaler.fit_transform(train_data['BILL_AMT4'].values.reshape(-1,1)) train_data['BILL_AMT5'] = scaler.fit_transform(train_data['BILL_AMT5'].values.reshape(-1,1)) train_data['BILL_AMT6'] = scaler.fit_transform(train_data['BILL_AMT6'].values.reshape(-1,1)) train_data['PAY_AMT1'] = scaler.fit_transform(train_data['PAY_AMT1'].values.reshape(-1,1)) train_data['PAY_AMT2'] = scaler.fit_transform(train_data['PAY_AMT2'].values.reshape(-1,1)) train_data['PAY_AMT3'] = scaler.fit_transform(train_data['PAY_AMT3'].values.reshape(-1,1)) train_data['PAY_AMT4'] = scaler.fit_transform(train_data['PAY_AMT4'].values.reshape(-1,1)) train_data['PAY_AMT5'] = scaler.fit_transform(train_data['PAY_AMT5'].values.reshape(-1,1)) train_data['PAY_AMT6'] = scaler.fit_transform(train_data['PAY_AMT6'].values.reshape(-1,1)) #处理测试集 test_data=pd.read_csv(r'E:\文档\我的文档\大数据\题目和模板\Test0904.csv') #先bining化 bins = [0,5,20,35,50,100] test_data['LIMIT_BAL_bins'] = pd.cut(test_data['LIMIT_BAL'] / 10000,bins) #factorize化 test_data['LIMIT_BAL_bins_id'] = pd.factorize(test_data['LIMIT_BAL_bins'])[0] #删掉LIMIT_BAL_bins列 #test_data = test_data.drop(['LIMIT_BAL_bins'],axis=1) #AGE以及BILL_AMT1到BILL_AMT6,PAY_AMT1到PAY_AMT6进行Scaling test_data['AGE'] = scaler.fit_transform(test_data['AGE'].values.reshape(-1,1)) test_data['BILL_AMT1'] = scaler.fit_transform(test_data['BILL_AMT1'].values.reshape(-1,1)) test_data['BILL_AMT2'] = scaler.fit_transform(test_data['BILL_AMT2'].values.reshape(-1,1)) test_data['BILL_AMT3'] = scaler.fit_transform(test_data['BILL_AMT3'].values.reshape(-1,1)) test_data['BILL_AMT4'] = scaler.fit_transform(test_data['BILL_AMT4'].values.reshape(-1,1)) test_data['BILL_AMT5'] = scaler.fit_transform(test_data['BILL_AMT5'].values.reshape(-1,1)) test_data['BILL_AMT6'] = scaler.fit_transform(test_data['BILL_AMT6'].values.reshape(-1,1)) test_data['PAY_AMT1'] = scaler.fit_transform(test_data['PAY_AMT1'].values.reshape(-1,1)) test_data['PAY_AMT2'] = scaler.fit_transform(test_data['PAY_AMT2'].values.reshape(-1,1)) test_data['PAY_AMT3'] = scaler.fit_transform(test_data['PAY_AMT3'].values.reshape(-1,1)) test_data['PAY_AMT4'] = scaler.fit_transform(test_data['PAY_AMT4'].values.reshape(-1,1)) test_data['PAY_AMT5'] = scaler.fit_transform(test_data['PAY_AMT5'].values.reshape(-1,1)) test_data['PAY_AMT6'] = scaler.fit_transform(test_data['PAY_AMT6'].values.reshape(-1,1)) #建模前处理 train_data_X = train_data.drop(['y'],axis=1) train_data_Y = train_data['y'] test_data_X = test_data.drop(['y'],axis=1) test_data_Y = test_data['y'] #解决clf.fit()时的浮点类型错误 train_data_X['LIMIT_BAL_bins']= pd.cut(train_data_X['LIMIT_BAL'] , 4).cat.codes test_data_X['LIMIT_BAL_bins']= pd.cut(test_data_X['LIMIT_BAL'] , 4).cat.codes #LogisticRegression（逻辑回归） #建立模型 from sklearn.linear_model import LogisticRegressionCV clf = LogisticRegressionCV( #penalty='l2',solver='lbfgs',n_jobs=2 ) clf.fit(train_data_X, train_data_Y) #模型预测与评价 from sklearn.metrics import roc_auc_score preds = clf.predict_proba(test_data_X)[:,1] print(roc_auc_score(test_data_Y, preds)) from sklearn.metrics import roc_curve, auc y_score = clf.fit(train_data_X, train_data_Y).predict_proba(test_data_X) fpr, tpr, thresholds = roc_curve(test_data_Y, y_score[:, 1]) roc_auc = auc(fpr, tpr) def drawRoc(roc_auc,fpr,tpr): plt.subplots(figsize=(7, 5.5)) plt.plot(fpr, tpr, color='darkorange', lw=2, label='ROC curve (area = %0.2f)' % roc_auc) plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--') plt.xlim([0.0, 1.0]) plt.ylim([0.0, 1.05]) plt.xlabel('False Positive Rate') plt.ylabel('True Positive Rate') plt.title('LogisticRegression ROC Curve') plt.legend(loc="lower right") plt.show() drawRoc(roc_auc, fpr, tpr) ''' #roc from sklearn.metrics import roc_curve, auc y_score = clf.fit(train_data_X, train_data_Y).predict_proba(test_data_X)[:,1] fpr,tpr,threshold = roc_curve(test_data_Y, y_score) roc_auc = auc(fpr,tpr) plt.figure() lw = 2 plt.figure(figsize=(10,10)) plt.plot(fpr, tpr, color='darkorange', lw=lw, label='ROC curve (area = %0.2f)' % roc_auc) ###假正率为横坐标，真正率为纵坐标做曲线 plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--') plt.xlim([0.0, 1.0]) plt.ylim([0.0, 1.05]) plt.xlabel('False Positive Rate') plt.ylabel('True Positive Rate') plt.title('Receiver operating characteristic example') plt.legend(loc="lower right") plt.show() ''' ''' #输出结果 test_data_ID = test_data['ID'] test_data_ID = pd.DataFrame(test_data_ID) test_proba=pd.DataFrame(preds) test_proba.columns=['proba'] test_target = pd.DataFrame(test_data_Y) test_target.columns = ['target'] test_re = pd.concat([test_data_ID,test_proba,test_target],axis=1) test_re.to_csv("test_LogisticRegression.csv",index=False,sep=',') ''' test_data_ID = test_data['ID'] test_data_ID = pd.DataFrame(test_data_ID) test_proba=pd.DataFrame(preds) test_proba.columns=['proba'] test_re = pd.concat([test_data_ID,test_proba],axis=1) ''' #test_target = pd.DataFrame(test_data_Y) #test_target.columns = ['target'] #test_re = pd.concat([test_data_ID,test_proba,test_target],axis=1) ''' test_target = test_proba test_target[(test_target <= 0.5)] = 0 test_target[(test_target > 0.5)] = 1 test_target.columns=['target'] test_re = pd.concat([test_re,test_target],axis=1) test_re.to_csv("测试test_RandomForest0904.csv",index=False,sep=',')