客户贷款逾期预测[3]-xgboost和lightgbm

任务

      根据客户贷款数据预测客户是否会逾期，1表示会，0表示不会。

实现

# -*- coding: utf-8 -*-
"""
Created on Thu Nov 15 13:02:11 2018

@author: keepi
"""

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import f1_score
import xgboost as xgb
import warnings
warnings.filterwarnings('ignore')
pd.set_option('display.max_row',1000)

#导入数据
data = pd.read_csv('data.csv',encoding='gb18030')
print("data.shape:",data.shape)
#数据处理
miss_rate = data.isnull().sum() / len(data)
print("缺失率：",miss_rate.sort_values(ascending=False))
X_num = data.select_dtypes('number').copy()
X_num.fillna(X_num.mean(),inplace=True)
print("数值型特征的shape：",X_num.shape)
print(X_num.columns)
X_num.drop(['Unnamed: 0','status'],axis=1,inplace=True)

X_str = data.select_dtypes(exclude='number').copy()
X_str.fillna(0,inplace=True)
print("非数值型特征：",X_str.columns)
print(X_str.head())

X_dummy = pd.get_dummies(X_str['reg_preference_for_trad'])
X = pd.concat([X_num,X_dummy],axis=1,sort=False)
y = data['status']

#划分训练集、测试集
X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.3,random_state=1117)

#归一化
ss = StandardScaler()
X_train_std = ss.fit_transform(X_train)
X_test_std = ss.transform(X_test)

print("f1_score:")

#xgboost
xgb_train = xgb.DMatrix(X_train_std,label = y_train)
xgb_test = xgb.DMatrix(X_test_std)
xgb_params = {
        'learning_rate':0.1,
        'n_estimators':1000,
        'max_depth':6,
        'min_child_weight':1,
        'gamma':0,
        'subsample':0.8,
        'colsample_bytree':0.8,
        'objective':'binary:logistic',
        'nthread':4,
        'scale_pos_weight':1,
        'seed':1118
        }
xgb_model = xgb.train(xgb_params, xgb_train, num_boost_round=xgb_params['n_estimators'])

test_xgb_pred_prob = xgb_model.predict(xgb_test)
test_xgb_pred = (test_xgb_pred_prob >= 0.5) + 0

print("xgboost:",f1_score(y_test,test_xgb_pred))

#xgboost sklearn版
from xgboost.sklearn import XGBClassifier
xgbc = XGBClassifier(**xgb_params)
xgbc.fit(X_train_std,y_train)
test_xgbc_pred = xgbc.predict(X_test_std)
print('xgbc:',f1_score(y_test,test_xgbc_pred))

#lightgbm
import lightgbm as lgb

lgb_params = {
        'learning_rate':0.1,
        'n_estimators':50,
        'max_depth':4,
        'min_child_weight':1,
        'gamma':0,
        'subsample':0.8,
        'colsample_bytree':0.8,
        'objective':'binary',
        'nthread':4,
        'scale_pos_weight':1,
        'seed':1117
        }
dtrain = lgb.Dataset(X_train_std,y_train)
lgb_model = lgb.train(lgb_params, dtrain, num_boost_round=lgb_params['n_estimators'])
test_lgb_pred_prob = lgb_model.predict(X_test_std)
test_lgb_pred = (test_lgb_pred_prob >=0.5) + 0
print('lgb:',f1_score(y_test,test_lgb_pred))

#lightgbm sklearn版
from lightgbm.sklearn import LGBMClassifier
lgb_model2 = LGBMClassifier(**lgb_params)
lgb_model2.fit(X_train_std,y_train)
test_lgbsk_pred = lgb_model2.predict(X_test_std)
print('lgb sklearn:',f1_score(y_test,test_lgbsk_pred))

结果：f1_score：

遇到的问题

         1.DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.

         这是一个关于numpy的警告，在空数组上弃用了真值检查，在新版的numpy中已经修复，这里我选择忽略警告。

import warnings
warnings.filterwarnings('ignore')

        现在只是学了如何调用xgboost和lightgbm，不知道如何调整参数，之后需要加强原理理解。

        还有如何选择评分指标也是一个问题。机器学习

参考

DeprecationWarning