模型调优

k折交叉验证

第一步，不重复抽样将原始数据随机分为 k 份。
第二步，每一次挑选其中 1 份作为测试集，剩余 k-1 份作为训练集用于模型训练。
第三步，重复第二步 k 次，这样每个子集都有一次机会作为测试集，其余机会作为训练集。
在每个训练集上训练后得到一个模型，
用这个模型在相应的测试集上测试，计算并保存模型的评估指标，
第四步，计算 k 组测试结果的平均值作为模型精度的估计，并作为当前 k 折交叉验证下模型的性能指标。

在这里我们采用5折交叉验证

网格搜索

GridSearchCV，它存在的意义就是自动调参，只要把参数输进去，就能给出最优化的结果和参数。但是这个方法适合于小数据集，一旦数据的量级上去了，很难得出结果。

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from xgboost import XGBClassifier
from sklearn.metrics import roc_auc_score
from sklearn.metrics import accuracy_score
from sklearn.metrics import precision_score
from sklearn.metrics import recall_score
from sklearn.metrics import f1_score
from sklearn.metrics import roc_curve,auc
from sklearn.model_selection import train_test_split,GridSearchCV
from sklearn.ensemble import RandomForestClassifier
from sklearn.ensemble import GradientBoostingClassifier
from lightgbm import LGBMClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.tree import DecisionTreeClassifier
from sklearn import svm

data_all = pd.read_csv('D:\data_all.csv',encoding ='gbk')

X = data_all.drop(['status'],axis = 1)
y = data_all['status']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3,random_state=2018)
#数据标准化
scaler = StandardScaler()
scaler.fit(X_train)
X_train = scaler.transform(X_train)
X_test = scaler.transform(X_test)
#LR
lr = LogisticRegression(random_state = 2018)
param = {'C':[1e-3,0.01,0.1,1,10,100,1e3], 'penalty':['l1', 'l2']}
grid = GridSearchCV(estimator=lr, param_grid=param, scoring='roc_auc', cv=5)
grid.fit(X_train,y_train)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#DecisionTree
dt = DecisionTreeClassifier(random_state = 2018)
param = {'criterion':['gini','entropy'],'splitter':['best','random'],'max_depth':[2,4,6,8],'max_features':['sqrt','log2',None]}
grid = GridSearchCV(estimator = dt, param_grid=param, scoring='roc_auc', cv=5)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#SVM
svc = svm.SVC(random_state = 2018)
param = {'C':[1e-2, 1e-1, 1, 10],'kernel':['linear','poly','rbf','sigmoid']}
grid = GridSearchCV(estimator = svc, param_grid=param, scoring='roc_auc', cv=5)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#RandomForest
rft = RandomForestClassifier()
param = {'n_estimators':[10,20,50,100],'criterion':['gini','entropy'],'max_depth':[2,4,6,8,10,None],'max_features':['sqrt','log2',None]}
grid = GridSearchCV(estimator = rft, param_grid=param, scoring='roc_auc', cv=5)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#GBDT
gb = GradientBoostingClassifier()
param = {'max_features':['sqrt','log2',None],'learning_rate':[0.01,0.1,0.5,1],'n_estimators':range(20,200,20),'subsample':[0.2,0.5,0.7,1.0]}
grid = GridSearchCV(estimator = gb, param_grid=param, scoring='roc_auc', cv=5)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#XGBoost
xgb_c = XGBClassifier()
param = {'n_estimators':range(20,200,20),'max_depth':[2,6,10],'reg_lambda':[0.2,0.5,1]}
grid = GridSearchCV(estimator = xgb_c, param_grid=param, scoring='roc_auc', cv=5)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))
#LightGBM
lgbm_c = LGBMClassifier()
param = {'learning_rate': [0.2,0.5,0.7], 'max_depth': range(1,10,2), 'n_estimators':range(20,100,10)}
grid = GridSearchCV(estimator = lgbm_c, param_grid=param, scoring='roc_auc', cv=5)
grid.fit(X_train,y_train)
print(grid.best_params_)
print( grid.best_score_)
print(grid.score(X_test,y_test))