金融风控-贷款违约预测

**摘要：**赛题以金融风控中的个人信贷为背景，要求选手根据贷款申请人的数据信息预测其是否有违约的可能，以此判断是否通过此项贷款，这是一个典型的分类问题。赛题以预测用户贷款是否违约为任务，数据集报名后可见并可下载，该数据来自某信贷平台的贷款记录，总数据量超过120w，包含47列变量信息，其中15列为匿名变量。为了保证比赛的公平性，将会从中抽取80万条作为训练集，20万条作为测试集A，20万条作为测试集B，同时会对employmentTitle、purpose、postCode和title等信息进行脱敏。比赛链接——天池

赛题理解

部分字段表，详细的请看赛题与数据中。

评测标准：

本次比赛的评价方法为AUC评估模型效果（越大越好）。现在来看看什么是AUC。

混淆矩阵（Confuse Matrix）

• 若一个实例是正类，并且被预测为正类，即为真正类TP(True Positive )
• 若一个实例是正类，但是被预测为负类，即为假负类FN(False Negative )
• 若一个实例是负类，但是被预测为正类，即为假正类FP(False Positive )
• 若一个实例是负类，并且被预测为负类，即为真负类TN(True Negative )

ROC空间将假正例率（FPR）定义为 X 轴，真正例率（TPR）定义为 Y 轴。

TPR：在所有实际为正例的样本中，被正确地判断为正例之比率。
$$TPR=\frac{TP}{TP+FN}$$
FPR：在所有实际为负例的样本中，被错误地判断为正例之比率。
$$FPR=\frac{FP}{FP+TN}$$

这是百度百科中关于AUC介绍的一张图。

AUC（Area Under Curve）被定义为ROC曲线下与坐标轴围成的面积，显然这个面积的数值不会大于1。又由于ROC曲线一般都处于y=x这条直线的上方，所以AUC的取值范围在0.5和1之间。AUC越接近1.0，检测方法真实性越高;等于0.5时，则真实性最低，无应用价值。

结果提交：

下面通过sklearn库中的一些函数来简单演示下上面提到的混淆矩阵，ROC，以及AUC

## 混淆矩阵
import numpy as np
from sklearn.metrics import confusion_matrix
y_pred = [0, 1, 0, 1]
y_true = [0, 1, 1, 0]
print('混淆矩阵:\n',confusion_matrix(y_true, y_pred))

# 输出
混淆矩阵:
[[1 1]
[1 1]]

其中[0, 0]表示的是TN，[0, 1]表示的是FP，[1, 0]表示的是FN，[1, 1]表示的是FP。

## ROC曲线
from sklearn.metrics import roc_curve
y_pred = [0, 1, 1, 0, 1, 1, 0, 1, 1, 1]
y_true = [0, 1, 1, 0, 1, 0, 1, 1, 0, 1]
FPR,TPR,thresholds=roc_curve(y_true, y_pred)
plt.title('ROC')
plt.plot(FPR, TPR,'b')
plt.plot([0,1],[0,1],'r--')
plt.ylabel('TPR')
plt.xlabel('FPR')

## AUC
import numpy as np
from sklearn.metrics import roc_auc_score
y_true = np.array([0, 0, 1, 1])
y_scores = np.array([0.1, 0.4, 0.35, 0.8])
print('AUC socre:',roc_auc_score(y_true, y_scores))
# AUC socre: 0.75

AUC的取值就是上面ROC蓝色曲线和下坐标轴构成的面积。

数据分析

首先导入数据分析及可视化会用到的库

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import datetime
import warnings
warnings.filterwarnings('ignore')

读取train.csv和testA.csv的数据，并查看数据的大小、特征、缺失值之类的。

print(train.info(), testA.info())

训练集train的大小为80万，testA的大小为20万行。

train.describe()

特征比较多，只能展示部分了。

missing = train.isnull().sum()
missing = missing[missing > 0]
# 并画出缺失率的图
missing_rate = missing/len(train)
missing_rate.plot.bar()

可以看出，train的缺失值还是不少的，但是占比不是很多。不过缺失值对于xgb，lgb等树模型来说可以直接空缺，树模型会自己优化，要是用别的模型还是要处理的。

one_value_fea = [col for col in train.columns if train[col].nunique() <= 1]
# 'policyCode'
one_value_fea_test = [col for col in testA.columns if testA[col].nunique() <= 1]
# 'policyCode'

然后我们发现policyCode这个记录不管是在训练集中还是测试集中 ，都只有一个值，那么这个就直接可以删去了，对模型丝毫没有影响。

查看特征的数值类型和对象类型：

• 特征一般都是由类别型特征和数值型特征组成
• 类别型特征有时具有非数值关系，有时也具有数值关系。比如‘grade’中的等级A，B，C等，是否只是单纯的分类，还是A优于其他要结合业务判断。
• 数值型特征本是可以直接入模的，但往往风控人员要对其做分箱，转化为WOE编码进而做标准评分卡等操作。从模型效果上来看，特征分箱主要是为了降低变量的复杂性，减少变量噪音对模型的影响，提高自变量和因变量的相关度。从而使模型更加稳定。

数值类型一般是能直接带进模型里面的，而对象模型也就是object是需要先处理一下的。

numerical_fea = list(train.select_dtypes(exclude=['object']).columns)
category_fea = list(train.select_dtypes(include=['object']).columns)

#每个数字特征得分布可视化
f = pd.melt(train, value_vars=numerical_fea)
g = sns.FacetGrid(f, col="variable",  col_wrap=2, sharex=False, sharey=False)
g = g.map(sns.distplot, "value")

特征工程

首先导入在这个过程中会用到的一些库，将数据读取进来。

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import datetime
from tqdm import tqdm
from sklearn.preprocessing import LabelEncoder
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2
from sklearn.preprocessing import MinMaxScaler
import xgboost as xgb
import lightgbm as lgb
from catboost import CatBoostRegressor
import warnings
from sklearn.model_selection import StratifiedKFold, KFold
from sklearn.metrics import accuracy_score, f1_score, roc_auc_score, log_loss
warnings.filterwarnings('ignore')


train =pd.read_csv('data/train.csv')
testA = pd.read_csv('data/testA.csv')

方便后序对特征的操作，在这里就把训练集的数据和测试集的数据拼接起来（后面再拆分开）。

data = pd.concat([train, testA], axis=0, ignore_index=True)

data['employmentLength'].value_counts(dropna=False).sort_index()

1 year        65671
10+ years    328525
2 years       90565
3 years       80163
4 years       59818
5 years       62645
6 years       46582
7 years       44230
8 years       45168
9 years       37866
< 1 year      80226
NaN           58541
Name: employmentLength, dtype: int64

从上面看出employmentLength是object类型，因此我们可以将字符串转成数字类型。

data['employmentLength'].replace(to_replace='10+ years', value='10 years', inplace=True)
data['employmentLength'].replace('< 1 year', '0 years', inplace=True)

def employmentLength_to_int(s):
    if pd.isnull(s):
        return s
    else:
        return np.int8(s.split()[0])
    
data['employmentLength'] = data['employmentLength'].apply(employmentLength_to_int)

10.0    328525
2.0      90565
0.0      80226
3.0      80163
1.0      65671
5.0      62645
4.0      59818
6.0      46582
8.0      45168
7.0      44230
9.0      37866
Name: employmentLength, dtype: int64

现在再来看看earliesCreditLine这列的数据长什么样子。

data['earliesCreditLine']

0         Aug-2001
1         May-2002
2         May-2006
3         May-1999
4         Aug-1977
            ...   
999995    Nov-2005
999996    Oct-2006
999997    Dec-2001
999998    Aug-2005
999999    Aug-2002
Name: earliesCreditLine, Length: 1000000, dtype: object

我们也可以将其中的月份去掉只取其中的年，操作如下

data['earliesCreditLine'] = data['earliesCreditLine'].apply(lambda s: int(s[-4:]))

结果就变成了数值类型的数据了。

0         2001
1         2002
2         2006
3         1999
4         1977
          ... 
999995    2005
999996    2006
999997    2001
999998    2005
999999    2002
Name: earliesCreditLine, Length: 1000000, dtype: int64

对一些类别特征做如下操作

# 类型数在2之上，又不是高维稀疏的
data = pd.get_dummies(data, columns=['grade', 'subGrade', 'homeOwnership', 'verificationStatus', 				'purpose', 'regionCode'], drop_first=True)
# 高维类别特征需要进行转换
for f in ['employmentTitle', 'postCode', 'title']:
    data[f+'_cnts'] = data.groupby([f])['id'].transform('count')
    data[f+'_rank'] = data.groupby([f])['id'].rank(ascending=False).astype(int)
    del data[f]

我们发现在训练集中没有n2.2，n2.3这列，而在测试集中有，因此我们可以选择删掉这两列（也可以不删，如果用的是树模型，树模型会自己优化缺失值）。

del data['n2.2']
del data['n2.3']
del data['id']

# id这列也没什么用，就一起删去了。

现在将训练集和测试集拆分开来。

features = [f for f in data.columns if f not in ['id','issueDate','isDefault']]

train = data[data.isDefault.notnull()].reset_index(drop=True)
test = data[data.isDefault.isnull()].reset_index(drop=True)

x_train = train[features]
x_test = test[features]

y_train = train['isDefault']

建模和融合

def cv_model(clf, train_x, train_y, test_x, clf_name):
    folds = 10
    seed = 2020
    kf = KFold(n_splits=folds, shuffle=True, random_state=seed)

    train = np.zeros(train_x.shape[0])
    test = np.zeros(test_x.shape[0])

    cv_scores = []

    for i, (train_index, valid_index) in enumerate(kf.split(train_x, train_y)):
        print('************************************ {} ************************************'.format(str(i+1)))
        trn_x, trn_y, val_x, val_y = train_x.iloc[train_index], train_y[train_index], train_x.iloc[valid_index], train_y[valid_index]

        if clf_name == "lgb":
            train_matrix = clf.Dataset(trn_x, label=trn_y)
            valid_matrix = clf.Dataset(val_x, label=val_y)

            params = {
                'boosting_type': 'gbdt',
                'objective': 'binary',
                'metric': 'auc',
                'min_child_weight': 5,
                'num_leaves': 2 ** 5,
                'lambda_l2': 10,
                'feature_fraction': 0.8,
                'bagging_fraction': 0.8,
                'bagging_freq': 4,
                'learning_rate': 0.1,
                'seed': 2020,
                'nthread': 28,
                'n_jobs':24,
                'silent': True,
                'verbose': -1,
            }

            model = clf.train(params, train_matrix, 50000, valid_sets=[train_matrix, valid_matrix], verbose_eval=200,early_stopping_rounds=200)
            val_pred = model.predict(val_x, num_iteration=model.best_iteration)
            test_pred = model.predict(test_x, num_iteration=model.best_iteration)
            
            # print(list(sorted(zip(features, model.feature_importance("gain")), key=lambda x: x[1], reverse=True))[:20])
                
        if clf_name == "xgb":
            train_matrix = clf.DMatrix(trn_x , label=trn_y)
            valid_matrix = clf.DMatrix(val_x , label=val_y)
            test_matrix = clf.DMatrix(test_x)
            
            params = {'booster': 'gbtree',
                      'objective': 'binary:logistic',
                      'eval_metric': 'auc',
                      'gamma': 1,
                      'min_child_weight': 1.5,
                      'max_depth': 5,
                      'lambda': 10,
                      'subsample': 0.7,
                      'colsample_bytree': 0.7,
                      'colsample_bylevel': 0.7,
                      'eta': 0.04,
                      'tree_method': 'exact',
                      'seed': 2020,
                      'nthread': 36,
                      "silent": True,
                      }
            
            watchlist = [(train_matrix, 'train'),(valid_matrix, 'eval')]
            
            model = clf.train(params, train_matrix, num_boost_round=50000, evals=watchlist, verbose_eval=200, early_stopping_rounds=200)
            val_pred  = model.predict(valid_matrix, ntree_limit=model.best_ntree_limit)
            test_pred = model.predict(test_matrix , ntree_limit=model.best_ntree_limit)
                 
        if clf_name == "cat":
            params = {'learning_rate': 0.05, 'depth': 5, 'l2_leaf_reg': 10, 'bootstrap_type': 'Bernoulli',
                      'od_type': 'Iter', 'od_wait': 50, 'random_seed': 11, 'allow_writing_files': False}
            
            model = clf(iterations=20000, **params)
            model.fit(trn_x, trn_y, eval_set=(val_x, val_y),
                      cat_features=[], use_best_model=True, verbose=500)
            
            val_pred  = model.predict(val_x)
            test_pred = model.predict(test_x)
            
        train[valid_index] = val_pred
        test = test_pred / kf.n_splits
        cv_scores.append(roc_auc_score(val_y, val_pred))
        
        print(cv_scores)
       
    print("%s_scotrainre_list:" % clf_name, cv_scores)
    print("%s_score_mean:" % clf_name, np.mean(cv_scores))
    print("%s_score_std:" % clf_name, np.std(cv_scores))
    return train, test

def lgb_model(x_train, y_train, x_test):
    lgb_train, lgb_test = cv_model(lgb, x_train, y_train, x_test, "lgb")
    return lgb_train, lgb_test

def xgb_model(x_train, y_train, x_test):
    xgb_train, xgb_test = cv_model(xgb, x_train, y_train, x_test, "xgb")
    return xgb_train, xgb_test

def cat_model(x_train, y_train, x_test):
    cat_train, cat_test = cv_model(CatBoostRegressor, x_train, y_train, x_test, "cat") 
    return cat_train, cat_test

rh_test = 0.5*lgb_test+0.5*cat_test
testA['isDefault'] = rh_test
testA[['id','isDefault']].to_csv('test_sub.csv', index=False)

因为xgboost模型的速度会比其它两个慢很多，我就没用这个去训练了。

baseline.py

# 导入需要的库
import pandas as pd
import os
import gc
import lightgbm as lgb
import xgboost as xgb
from catboost import CatBoostRegressor
from sklearn.linear_model import SGDRegressor, LinearRegression, Ridge
from sklearn.preprocessing import MinMaxScaler
import math
import numpy as np
from tqdm import tqdm
from sklearn.model_selection import StratifiedKFold, KFold
from sklearn.metrics import accuracy_score, f1_score, roc_auc_score, log_loss
import matplotlib.pyplot as plt
import time
import warnings
warnings.filterwarnings('ignore')

# 导入数据
train = pd.read_csv('data/train.csv')
testA = pd.read_csv('data/testA.csv')

data = pd.concat([train, testA], axis=0, ignore_index=True)

data['employmentLength'].replace(to_replace='10+ years', value='10 years', inplace=True)
data['employmentLength'].replace('< 1 year', '0 years', inplace=True)

def employmentLength_to_int(s):
    if pd.isnull(s):
        return s
    else:
        return np.int8(s.split()[0])
    
data['employmentLength'] = data['employmentLength'].apply(employmentLength_to_int)
data['earliesCreditLine'] = data['earliesCreditLine'].apply(lambda s: int(s[-4:]))

data = pd.get_dummies(data, columns=['grade', 'subGrade', 'homeOwnership', 'verificationStatus', 'purpose', 'regionCode'], drop_first=True)


for f in ['employmentTitle', 'postCode', 'title']:
    data[f+'_cnts'] = data.groupby([f])['id'].transform('count')
    data[f+'_rank'] = data.groupby([f])['id'].rank(ascending=False).astype(int)
    del data[f]
    
del data['n2.2']
del data['n2.3']
del data['id']

features = [f for f in data.columns if f not in ['id','issueDate','isDefault']]

train = data[data.isDefault.notnull()].reset_index(drop=True)
test = data[data.isDefault.isnull()].reset_index(drop=True)

x_train = train[features]
x_test = test[features]
y_train = train['isDefault']

def cv_model(clf, train_x, train_y, test_x, clf_name):
    folds = 10
    seed = 2020
    kf = KFold(n_splits=folds, shuffle=True, random_state=seed)

    train = np.zeros(train_x.shape[0])
    test = np.zeros(test_x.shape[0])

    cv_scores = []

    for i, (train_index, valid_index) in enumerate(kf.split(train_x, train_y)):
        print('************************************ {} ************************************'.format(str(i+1)))
        trn_x, trn_y, val_x, val_y = train_x.iloc[train_index], train_y[train_index], train_x.iloc[valid_index], train_y[valid_index]

        if clf_name == "lgb":
            train_matrix = clf.Dataset(trn_x, label=trn_y)
            valid_matrix = clf.Dataset(val_x, label=val_y)

            params = {
                'boosting_type': 'gbdt',
                'objective': 'binary',
                'metric': 'auc',
                'min_child_weight': 5,
                'num_leaves': 2 ** 5,
                'lambda_l2': 10,
                'feature_fraction': 0.8,
                'bagging_fraction': 0.8,
                'bagging_freq': 4,
                'learning_rate': 0.1,
                'seed': 2020,
                'nthread': 28,
                'n_jobs':24,
                'silent': True,
                'verbose': -1,
            }

            model = clf.train(params, train_matrix, 50000, valid_sets=[train_matrix, valid_matrix], verbose_eval=200,early_stopping_rounds=200)
            val_pred = model.predict(val_x, num_iteration=model.best_iteration)
            test_pred = model.predict(test_x, num_iteration=model.best_iteration)
            
            # print(list(sorted(zip(features, model.feature_importance("gain")), key=lambda x: x[1], reverse=True))[:20])
                
        if clf_name == "xgb":
            train_matrix = clf.DMatrix(trn_x , label=trn_y)
            valid_matrix = clf.DMatrix(val_x , label=val_y)
            test_matrix = clf.DMatrix(test_x)
            
            params = {'booster': 'gbtree',
                      'objective': 'binary:logistic',
                      'eval_metric': 'auc',
                      'gamma': 1,
                      'min_child_weight': 1.5,
                      'max_depth': 5,
                      'lambda': 10,
                      'subsample': 0.7,
                      'colsample_bytree': 0.7,
                      'colsample_bylevel': 0.7,
                      'eta': 0.04,
                      'tree_method': 'exact',
                      'seed': 2020,
                      'nthread': 36,
                      "silent": True,
                      }
            
            watchlist = [(train_matrix, 'train'),(valid_matrix, 'eval')]
            
            model = clf.train(params, train_matrix, num_boost_round=50000, evals=watchlist, verbose_eval=200, early_stopping_rounds=200)
            val_pred  = model.predict(valid_matrix, ntree_limit=model.best_ntree_limit)
            test_pred = model.predict(test_matrix , ntree_limit=model.best_ntree_limit)
                 
        if clf_name == "cat":
            params = {'learning_rate': 0.05, 'depth': 5, 'l2_leaf_reg': 10, 'bootstrap_type': 'Bernoulli',
                      'od_type': 'Iter', 'od_wait': 50, 'random_seed': 11, 'allow_writing_files': False}
            
            model = clf(iterations=20000, **params)
            model.fit(trn_x, trn_y, eval_set=(val_x, val_y),
                      cat_features=[], use_best_model=True, verbose=500)
            
            val_pred  = model.predict(val_x)
            test_pred = model.predict(test_x)
            
        train[valid_index] = val_pred
        test = test_pred / kf.n_splits
        cv_scores.append(roc_auc_score(val_y, val_pred))
        
        print(cv_scores)
       
    print("%s_scotrainre_list:" % clf_name, cv_scores)
    print("%s_score_mean:" % clf_name, np.mean(cv_scores))
    print("%s_score_std:" % clf_name, np.std(cv_scores))
    return train, test
    
def lgb_model(x_train, y_train, x_test):
    lgb_train, lgb_test = cv_model(lgb, x_train, y_train, x_test, "lgb")
    return lgb_train, lgb_test

def xgb_model(x_train, y_train, x_test):
    xgb_train, xgb_test = cv_model(xgb, x_train, y_train, x_test, "xgb")
    return xgb_train, xgb_test

def cat_model(x_train, y_train, x_test):
    cat_train, cat_test = cv_model(CatBoostRegressor, x_train, y_train, x_test, "cat") 
    return cat_train, cat_test

lgb_train, lgb_test = lgb_model(x_train, y_train, x_test)
cat_train, cat_test = cat_model(x_train, y_train, x_test)
rh_test = 0.5*lgb_test+0.5*cat_test
testA['isDefault'] = rh_test
testA[['id','isDefault']].to_csv('test_sub.csv', index=False)