6种分类模型比较
丹丹
 2019-08-07 00:00:00    1.9k Words  11 Mins  

数据集载入

1
2
3
4
5
6
7
8
9
# -*- coding: utf-8 -*-
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
import seaborn as sns
import warnings
warnings.filterwarnings('ignore')

df = pd.read_csv(u"2019-08-01_金融数据描述_data1.csv",encoding = 'gbk')
1
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4754 entries, 0 to 4753
Data columns (total 90 columns):
Unnamed: 0                                    4754 non-null int64
custid                                        4754 non-null int64
trade_no                                      4754 non-null object
bank_card_no                                  4754 non-null object
low_volume_percent                            4752 non-null float64
middle_volume_percent                         4752 non-null float64
take_amount_in_later_12_month_highest         4754 non-null int64
trans_amount_increase_rate_lately             4751 non-null float64
trans_activity_month                          4752 non-null float64
trans_activity_day                            4752 non-null float64
transd_mcc                                    4752 non-null float64
trans_days_interval_filter                    4746 non-null float64
trans_days_interval                           4752 non-null float64
regional_mobility                             4752 non-null float64
student_feature                               1756 non-null float64
repayment_capability                          4754 non-null int64
is_high_user                                  4754 non-null int64
number_of_trans_from_2011                     4752 non-null float64
first_transaction_time                        4752 non-null float64
historical_trans_amount                       4754 non-null int64
historical_trans_day                          4752 non-null float64
rank_trad_1_month                             4752 non-null float64
trans_amount_3_month                          4754 non-null int64
avg_consume_less_12_valid_month               4752 non-null float64
abs                                           4754 non-null int64
top_trans_count_last_1_month                  4752 non-null float64
avg_price_last_12_month                       4754 non-null int64
avg_price_top_last_12_valid_month             4650 non-null float64
reg_preference_for_trad                       4752 non-null object
trans_top_time_last_1_month                   4746 non-null float64
trans_top_time_last_6_month                   4746 non-null float64
consume_top_time_last_1_month                 4746 non-null float64
consume_top_time_last_6_month                 4746 non-null float64
cross_consume_count_last_1_month              4328 non-null float64
trans_fail_top_count_enum_last_1_month        4738 non-null float64
trans_fail_top_count_enum_last_6_month        4738 non-null float64
trans_fail_top_count_enum_last_12_month       4738 non-null float64
consume_mini_time_last_1_month                4728 non-null float64
max_cumulative_consume_later_1_month          4754 non-null int64
max_consume_count_later_6_month               4746 non-null float64
railway_consume_count_last_12_month           4742 non-null float64
pawns_auctions_trusts_consume_last_1_month    4754 non-null int64
pawns_auctions_trusts_consume_last_6_month    4754 non-null int64
jewelry_consume_count_last_6_month            4742 non-null float64
status                                        4754 non-null int64
source                                        4754 non-null object
first_transaction_day                         4752 non-null float64
trans_day_last_12_month                       4752 non-null float64
id_name                                       4478 non-null object
apply_score                                   4450 non-null float64
apply_credibility                             4450 non-null float64
query_org_count                               4450 non-null float64
query_finance_count                           4450 non-null float64
query_cash_count                              4450 non-null float64
query_sum_count                               4450 non-null float64
latest_query_time                             4450 non-null object
latest_one_month_apply                        4450 non-null float64
latest_three_month_apply                      4450 non-null float64
latest_six_month_apply                        4450 non-null float64
loans_score                                   4457 non-null float64
loans_credibility_behavior                    4457 non-null float64
loans_count                                   4457 non-null float64
loans_settle_count                            4457 non-null float64
loans_overdue_count                           4457 non-null float64
loans_org_count_behavior                      4457 non-null float64
consfin_org_count_behavior                    4457 non-null float64
loans_cash_count                              4457 non-null float64
latest_one_month_loan                         4457 non-null float64
latest_three_month_loan                       4457 non-null float64
latest_six_month_loan                         4457 non-null float64
history_suc_fee                               4457 non-null float64
history_fail_fee                              4457 non-null float64
latest_one_month_suc                          4457 non-null float64
latest_one_month_fail                         4457 non-null float64
loans_long_time                               4457 non-null float64
loans_latest_time                             4457 non-null object
loans_credit_limit                            4457 non-null float64
loans_credibility_limit                       4457 non-null float64
loans_org_count_current                       4457 non-null float64
loans_product_count                           4457 non-null float64
loans_max_limit                               4457 non-null float64
loans_avg_limit                               4457 non-null float64
consfin_credit_limit                          4457 non-null float64
consfin_credibility                           4457 non-null float64
consfin_org_count_current                     4457 non-null float64
consfin_product_count                         4457 non-null float64
consfin_max_limit                             4457 non-null float64
consfin_avg_limit                             4457 non-null float64
latest_query_day                              4450 non-null float64
loans_latest_day                              4457 non-null float64
dtypes: float64(70), int64(13), object(7)
memory usage: 3.3+ MB

特征预处理

删除无用

1
2
delete = ['Unnamed: 0', 'custid', 'trade_no', 'bank_card_no','id_name','latest_query_time','source','loans_latest_time','first_transaction_time', 'student_feature']
df = df.drop(delete,axis=1)

处理分类型特征

1
2
from sklearn.preprocessing import LabelEncoder
df['reg_preference_for_trad'] = LabelEncoder().fit_transform(df['reg_preference_for_trad'].astype(str))

使用众数填充

1
2
3
4
5
from sklearn.preprocessing import Imputer
for i in range(df.shape[1]):
    feature = df.iloc[:,i].values.reshape(-1,1)
    imp_mode = Imputer(strategy='most_frequent')
    df.iloc[:,i] = imp_mode.fit_transform(feature)

数据划分

1
2
X = df[:].drop("status",axis=1)
y = df["status"]
1
2
3
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, shuffle = False)

数据归一化

1
2
3
from sklearn.preprocessing import minmax_scale
X_train = minmax_scale(X_train)
X_test =  minmax_scale(X_test)

建模与预测

使用一般建模方法

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
from sklearn.metrics import *
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import cross_val_score

LR = LogisticRegression()
LR = LR.fit(X_train, y_train)
# LR = cross_val_score(LR, X_train, y_train, cv = 5)

svc = SVC(kernel='linear', probability = True)
svc = svc.fit(X_train, y_train)
# svc = cross_val_score(svc,X_train,y_train, cv = 5)

DT = DecisionTreeClassifier(max_depth = 6)
DT = DT.fit(X_train, y_train)
# DT = cross_val_score(DT, X_train, y_train, cv = 5)

RF = RandomForestClassifier()
RF = RF.fit(X_train, y_train)
# RF = cross_val_score(RF, X_train, y_train, cv = 5)

KNN = KNeighborsClassifier()
KNN = KNN.fit(X_train, y_train)
# LR = cross_val_score(LR, X_train, y_train, cv = 5)

GBDT = GradientBoostingClassifier()
GBDT = GBDT.fit(X_train, y_train)
# LR = cross_val_score(LR, X_train, y_train, cv = 5)

names = ["LR", "SVC", 'DT', "RF", "KNN", "GBDT"]
models = [LR, svc, DT, RF, KNN, GBDT]
evaluates = ['accuracy', 'precision', 'recall', 'f1', 'roc_auc']
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
df_list = []
for name,model in zip(names,models):
    y_train_pred = model.predict(X_train)
    y_test_pred = model.predict(X_test)
    
    #accuracy
    train_accuracy = model.score(X_train,y_train)
    test_accuracy = model.score(X_test,y_test)
    
    
    #precision   
    train_precision = precision_score(y_train,y_train_pred)
    test_precision = precision_score(y_test,y_test_pred)   
    
    #recall
    train_recall = recall_score(y_train,y_train_pred)
    test_recall = recall_score(y_test,y_test_pred)   
    
    #f1
    train_f1 = f1_score(y_train,y_train_pred)
    test_f1 = f1_score(y_test,y_test_pred)   
    
    #auc
    y_train_pred = model.predict_proba(X_train)[:,1]
    y_test_pred = model.predict_proba(X_test)[:,1]
    
    train_auc = roc_auc_score(y_train,y_train_pred)
    test_auc = roc_auc_score(y_test,y_test_pred)
    
    print('{} 训练集： accuracy:{:.3},precision:{:.3}, recall:{:.3}, f1:{:.3}, auc:{:.3}'.format(name,train_accuracy,train_precision,train_recall,train_f1,train_auc))
    print('{} 测试集： accuracy:{:.3},precision:{:.3}, recall:{:.3}, f1:{:.3}, auc:{:.3}'.format(name,test_accuracy,test_precision,test_recall,test_f1,test_auc))
    print('\n')
    df = pd.DataFrame(np.array([train_accuracy,train_precision,train_recall,train_f1,train_auc,test_accuracy,test_precision,test_recall,test_f1,test_auc]).reshape(2,-1),
                  index = ['train','test'],
                  columns = ['Accuracy','Precision','Recall','F1-Score','AUC-Score'])
    df_list.append(df)

pd.concat(df_list,axis=0,keys=names)
LR 训练集： accuracy:0.801,precision:0.742, recall:0.279, f1:0.406, auc:0.8
LR 测试集： accuracy:0.789,precision:0.752, recall:0.365, f1:0.491, auc:0.787


SVC 训练集： accuracy:0.793,precision:0.809, recall:0.196, f1:0.316, auc:0.809
SVC 测试集： accuracy:0.771,precision:0.843, recall:0.222, f1:0.351, auc:0.796


DT 训练集： accuracy:0.829,precision:0.723, recall:0.487, f1:0.582, auc:0.832
DT 测试集： accuracy:0.727,precision:0.516, recall:0.361, f1:0.425, auc:0.707


RF 训练集： accuracy:0.981,precision:0.999, recall:0.924, f1:0.96, auc:0.999
RF 测试集： accuracy:0.722,precision:0.509, recall:0.218, f1:0.305, auc:0.673


KNN 训练集： accuracy:0.816,precision:0.741, recall:0.379, f1:0.501, auc:0.848
KNN 测试集： accuracy:0.708,precision:0.445, recall:0.184, f1:0.261, auc:0.623


GBDT 训练集： accuracy:0.859,precision:0.867, recall:0.497, f1:0.632, auc:0.913
GBDT 测试集： accuracy:0.754,precision:0.573, recall:0.474, f1:0.519, auc:0.771
Accuracy Precision Recall F1-Score AUC-Score
LR train 0.800684 0.742120 0.279396 0.405956 0.799942
test 0.788644 0.751938 0.364662 0.491139 0.786801
SVC train 0.792795 0.808889 0.196332 0.315972 0.809276
test 0.770768 0.842857 0.221805 0.351190 0.796191
DT train 0.829345 0.722756 0.486516 0.581560 0.831990
test 0.726604 0.516129 0.360902 0.424779 0.706676
RF train 0.981331 0.998834 0.924488 0.960224 0.999103
test 0.722397 0.508772 0.218045 0.305263 0.672949
KNN train 0.816198 0.740506 0.378641 0.501071 0.847694
test 0.707676 0.445455 0.184211 0.260638 0.623415
GBDT train 0.858796 0.866541 0.497303 0.631940 0.912630
test 0.753943 0.572727 0.473684 0.518519 0.770979 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
def draw_roc_curve(train_pre_proba,test_pre_proba,train_auc,test_auc,model_name,num):
    fpr,tpr,roc_auc = train_pre_proba
    test_fpr,test_tpr,test_roc_auc = test_pre_proba
    
    plt.figure()
    lw = 2
    plt.plot(fpr, tpr, color='darkorange',
             lw=lw, label='ROC curve (area = %0.2f)' % train_auc)
    plt.plot(test_fpr, test_tpr, color='red',
         lw=lw, label='ROC curve (area = %0.2f)' %test_auc)
    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('Roc example '+ model_name)
    plt.legend(loc="lower right")
    plt.savefig("../img/2019-08-07_5种分类模型比较_{}.png".format(num))
    plt.close()
    
for num,name,model in zip(range(1,7),names,models):
    
    y_train_pred = model.predict_proba(X_train)[:,1]
    y_test_pred = model.predict_proba(X_test)[:,1]


    train_roc = roc_curve(y_train,y_train_pred)
    test_roc = roc_curve(y_test,y_test_pred)

    train_auc = roc_auc_score(y_train,y_train_pred)
    test_auc = roc_auc_score(y_test,y_test_pred)

    draw_roc_curve(train_roc,test_roc,train_auc,test_auc,name,num)

使用k-folds交叉建模

1
2
3
4
5
6
7
8
9
10
11
12
13
14
from sklearn.model_selection import KFold

def run_cv(X,y,clf_class,**kwargs):
    kf = KFold(n_splits = 5, shuffle = False, random_state = 0)
    y_pred = y.copy()
    clf = clf_class(**kwargs)
    
    for train_index , test_index in kf.split(X):
        X_train, X_test = X[train_index], X[test_index]
        y_train, y_test = y[train_index], y[test_index]
        
        clf.fit(X_train,y_train)
        y_pred[test_index] = clf.predict(X_test)
    return y_pred
1
2
3
4
5
6
7
8
9
10
from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier
from sklearn.svm import SVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import cross_val_score
from sklearn.neighbors import KNeighborsClassifier 
from sklearn.linear_model import LogisticRegression

LR_CV_result = run_cv(X_train,y_train,LogisticRegression)
RF_CV_result = run_cv(X_train,y_train,RandomForestClassifier)
KNN_CV_result = run_cv(X_train,y_train,KNeighborsClassifier)
1
2
3
4
5
6
def accuracy(y_true,y_pred):
    return np.mean(y_true == y_pred)

print ("Logistic Regression (L2 is default): " + str(accuracy(y_train, LR_CV_result)))
print ("Random forest: " + str(accuracy(y_train, RF_CV_result)))
print ("K-nearest-neighbors: " + str(accuracy(y_train, KNN_CV_result)))
Logistic Regression (L2 is default): 0.7933210623192216
Random forest: 0.7812253484091507
K-nearest-neighbors: 0.7520378648435446

参考：

  1. DataWhale数据挖掘实战营
  2. 吴裕雄 PYTHON 机器学习——集成学习梯度提升决策树GRADIENTBOOSTINGCLASSIFIER分类模型
  3. 使用5种分类模型进行用户贷款逾期预测
  4. sklearn.model_selection.KFold