In [ ]:
import pandas as pd
import matplotlib as plt
import numpy as np
train=pd.read_csv("train.csv")
In [ ]:
# 移除无用特征
train=train.drop(["ID"],axis=1)
#数据都为RG+数字，现将RG除去，只留数字
train['Region_Code']=train['Region_Code'].apply(lambda x:x[2:]).astype('int64')
In [ ]:
# 可视化各项指标
import seaborn as sns
import matplotlib.pyplot as plt
plt.figure(figsize=(20, 6))
sns.countplot(x = "Age", data = train)
Out[ ]:
<Axes: xlabel='Age', ylabel='count'>
In [ ]:
plt.figure(figsize=(20, 6))
sns.countplot(x = "Vintage", data = train)
Out[ ]:
<Axes: xlabel='Vintage', ylabel='count'>
In [ ]:
plt.figure(figsize=(20, 6))
sns.countplot(x = "Occupation", data = train)
Out[ ]:
<Axes: xlabel='Occupation', ylabel='count'>
In [ ]:
sns.countplot(x = "Credit_Product", data = train)
Out[ ]:
<Axes: xlabel='Credit_Product', ylabel='count'>
In [ ]:
sns.countplot(x = "Is_Active", data = train)
Out[ ]:
<Axes: xlabel='Is_Active', ylabel='count'>
In [ ]:
sns.countplot(x = "Is_Lead", data = train)
# 问题出现了，需要预测的部分Is_Lead=1的样本量太少了，需要欠采样多数类
Out[ ]:
<Axes: xlabel='Is_Lead', ylabel='count'>
In [ ]:
sns.countplot(x = "Channel_Code", data = train)
Out[ ]:
<Axes: xlabel='Channel_Code', ylabel='count'>
In [ ]:
from sklearn.utils import resample
# 欠采样多数类
df_majority = train[train['Is_Lead']==0]
df_minority = train[train['Is_Lead']==1]
df_majority_undersampled = resample(df_majority,replace=True,n_samples=len(df_minority),random_state=0)
# 结合少数类和过采样的多数类
df_undersampled = pd.concat([df_minority, df_majority_undersampled])
df_undersampled['Is_Lead'].value_counts()
train=df_undersampled
In [ ]:
# 将Gender、Credit_Product、Is_Active改为0，1表示
train["Gender"] = train["Gender"].replace({"Male": 1, "Female": 0}).astype("int32")
train["Is_Active"] = train["Is_Active"].replace({"Yes": 1, "No": 0}).astype("int32")
train["Credit_Product"] = train["Credit_Product"].replace({"Yes": 1, "No": 0})
train
Out[ ]:
Gender Age Region_Code Occupation Channel_Code Vintage Credit_Product Avg_Account_Balance Is_Active Is_Lead
4 0 84 279 Other X2 93 0.0 555199 0 1
11 0 28 270 Salaried X1 27 0.0 743781 0 1
23 0 56 254 Self_Employed X2 97 NaN 3360390 1 1
26 1 42 251 Self_Employed X2 15 0.0 327555 1 1
34 1 48 276 Self_Employed X3 33 0.0 1680907 1 1
... ... ... ... ... ... ... ... ... ... ...
116375 1 49 268 Self_Employed X2 92 1.0 1763846 0 0
162838 1 32 252 Salaried X1 26 0.0 1016709 0 0
27157 1 48 282 Self_Employed X3 91 0.0 912775 0 0
84102 1 36 267 Salaried X3 33 0.0 523951 0 0
16567 0 27 254 Other X1 26 0.0 870064 0 0

93154 rows × 10 columns

In [ ]:
# 将Occupation进行one-hot编码
encoded_data = pd.get_dummies(train["Occupation"], prefix="Occupation").replace({True: 1, False: 0})
train = pd.concat([train.iloc[:, :-1], encoded_data, train.iloc[:, -1]], axis=1)
train=train.drop(["Occupation"],axis=1)
train
Out[ ]:
Gender Age Region_Code Channel_Code Vintage Credit_Product Avg_Account_Balance Is_Active Occupation_Entrepreneur Occupation_Other Occupation_Salaried Occupation_Self_Employed Is_Lead
4 0 84 279 X2 93 0.0 555199 0 0 1 0 0 1
11 0 28 270 X1 27 0.0 743781 0 0 0 1 0 1
23 0 56 254 X2 97 NaN 3360390 1 0 0 0 1 1
26 1 42 251 X2 15 0.0 327555 1 0 0 0 1 1
34 1 48 276 X3 33 0.0 1680907 1 0 0 0 1 1
... ... ... ... ... ... ... ... ... ... ... ... ... ...
116375 1 49 268 X2 92 1.0 1763846 0 0 0 0 1 0
162838 1 32 252 X1 26 0.0 1016709 0 0 0 1 0 0
27157 1 48 282 X3 91 0.0 912775 0 0 0 0 1 0
84102 1 36 267 X3 33 0.0 523951 0 0 0 1 0 0
16567 0 27 254 X1 26 0.0 870064 0 0 1 0 0 0

93154 rows × 13 columns

In [ ]:
# 将Channel_Code进行one-hot编码
train = pd.get_dummies(train, columns=['Channel_Code'], prefix=['X']).replace({True: 1, False: 0})
# 获取倒数第五列的列名
column_to_move = train.columns[-5]
# 将倒数第五列保存到变量中
column_data = train[column_to_move]
# 删除倒数第五列
train = train.drop(column_to_move, axis=1)
# 将倒数第五列重新添加到DataFrame的最后
train[column_to_move] = column_data
train
Out[ ]:
Gender Age Region_Code Vintage Credit_Product Avg_Account_Balance Is_Active Occupation_Entrepreneur Occupation_Other Occupation_Salaried Occupation_Self_Employed X_X1 X_X2 X_X3 X_X4 Is_Lead
4 0 84 279 93 0.0 555199 0 0 1 0 0 0 1 0 0 1
11 0 28 270 27 0.0 743781 0 0 0 1 0 1 0 0 0 1
23 0 56 254 97 NaN 3360390 1 0 0 0 1 0 1 0 0 1
26 1 42 251 15 0.0 327555 1 0 0 0 1 0 1 0 0 1
34 1 48 276 33 0.0 1680907 1 0 0 0 1 0 0 1 0 1
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
116375 1 49 268 92 1.0 1763846 0 0 0 0 1 0 1 0 0 0
162838 1 32 252 26 0.0 1016709 0 0 0 1 0 1 0 0 0 0
27157 1 48 282 91 0.0 912775 0 0 0 0 1 0 0 1 0 0
84102 1 36 267 33 0.0 523951 0 0 0 1 0 0 0 1 0 0
16567 0 27 254 26 0.0 870064 0 0 1 0 0 1 0 0 0 0

93154 rows × 16 columns

In [ ]:
#查看各特征值与是否可能成为信用卡客户的相关性
df_corr_lead=train.corr()[u'Is_Lead'].sort_values(ascending=False)
df_corr_lead=df_corr_lead.drop(['Is_Lead'],axis=0)
#对其进行可视化
plt.figure(figsize=(30,12),)
plt.bar(list(df_corr_lead.index),df_corr_lead.values)
for i,j in zip(list(df_corr_lead.index),df_corr_lead.values):
    plt.text(i,j,'%s'%('{:.3f}'.format(j)),ha='center',size=15)
plt.show()
In [ ]:
# 判断是否存在缺失值
pd.isnull(train).any()
# 发现Credit_Product部分缺失，采用预测的方式进行填补
Out[ ]:
Gender                      False
Age                         False
Region_Code                 False
Vintage                     False
Credit_Product               True
Avg_Account_Balance         False
Is_Active                   False
Occupation_Entrepreneur     False
Occupation_Other            False
Occupation_Salaried         False
Occupation_Self_Employed    False
X_X1                        False
X_X2                        False
X_X3                        False
X_X4                        False
Is_Lead                     False
dtype: bool
In [ ]:
# 筛选空值部分
missing_credit_product = train[train["Credit_Product"].isnull()]
# 筛选非空值部分
non_missing_credit_product = train[train["Credit_Product"].notnull()]
In [ ]:
#查看各特征值与是否可能成为信用卡客户的相关性
df_corr_lead=non_missing_credit_product.corr()[u'Credit_Product'].sort_values(ascending=False)
df_corr_lead=df_corr_lead.drop(['Is_Lead','Credit_Product'],axis=0)
#对其进行可视化
plt.figure(figsize=(30,12),)
plt.bar(list(df_corr_lead.index),df_corr_lead.values)
for i,j in zip(list(df_corr_lead.index),df_corr_lead.values):
    plt.text(i,j,'%s'%('{:.3f}'.format(j)),ha='center',size=15)
plt.show()
In [ ]:
# 对数据集进行划分
from sklearn.model_selection import train_test_split
data=non_missing_credit_product.drop(["Is_Lead", "Credit_Product"], axis=1)
x_train,x_test,y_train,y_test=train_test_split(data,non_missing_credit_product["Credit_Product"],test_size=0.2,random_state=22)
In [ ]:
# 训练集与测试集标准化
from sklearn.preprocessing import StandardScaler
transfer=StandardScaler()
x_train_new=transfer.fit_transform(x_train)
x_test_new=transfer.transform(x_test)
In [ ]:
# # 主成分分析降维
from sklearn.decomposition import PCA
# transfer=PCA(n_components=0.95)
# x_train_new=transfer.fit_transform(x_train_new)
# x_test_new=transfer.transform(x_test_new)
In [ ]:
import lightgbm as lgb
from sklearn.metrics import roc_auc_score
# 创建LightGBM分类器
lgb_classifier = lgb.LGBMClassifier(n_estimators=80, random_state=40)
# 训练LightGBM分类器
lgb_classifier.fit(x_train_new, y_train)
# 在测试集上进行预测
y_pred_prob = lgb_classifier.predict_proba(x_test_new)[:, 1]  # 预测样本属于正类的概率
# 计算AUC
auc_score = roc_auc_score(y_test, y_pred_prob)
print("AUC Score:", auc_score)

AUC Score: 0.7671483718923551
In [ ]:
# 至此我们发现，采用lightgbm预估正确率最高，因此，采用此方法来填充空值
data=non_missing_credit_product.drop(["Is_Lead", "Credit_Product"], axis=1)
data_ans=non_missing_credit_product["Credit_Product"]
test=missing_credit_product.drop(["Is_Lead", "Credit_Product"], axis=1)
# 标准化
transfer=StandardScaler()
data_new=transfer.fit_transform(data)
test_new=transfer.transform(test)
# # 主成分分析降维
# transfer=PCA(n_components=0.95)
# data_new=transfer.fit_transform(data_new)
# test_new=transfer.transform(test_new)
In [ ]:
lgb_classifier = lgb.LGBMClassifier(n_estimators=80, random_state=42)
# 训练LightGBM分类器
lgb_classifier.fit(data_new, data_ans)
# 在测试集上进行预测
y_pred_prob = lgb_classifier.predict_proba(test_new)[:,1] # 预测样本属于正类的概率
# 填充 
y_pred_prob_series = pd.Series(y_pred_prob, index=missing_credit_product.index)
missing_credit_product.loc[missing_credit_product["Credit_Product"].isnull(), "Credit_Product"] = y_pred_prob_series
# 拼接回原来的训练集
train=pd.concat([missing_credit_product, non_missing_credit_product])
In [ ]:
# 开始预测目标值
data=train.drop(["Is_Lead"], axis=1)
x_train,x_test,y_train,y_test=train_test_split(data,train["Is_Lead"],test_size=0.25,random_state=42)
In [ ]:
# 训练集与测试集标准化
transfer=StandardScaler()
x_train_new=transfer.fit_transform(x_train)
x_test_new=transfer.transform(x_test)
In [ ]:
from bayes_opt import BayesianOptimization

# Define the function to optimize (AUC score) using Bayesian optimization
def lgb_cv(n_estimators, learning_rate, max_depth, num_leaves, min_child_samples, subsample, colsample_bytree):
    # Convert integer hyperparameters to integers
    n_estimators = int(n_estimators)
    max_depth = int(max_depth)
    num_leaves = int(num_leaves)
    min_child_samples = int(min_child_samples)

    # Create a LightGBM classifier with the given hyperparameters
    lgb_classifier = lgb.LGBMClassifier(
        n_estimators=n_estimators,
        learning_rate=learning_rate,
        max_depth=max_depth,
        num_leaves=num_leaves,
        min_child_samples=min_child_samples,
        subsample=subsample,
        colsample_bytree=colsample_bytree,
        random_state=40
    )
    
    # Train the classifier on the training data
    lgb_classifier.fit(x_train_new, y_train)
    
    # Make predictions on the test data
    y_pred_prob = lgb_classifier.predict_proba(x_test_new)[:, 1]
    
    # Calculate the AUC score
    auc_score = roc_auc_score(y_test, y_pred_prob)
    
    return auc_score

# Define the hyperparameter ranges for Bayesian optimization
pbounds = {
    'n_estimators': (150, 220),
    'learning_rate': (0.02, 0.09),
    'max_depth': (10, 15),
    'num_leaves': (20, 100),
    'min_child_samples': (5, 50),
    'subsample': (0.5, 0.7),
    'colsample_bytree': (0.6, 1.0)
}

# Create the BayesianOptimization object with the function to optimize and the hyperparameter bounds
optimizer = BayesianOptimization(
    f=lgb_cv,
    pbounds=pbounds,
    random_state=42
)

# Perform Bayesian optimization
optimizer.maximize(init_points=10, n_iter=100)

# Get the optimal hyperparameters and corresponding AUC score
best_params = optimizer.max['params']
best_auc = optimizer.max['target']

print("Optimal Hyperparameters:")
print(best_params)
print("Best AUC Score:", best_auc)

|   iter    |  target   | colsam... | learni... | max_depth | min_ch... | n_esti... | num_le... | subsample |
-------------------------------------------------------------------------------------------------------------
| 1         | 0.8734    | 0.7498    | 0.08655   | 13.66     | 31.94     | 160.9     | 32.48     | 0.5116    |
| 2         | 0.8748    | 0.9465    | 0.06208   | 13.54     | 5.926     | 217.9     | 86.6      | 0.5425    |
| 3         | 0.8737    | 0.6727    | 0.03284   | 11.52     | 28.61     | 180.2     | 43.3      | 0.6224    |
| 4         | 0.8738    | 0.6558    | 0.04045   | 11.83     | 25.52     | 205.0     | 35.97     | 0.6028    |
| 5         | 0.8738    | 0.837     | 0.02325   | 13.04     | 12.67     | 154.6     | 95.91     | 0.6931    |
| 6         | 0.8733    | 0.9234    | 0.04132   | 10.49     | 35.79     | 180.8     | 29.76     | 0.599     |
| 7         | 0.874     | 0.6138    | 0.08365   | 11.29     | 34.81     | 171.8     | 61.61     | 0.6093    |
| 8         | 0.8734    | 0.6739    | 0.08787   | 13.88     | 47.28     | 212.6     | 67.83     | 0.6844    |
| 9         | 0.8739    | 0.6354    | 0.03372   | 10.23     | 19.64     | 177.2     | 41.71     | 0.6657    |
| 10        | 0.8738    | 0.7427    | 0.03967   | 12.71     | 11.34     | 206.2     | 25.96     | 0.6974    |
| 11        | 0.8746    | 0.7124    | 0.07015   | 14.43     | 5.014     | 198.1     | 81.36     | 0.542     |
| 12        | 0.8737    | 0.6257    | 0.02763   | 14.48     | 43.25     | 202.6     | 42.31     | 0.5996    |
| 13        | 0.874     | 0.7703    | 0.04386   | 14.52     | 26.26     | 216.0     | 43.78     | 0.6361    |
| 14        | 0.8736    | 0.9994    | 0.02358   | 13.91     | 38.4      | 205.5     | 47.94     | 0.5929    |
| 15        | 0.8745    | 0.8014    | 0.04505   | 13.5      | 5.988     | 218.5     | 88.25     | 0.5013    |
| 16        | 0.875     | 0.8223    | 0.05303   | 14.97     | 6.37      | 217.5     | 83.72     | 0.6993    |
| 17        | 0.8741    | 0.8913    | 0.04304   | 11.42     | 5.653     | 211.1     | 79.25     | 0.6315    |
| 18        | 0.8746    | 0.6002    | 0.06954   | 12.52     | 12.02     | 219.9     | 81.73     | 0.5439    |
| 19        | 0.8744    | 0.9756    | 0.04545   | 13.96     | 10.64     | 210.9     | 87.23     | 0.645     |
| 20        | 0.8737    | 0.905     | 0.06268   | 12.57     | 6.198     | 218.9     | 80.1      | 0.6714    |
| 21        | 0.8747    | 0.6232    | 0.07755   | 12.08     | 11.6      | 217.2     | 85.4      | 0.6758    |
| 22        | 0.8747    | 0.6895    | 0.07514   | 14.99     | 8.508     | 216.5     | 85.29     | 0.5305    |
| 23        | 0.8753    | 0.6996    | 0.03679   | 14.46     | 5.073     | 215.9     | 84.06     | 0.6093    |
| 24        | 0.8742    | 0.8952    | 0.05149   | 13.56     | 5.241     | 211.5     | 83.69     | 0.578     |
| 25        | 0.8741    | 0.8848    | 0.0596    | 10.69     | 15.64     | 219.9     | 85.7      | 0.6331    |
| 26        | 0.874     | 0.6212    | 0.07173   | 10.86     | 12.13     | 213.9     | 83.68     | 0.6485    |
| 27        | 0.8745    | 0.7315    | 0.06917   | 11.47     | 6.007     | 217.3     | 84.75     | 0.6004    |
| 28        | 0.8751    | 0.7557    | 0.04125   | 14.81     | 11.88     | 215.1     | 90.89     | 0.5178    |
| 29        | 0.8747    | 0.7723    | 0.07678   | 14.28     | 10.37     | 216.0     | 93.3      | 0.6447    |
| 30        | 0.8744    | 0.7857    | 0.05494   | 13.44     | 15.64     | 215.3     | 90.32     | 0.6147    |
| 31        | 0.8745    | 0.92      | 0.06649   | 13.9      | 8.149     | 213.1     | 90.06     | 0.6466    |
| 32        | 0.8745    | 0.8119    | 0.07494   | 14.23     | 13.55     | 211.7     | 93.92     | 0.6124    |
| 33        | 0.8744    | 0.6       | 0.09      | 15.0      | 11.77     | 218.2     | 89.36     | 0.7       |
| 34        | 0.8742    | 0.895     | 0.08347   | 12.83     | 11.21     | 215.5     | 90.76     | 0.6061    |
| 35        | 0.8751    | 0.6597    | 0.04563   | 14.73     | 12.94     | 211.1     | 91.38     | 0.646     |
| 36        | 0.8743    | 0.7504    | 0.02135   | 14.94     | 12.33     | 214.2     | 87.38     | 0.5756    |
| 37        | 0.8745    | 0.8825    | 0.04539   | 14.14     | 14.27     | 207.1     | 91.44     | 0.6077    |
| 38        | 0.8752    | 0.6253    | 0.0795    | 14.62     | 11.91     | 214.2     | 93.05     | 0.6013    |
| 39        | 0.8738    | 0.9372    | 0.03267   | 13.28     | 9.903     | 209.2     | 92.21     | 0.5008    |
| 40        | 0.8744    | 0.7772    | 0.08631   | 14.12     | 12.91     | 216.2     | 96.2      | 0.6497    |
| 41        | 0.8748    | 0.8146    | 0.06851   | 14.44     | 11.22     | 213.1     | 90.65     | 0.5733    |
| 42        | 0.8747    | 0.6231    | 0.07336   | 14.2      | 12.8      | 213.4     | 92.05     | 0.6644    |
| 43        | 0.8746    | 0.6494    | 0.06806   | 14.44     | 13.33     | 216.5     | 91.56     | 0.6972    |
| 44        | 0.8748    | 0.9791    | 0.08346   | 13.78     | 6.643     | 215.0     | 84.97     | 0.5984    |
| 45        | 0.8742    | 0.7886    | 0.07413   | 14.48     | 10.95     | 213.0     | 95.36     | 0.6686    |
| 46        | 0.8742    | 0.7596    | 0.07788   | 11.75     | 10.55     | 219.3     | 83.94     | 0.5641    |
| 47        | 0.8746    | 0.6756    | 0.02833   | 13.49     | 15.83     | 210.7     | 89.41     | 0.5021    |
| 48        | 0.8744    | 0.63      | 0.02798   | 14.0      | 15.38     | 209.6     | 92.5      | 0.5348    |
| 49        | 0.875     | 0.8137    | 0.07083   | 14.65     | 6.137     | 215.1     | 87.86     | 0.5007    |
| 50        | 0.8744    | 0.9191    | 0.05873   | 14.79     | 5.083     | 214.4     | 81.14     | 0.5918    |
| 51        | 0.8745    | 0.9163    | 0.06364   | 13.57     | 14.65     | 219.8     | 78.71     | 0.6571    |
| 52        | 0.8744    | 0.9472    | 0.07874   | 14.79     | 14.37     | 217.7     | 81.24     | 0.5633    |
| 53        | 0.8742    | 0.7956    | 0.06696   | 11.6      | 5.409     | 214.1     | 89.53     | 0.5687    |
| 54        | 0.8743    | 0.8097    | 0.06567   | 13.67     | 5.06      | 216.8     | 86.38     | 0.6274    |
| 55        | 0.8746    | 0.8226    | 0.06416   | 14.35     | 8.553     | 218.1     | 85.89     | 0.6583    |
| 56        | 0.8745    | 0.8056    | 0.04099   | 14.09     | 5.906     | 216.5     | 84.59     | 0.671     |
| 57        | 0.8742    | 0.8053    | 0.02318   | 12.67     | 8.146     | 216.7     | 85.79     | 0.6145    |
| 58        | 0.8751    | 0.8743    | 0.03616   | 15.0      | 13.5      | 211.5     | 90.18     | 0.5       |
| 59        | 0.8742    | 0.9308    | 0.03037   | 13.63     | 6.562     | 213.2     | 86.4      | 0.5806    |
| 60        | 0.8747    | 0.7161    | 0.03201   | 14.29     | 8.196     | 215.1     | 89.02     | 0.6161    |
| 61        | 0.8749    | 0.632     | 0.06484   | 12.5      | 6.597     | 219.9     | 86.28     | 0.5258    |
| 62        | 0.8747    | 0.8629    | 0.08885   | 13.72     | 6.881     | 214.5     | 83.6      | 0.663     |
| 63        | 0.8747    | 0.9029    | 0.03119   | 14.99     | 11.65     | 215.9     | 94.67     | 0.5856    |
| 64        | 0.8748    | 0.6282    | 0.05371   | 14.61     | 5.019     | 215.3     | 82.19     | 0.6261    |
| 65        | 0.8737    | 0.9841    | 0.08831   | 14.56     | 5.695     | 219.4     | 86.62     | 0.6388    |
| 66        | 0.8747    | 0.8082    | 0.06228   | 14.46     | 7.694     | 216.7     | 82.84     | 0.5904    |
| 67        | 0.8742    | 0.7953    | 0.05921   | 14.73     | 7.589     | 218.9     | 84.19     | 0.5339    |
| 68        | 0.8747    | 0.9383    | 0.03091   | 14.6      | 9.09      | 215.2     | 83.98     | 0.5664    |
| 69        | 0.8744    | 0.7433    | 0.06533   | 14.77     | 13.16     | 215.0     | 91.8      | 0.6959    |
| 70        | 0.8743    | 0.6451    | 0.0207    | 13.45     | 11.34     | 214.6     | 92.95     | 0.5836    |
| 71        | 0.8753    | 0.7226    | 0.06007   | 14.66     | 10.97     | 215.0     | 89.98     | 0.6399    |
| 72        | 0.8743    | 0.9938    | 0.07436   | 14.68     | 9.313     | 216.3     | 87.29     | 0.5776    |
| 73        | 0.874     | 0.9748    | 0.07583   | 13.9      | 14.43     | 212.6     | 90.3      | 0.5152    |
| 74        | 0.8744    | 0.7307    | 0.02372   | 14.84     | 11.83     | 214.3     | 90.69     | 0.5307    |
| 75        | 0.8744    | 0.961     | 0.08507   | 14.93     | 11.63     | 211.5     | 90.48     | 0.5398    |
| 76        | 0.8742    | 0.9387    | 0.04062   | 13.67     | 10.42     | 212.9     | 90.44     | 0.5167    |
| 77        | 0.8747    | 0.9462    | 0.06496   | 14.98     | 9.483     | 214.0     | 90.48     | 0.6503    |
| 78        | 0.8748    | 0.6325    | 0.07283   | 13.27     | 12.67     | 210.6     | 90.98     | 0.6301    |
| 79        | 0.8744    | 0.8623    | 0.04145   | 14.69     | 9.27      | 217.2     | 84.96     | 0.6807    |
| 80        | 0.8748    | 0.8571    | 0.05822   | 13.84     | 12.55     | 212.1     | 92.87     | 0.6403    |
| 81        | 0.8744    | 0.8831    | 0.07051   | 14.74     | 10.45     | 215.1     | 89.62     | 0.5941    |
| 82        | 0.8744    | 0.66      | 0.05258   | 14.17     | 35.36     | 186.6     | 76.76     | 0.6016    |
| 83        | 0.8744    | 0.8986    | 0.06332   | 13.63     | 7.139     | 216.9     | 82.52     | 0.6816    |
| 84        | 0.8747    | 0.6495    | 0.04659   | 11.62     | 7.195     | 178.1     | 61.46     | 0.6958    |
| 85        | 0.8747    | 0.7534    | 0.08182   | 13.27     | 5.733     | 215.1     | 87.0      | 0.6505    |
| 86        | 0.8751    | 0.668     | 0.05366   | 13.99     | 6.288     | 214.6     | 88.31     | 0.5067    |
| 87        | 0.8734    | 0.9939    | 0.08022   | 13.06     | 5.231     | 214.4     | 87.75     | 0.58      |
| 88        | 0.8748    | 0.9781    | 0.0565    | 14.86     | 12.92     | 211.7     | 90.21     | 0.6131    |
| 89        | 0.8747    | 0.7613    | 0.07047   | 14.14     | 11.54     | 213.5     | 91.76     | 0.584     |
| 90        | 0.8741    | 0.8456    | 0.02431   | 13.99     | 12.98     | 211.3     | 92.18     | 0.6626    |
| 91        | 0.8739    | 0.9985    | 0.0696    | 12.48     | 12.17     | 216.5     | 85.21     | 0.6277    |
| 92        | 0.8745    | 0.6555    | 0.08594   | 13.62     | 8.912     | 215.2     | 82.78     | 0.5465    |
| 93        | 0.8731    | 0.9326    | 0.08337   | 13.47     | 48.39     | 167.0     | 52.2      | 0.6685    |
| 94        | 0.8735    | 0.7645    | 0.07311   | 10.61     | 45.97     | 191.9     | 33.88     | 0.5132    |
| 95        | 0.874     | 0.9073    | 0.05692   | 13.63     | 32.84     | 197.7     | 64.32     | 0.6277    |
| 96        | 0.8739    | 0.6939    | 0.06224   | 12.14     | 24.31     | 193.0     | 94.21     | 0.5147    |
| 97        | 0.8745    | 0.7277    | 0.03087   | 13.6      | 5.01      | 217.1     | 84.45     | 0.5553    |
| 98        | 0.8746    | 0.6408    | 0.02641   | 14.31     | 10.95     | 215.6     | 94.29     | 0.5391    |
| 99        | 0.8751    | 0.6559    | 0.05133   | 14.74     | 13.15     | 212.1     | 92.64     | 0.5069    |
| 100       | 0.8756    | 0.6109    | 0.08053   | 13.71     | 7.263     | 215.9     | 87.0      | 0.6203    |
| 101       | 0.8746    | 0.9861    | 0.05746   | 14.85     | 13.42     | 211.9     | 90.86     | 0.5079    |
| 102       | 0.8743    | 0.8844    | 0.0206    | 14.27     | 5.322     | 215.4     | 82.88     | 0.5666    |
| 103       | 0.8742    | 0.8053    | 0.08119   | 14.35     | 6.977     | 215.7     | 84.78     | 0.642     |
| 104       | 0.8748    | 0.8164    | 0.06701   | 15.0      | 8.916     | 216.3     | 85.07     | 0.5066    |
| 105       | 0.875     | 0.6301    | 0.0881    | 12.55     | 7.67      | 216.0     | 88.0      | 0.6923    |
| 106       | 0.8749    | 0.6827    | 0.07945   | 14.03     | 11.88     | 211.2     | 89.78     | 0.6455    |
| 107       | 0.8743    | 0.64      | 0.04904   | 11.5      | 48.04     | 183.8     | 53.34     | 0.63      |
| 108       | 0.8735    | 0.7001    | 0.0673    | 14.01     | 46.95     | 179.4     | 39.14     | 0.6673    |
| 109       | 0.8742    | 0.8836    | 0.05713   | 13.81     | 12.01     | 215.5     | 90.47     | 0.5395    |
| 110       | 0.8732    | 0.9107    | 0.02603   | 10.9      | 46.57     | 171.9     | 82.75     | 0.5198    |
=============================================================================================================
Optimal Hyperparameters:
{'colsample_bytree': 0.610865775300284, 'learning_rate': 0.0805284415086373, 'max_depth': 13.707039129713081, 'min_child_samples': 7.262616569392997, 'n_estimators': 215.9489533887009, 'num_leaves': 87.00033587605007, 'subsample': 0.620344288616325}
Best AUC Score: 0.8755665739616039