非均衡样本下的信用卡欺诈分析
本文是针对一份kaggle上信用卡数据的建模分析，主要内容包含：
- 理解数据：通过直方图、箱型图等辅助理解数据分布
- 预处理：归一化和分布情况；数据分割
- 随机采样：上采样和下采样，主要是欠采样（下采样）
- 异常检测：如何从数据中找到异常点，并且进行删除
- 数据建模：利用逻辑回归和神经网络进行建模分析
- 模型评价：分类模型的多种评价指标
原notebook地址为：https://www.kaggle.com/code/janiobachmann/credit-fraud-dealing-with-imbalanced-datasets/notebook
所谓的【非均衡】：信用卡数据中欺诈和非欺诈的比例是不均衡的，肯定是非欺诈的比例占据绝大多数。本文提供一种方法：如何处理这种极度不均衡的数据
导入库
导入各种库和包：绘图、特征工程、降维、分类模型、评价指标相关等
代码语言：javascript复制import numpy as np
import pandas as pd
import tensorflow as tf
import plotly_express as px
import plotly.graph_objects as go
# 子图
from plotly.subplots import make_subplots
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
import seaborn as sns
# 降维
from sklearn.manifold import TSNE
from sklearn.decomposition import PCA, TruncatedSVD
import time
plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体
plt.rcParams["axes.unicode_minus"]=False #正常显示负号
# 分类库
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
# 特征工程相关的库
from sklearn.model_selection import train_test_split
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold, StratifiedKFold
from sklearn.pipeline import make_pipeline
from imblearn.pipeline import make_pipeline as imbalanced_make_pipeline
# 上采样
from imblearn.over_sampling import SMOTE
# 欠采样
from imblearn.under_sampling import NearMiss
from imblearn.metrics import classification_report_imbalanced
from sklearn.metrics import precision_score, recall_score, f1_score, roc_auc_score, accuracy_score, classification_report
# 统计数量
from collections import Counter
import collections
import warnings
warnings.filterwarnings("ignore")基本信息
读取数据，查看基本信息
数据的形状如下：
In [3]:
代码语言：javascript复制df.shapeOut[3]:
代码语言：javascript复制(284807, 31)In [4]:
代码语言：javascript复制# 缺失值的最大值
df.isnull().sum().max()Out[4]:
代码语言：javascript复制0结果表明是没有缺失值的。
下面是查看数据中字段的相关类型，我们发现有30个float64类型，1个int64类型
In [5]:
代码语言：javascript复制pd.value_counts(df.dtypes)Out[5]:
代码语言：javascript复制float64 30
int64 1
dtype: int64In [6]:
代码语言：javascript复制columns = df.columns
columnsOut[6]:
代码语言：javascript复制Index(['Time', 'V1', 'V2', 'V3', 'V4', 'V5', 'V6', 'V7', 'V8', 'V9', 'V10',
'V11', 'V12', 'V13', 'V14', 'V15', 'V16', 'V17', 'V18', 'V19', 'V20',
'V21', 'V22', 'V23', 'V24', 'V25', 'V26', 'V27', 'V28', 'Amount',
'Class'],
dtype='object')查看数据的统计信息：
代码语言：javascript复制df.describe()正负样本不均衡
In [8]:
代码语言：javascript复制df["Class"].value_counts(normalize=True)Out[8]:
代码语言：javascript复制0 0.998273 # 不欺诈
1 0.001727 # 欺诈
Name: Class, dtype: float64我们发现属于0类的样本远高于属于1的样本，非常地不均衡。这就是本文重点关注的问题。
In [9]:
代码语言：javascript复制# 绘图
colors = ["red", "blue"]
sns.countplot("Class", data=df, palette=colors)
plt.title("Class Distributions n (0-No Fraud & 1-Fraud)")
plt.show()通过柱状图也能够明显观察到非欺诈-0 和 欺诈-1的比例是极度不均衡的。
查看特征分布
部分特征的分布，发现存在偏态状况：
直方图分布
In [10]:
代码语言：javascript复制fig, ax = plt.subplots(1,2,figsize=(18,6))
amount_val = df["Amount"].values
time_val = df["Time"].values
sns.distplot(amount_val, ax=ax[0], color="r")
ax[0].set_title("Amount", fontsize=14)
ax[0].set_xlim([min(amount_val), max(amount_val)]) # 设置范围
sns.distplot(time_val, ax=ax[1], color="b")
ax[1].set_title("Time", fontsize=14)
ax[1].set_xlim([min(time_val), max(time_val)]) # 设置范围
plt.show()观察两个字段Amount和Time在不同取值下的分布情况，发现：
- Amount的偏态现象严重，极大多数的数据集中在左侧
- Time中，数据主要集中在两个阶段
特征分布箱型图
查看每个特征取值的箱型图：
代码语言：javascript复制# 两个基本参数：设置行、列
fig = make_subplots(rows=5, cols=6)
for i, v in enumerate(parameters):
r = i // 6 1
c = (i 1) % 6
if c == 0:
fig.add_trace(go.Box(y=df[v].tolist(),name=v),
row=r, col=6)
else:
fig.add_trace(go.Box(y=df[v].tolist(),name=v),
row=r, col=c)
fig.update_layout(width=800, height=600)
fig.show()数据预处理
数据缩放和分布
针对Amount和Time字段的归一化操作。原始数据中的其他字段已经进行了归一化的操作。
- StandardScaler：将数据减去均值除以标准差
- RobustScaler：如果数据有离群点，有对数据中心化和数据的缩放鲁棒性更强的参数
In [13]:
代码语言：javascript复制from sklearn.preprocessing import StandardScaler, RobustScaler
# ss = StandardScaler()
rs = RobustScaler()
# 好方法
df['scaled_amount'] = rs.fit_transform(df['Amount'].values.reshape(-1,1))
df['scaled_time'] = rs.fit_transform(df['Time'].values.reshape(-1,1))In [14]:
删除原始字段，使用归一化后的字段和数据
代码语言：javascript复制df['Amount'].values.reshape(-1,1) # 个人添加技巧1：新字段位置
将新生成的字段放在最前面
代码语言：javascript复制# 把两个缩放的字段放在最前面
# 1、单独提出来
scaled_amount = df['scaled_amount']
scaled_time = df['scaled_time']
# 2、删除原字段信息
df.drop(['scaled_amount', 'scaled_time'], axis=1, inplace=True)
# 3、插入
df.insert(0, 'scaled_amount', scaled_amount)
df.insert(1, 'scaled_time', scaled_time)分割数据（基于原DataFrame）
在开始进行随机欠采样之前，我们需要将原始数据进行分割。
尽管我们会对数据进行欠采样和上采样，但是我们希望在测试的时候，仍然是使用原始的数据集（原来的数据顺序）
In [18]:
代码语言：javascript复制from sklearn.model_selection import train_test_split
from sklearn.model_selection import StratifiedShuffleSplit查看Class中0-no fraud和1-fraud的比例：
In [19]:
代码语言：javascript复制df["Class"].value_counts(normalize=True)Out[19]:
代码语言：javascript复制0 0.998273
1 0.001727
Name: Class, dtype: float64生成特征数据集X和标签数据y：
In [20]:
代码语言：javascript复制X = df.drop("Class", axis=1)
y = df["Class"]In [21]:
技巧2：生成随机索引
代码语言：javascript复制sfk = StratifiedKFold(
n_splits=5, # 生成5份
random_state=None,
shuffle=False)
for train_index, test_index in sfk.split(X,y):
# 随机生成的index
print(train_index)
print("------------")
print(test_index)
# 根据随机生成的索引再生成数据
original_X_train = X.iloc[train_index]
original_X_test = X.iloc[test_index]
original_y_train = y.iloc[train_index]
original_y_test = y.iloc[test_index]
[ 30473 30496 31002 ... 284804 284805 284806]
------------
[ 0 1 2 ... 57017 57018 57019]
[ 0 1 2 ... 284804 284805 284806]
------------
[ 30473 30496 31002 ... 113964 113965 113966]
[ 0 1 2 ... 284804 284805 284806]
------------
[ 81609 82400 83053 ... 170946 170947 170948]
[ 0 1 2 ... 284804 284805 284806]
------------
[150654 150660 150661 ... 227866 227867 227868]
[ 0 1 2 ... 227866 227867 227868]
------------
[212516 212644 213092 ... 284804 284805 284806]将生成的数据转成numpy数组：
In [22]:
代码语言：javascript复制original_Xtrain = original_X_train.values
original_Xtest = original_X_test.values
original_ytrain = original_y_train.values
original_ytest = original_y_test.values查看训练集 original_ytrain 和 original_ytest 的唯一值以及每个唯一值所占的比例：
In [23]:
技巧3：数据唯一值及比例
代码语言：javascript复制# 训练集
# 针对的是numpy数组
train_unique_label, train_counts_label = np.unique(original_ytrain, return_counts=True)
# 测试集
test_unique_label, test_counts_label = np.unique(original_ytest, return_counts=True)In [24]:
代码语言：javascript复制print(train_counts_label / len(original_ytrain))
print(test_counts_label / len(original_ytest))
[0.99827076 0.00172924]
[0.99827952 0.00172048]欠采样
原理
欠采样也称之为下采样，主要是通过删除原数据中类别较多的数据，从而和类别少的数据达到平衡，以免造成模型的过拟合。
步骤
- 确定数据不平衡度是多少：通过value_counts()来统计，查看每个类别的数量和占比
- 在本例中一旦我们确定了fraud的数量，我们就需要将no-fraud的数量采样和其相同，形成50%：50%
- 实施采样之后，随机打乱采样的子样本
缺点
下采样会造成数据信息的缺失。比如原数据中no-fraud有284315条数据，但是经过欠采样只有492，大量的数据被放弃了。
取出欺诈的数据，同时从非欺诈中取出相同长度的数据：
代码语言：javascript复制# 欺诈的数据
fraud_df = df[df["Class"] == 1]
# 从非欺诈的数据中取出相同的长度len(fraud_df)
no_fraud_df = df[df["Class"] == 0][:len(fraud_df)]
# 492 492
normal_distributed_df = pd.concat([fraud_df, no_fraud_df])
normal_distributed_df.shape
# 再次随机打乱数据
new_df = normal_distributed_df.sample(frac=1, random_state=123)均匀分布
现在我们发现样本是均匀的：
In [28]:
代码语言：javascript复制# 显示数量
new_df["Class"].value_counts()Out[28]:
代码语言：javascript复制1 492
0 492
Name: Class, dtype: int64In [29]:
代码语言：javascript复制# 显示比例
new_df["Class"].value_counts(normalize=True)Out[29]:
代码语言：javascript复制1 0.5
0 0.5
Name: Class, dtype: float64In [30]:
当我们再次查看数据分布的时候发现：已经是均匀分布了
代码语言：javascript复制sns.countplot("Class",
data=new_df,
palette=colors)
plt.title("Equally Distributed Classes", fontsize=12)
plt.show()相关性分析
相关性分析主要是通过相关系数矩阵来实现的。下面绘制基于原始数据和欠采样数据的相关系数矩阵图：
系数矩阵热力图
In [31]:
代码语言：javascript复制f, (ax1, ax2) = plt.subplots(2,1,figsize=(24, 20))
# 原始数据df
corr = df.corr()
sns.heatmap(corr, cmap="coolwarm_r",annot_kws={"size":20}, ax=ax1)
ax1.set_title("Imbalanced Correlation Matrix", fontsize=14)
# 欠采样数据new_df
new_corr = new_df.corr()
sns.heatmap(new_corr, cmap="coolwarm_r",annot_kws={"size":20}, ax=ax2)
ax2.set_title("SubSample Correlation Matrix", fontsize=14)
plt.show()小结：
- 正相关：特征V2、V4、V11、V19是正相关的。值越大，结果越可能出现fraud
- 负相关：特征V17, V14, V12 和 V10 是负相关的；值越小，结果越可能出现fraud
箱型图
In [32]:
负相关的特征箱型图
代码语言：javascript复制# 负相关的数据
# 生成4个子图
f, axes = plt.subplots(ncols=4, figsize=(20,4))
sns.boxplot(x="Class", # 分类的类别 0-1
y="V17", # 选择某个字段进行绘图
data=new_df, # 绘图的数据
palette=colors,
ax=axes[0]) # 选择某个ax来绘图
axes[0].set_title('V17') # 子图标题
sns.boxplot(x="Class",
y="V14",
data=new_df,
palette=colors,
ax=axes[1])
axes[1].set_title('V14')
sns.boxplot(x="Class",
y="V12",
data=new_df,
palette=colors,
ax=axes[2])
axes[2].set_title('V12')
sns.boxplot(x="Class",
y="V10",
data=new_df,
palette=colors,
ax=axes[3])
axes[3].set_title('V10')
plt.show()正相关特征的箱型图：
代码语言：javascript复制# 正相关
f, axes = plt.subplots(ncols=4, figsize=(20,4))
sns.boxplot(x="Class",
y="V2",
data=new_df,
palette=colors,
ax=axes[0])
axes[0].set_title('V2')
sns.boxplot(x="Class",
y="V4",
data=new_df,
palette=colors,
ax=axes[1])
axes[1].set_title('V4')
sns.boxplot(x="Class",
y="V11",
data=new_df,
palette=colors,
ax=axes[2])
axes[2].set_title('V11')
sns.boxplot(x="Class",
y="V19",
data=new_df,
palette=colors,
ax=axes[3])
axes[3].set_title('V19')
plt.show()异常检测
目的
异常检测的目的主要是：发现数据中的离群点来进行删除。
方法
- IQR：我们通过第75个百分位和第25个百分位之间的差异来计算。我们的目标是创建一个超过第75和 25 个百分位的阈值，以防某些实例超过此阈值，如果超过阈值该实例将被删除。
- 箱型图boxplot：除了很容易看到第 25 和第 75 个百分位数（正方形的两端）之外，还很容易看到极端异常值（超出下限和上限的点）
异常值去除权衡
在通过四分位法删除异常值的时候，我们通过将一个数字（例如1.5）乘以（四分位距）来确定阈值。该阈值越高，检测到的异常值越少，反之检测到的异常值越多。这个比例（例如1.5）我们可以在实际进行控制
特征分布-直方图
In [34]:
代码语言：javascript复制# 查看3个特征的分布
from scipy.stats import norm
f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20,6))
# 生成绘图数据：numpy数组
v14_fraud = new_df["V14"].loc[new_df["Class"] == 1].values
sns.distplot(v14_fraud, # 传入数据
ax=ax1, # 选择子图
fit=norm, # 拟合：正态化
color="#FB8861")
ax1.set_title("V14", fontsize=14)
v12_fraud = new_df["V12"].loc[new_df["Class"] == 1].values
sns.distplot(v12_fraud,
ax=ax2,
fit=norm,
color="#56F9BB")
ax2.set_title("V12", fontsize=14)
v10_fraud = new_df["V10"].loc[new_df["Class"] == 1].values
sns.distplot(v10_fraud,
ax=ax3,
fit=norm,
color="#C5B3F9")
ax2.set_title("V10", fontsize=14)
plt.show(技巧4：删除离群点
删除3个特征下的离群点，以V12为例：
In [35]:
第一步先确定上下分位数的值：
代码语言：javascript复制# 数组
v12_fraud = new_df["V12"].loc[new_df["Class"] == 1]
# 25%和75%分位数
q1, q3 = v12_fraud.quantile(0.25), v12_fraud.quantile(0.75)
iqr = q3 - q1In [36]:
代码语言：javascript复制# 确定上下限
v12_cut_off = iqr * 1.5
v12_lower = q1 - v12_cut_off
v12_upper = q3 v12_cut_off
print(v12_lower)
print(v12_upper)
-17.25930926645337
5.597044719256134In [37]:
找出满足要求的离群点的数据
代码语言：javascript复制# 确定离群点
outliers = [x for x in v12_fraud if x < v12_lower or x > v12_upper]
print(outliers)
print("------------")
print("离群点数量：",len(outliers))
[-17.6316063138707, -17.7691434633638, -18.6837146333443,
-18.5536970096458, -18.0475965708216, -18.4311310279993]
------------
离群点数量： 6对单个列字段的数据执行下面删除离群点的操作：
In [38]:
代码语言：javascript复制# 技巧：如何删除异常值
new_df = new_df.drop(new_df[(new_df["V12"] > v12_upper) | (new_df["V12"] < v12_lower)].index)
new_df对其他的特征执行相同的操作：
可以看到：欠采样之后的数据原本是984，现在变成了978条数据，删除了6个离群点的数据
In [39]:
代码语言：javascript复制# 对V10和V14执行同样的操作
# 数组
v14_fraud = new_df["V14"].loc[new_df["Class"] == 1]
q1, q3 = v14_fraud.quantile(0.25), v14_fraud.quantile(0.75)
iqr = q3 - q1
v14_cut_off = iqr * 1.5
v14_lower = q1 - v14_cut_off
v14_upper = q3 v14_cut_off
outliers = [x for x in v14_fraud if x < v14_lower or x > v14_upper]
new_df = new_df.drop(new_df[(new_df["V14"] > v14_upper) | (new_df["V14"] < v14_lower)].index)In [40]:
代码语言：javascript复制# 对V10和V14执行同样的操作
# 数组
v10_fraud = new_df["V10"].loc[new_df["Class"] == 1]
q1, q3 = v10_fraud.quantile(0.25), v10_fraud.quantile(0.75)
iqr = q3 - q1
v10_cut_off = iqr * 1.5
v10_lower = q1 - v10_cut_off
v10_upper = q3 v10_cut_off
outliers = [x for x in v10_fraud if x < v10_lower or x > v10_upper]
new_df = new_df.drop(new_df[(new_df["V10"] > v10_upper) | (new_df["V10"] < v10_lower)].index)查看删除了异常点后的数据：
In [42]:
代码语言：javascript复制f, (ax1, ax2, ax3) = plt.subplots(1,3,figsize=(20,10))
colors = ['#B3F9C5', '#f9c5b3']
sns.boxplot(x="Class", # 分类字段
y="V14", # y轴的取值
data=new_df, # 数据框
ax=ax1, # 子图
palette=colors)
ax1.set_title("V14", fontsize=14) # 标题
ax1.annotate("Fewer extreme", # 注解
xy=(0.98,-17.5),
xytext=(0,-12),
arrowprops=dict(facecolor="black"),
fontsize=14)
sns.boxplot(x="Class", y="V12", data=new_df, ax=ax2, palette=colors)
ax2.set_title("V12", fontsize=14)
ax2.annotate("Fewer extreme",
xy=(0.98,-17),
xytext=(0,-12),
arrowprops=dict(facecolor="black"),
fontsize=14)
sns.boxplot(x="Class", y="V10", data=new_df, ax=ax3, palette=colors)
ax3.set_title("V10", fontsize=14)
ax3.annotate("Fewer extreme", # 注释名称
xy=(0.98,-16.5), # 位置
xytext=(0,-12), # 注释文本的坐标点，二维元组，默认xy
arrowprops=dict(facecolor="black"), # 箭头颜色
fontsize=14)
plt.show()降维和聚类
理解t-SNE
详细地址：https://www.youtube.com/watch?v=NEaUSP4YerM
欠采样数据降维
对3种不同方法实施欠采样：
In [43]:
代码语言：javascript复制X = new_df.drop("Class", axis=1)
y = new_df["Class"]
# t-SNE降维
t0 = time.time()
X_reduced_tsne = TSNE(n_components=2,
random_state=42).fit_transform(X.values)
t1 = time.time()
print("T-SNE: ", (t1 - t0))
T-SNE: 5.750015020370483In [44]:
代码语言：javascript复制# PCA降维
t0 = time.time()
X_reduced_pca = PCA(n_components=2,
random_state=42).fit_transform(X.values)
t1 = time.time()
print("PCA: ", (t1 - t0))
PCA: 0.02214193344116211In [45]:
代码语言：javascript复制# TruncatedSVD降维
t0 = time.time()
X_reduced_svd = TruncatedSVD(n_components=2,
algorithm="randomized",
random_state=42).fit_transform(X.values)
t1 = time.time()
print("TruncatedSVD: ", (t1 - t0))
TruncatedSVD: 0.01066279411315918绘图
In [46]:
代码语言：javascript复制f, (ax1, ax2, ax3) = plt.subplots(1,3,figsize=(24,6))
# 标题设置
f.suptitle("Clusters using Dimensionality Reduction", fontsize=14)
blue_patch = mpatches.Patch(color="#0A0AFF", label="No Fraud")
red_patch = mpatches.Patch(color="#AF0000", label="Fraud")
# t-SNE
ax1.scatter(X_reduced_tsne[:,0],
X_reduced_tsne[:,1],
c=(y==0),
cmap="coolwarm",
label="No Fraud",
linewidths=2
)
ax1.scatter(X_reduced_tsne[:,0],
X_reduced_tsne[:,1],
c=(y==0),
cmap="coolwarm",
label="Fraud",
linewidths=2
)
ax1.set_title("t-SNE", fontsize=14) # 子图标题设置
ax1.grid(True) # 设置网格
ax1.legend(handles=[blue_patch,red_patch]) # 设置图例
# PCA
ax2.scatter(X_reduced_pca[:,0],
X_reduced_pca[:,1],
c=(y==0),
cmap="coolwarm",
label="No Fraud",
linewidths=2
)
ax2.scatter(X_reduced_pca[:,0],
X_reduced_pca[:,1],
c=(y==0),
cmap="coolwarm",
label="Fraud",
linewidths=2
)
ax2.set_title("PCA", fontsize=14) # 标题设置
ax2.grid(True) # 设置网格
ax2.legend(handles=[blue_patch,red_patch])
# TruncatedSVD
ax3.scatter(X_reduced_svd[:,0],
X_reduced_svd[:,1],
c=(y==0),
cmap="coolwarm",
label="No Fraud",
linewidths=2
)
ax3.scatter(X_reduced_svd[:,0],
X_reduced_svd[:,1],
c=(y==0),
cmap="coolwarm",
label="Fraud",
linewidths=2
)
ax3.set_title("TruncatedSVD", fontsize=14) # 标题设置
ax3.grid(True) # 设置网格
ax3.legend(handles=[blue_patch,red_patch])
plt.show()基于欠采样的分类建模
4个分类模型
采用4个不同模型的分类来训练数据，看哪个模型在欺诈数据上表现的更好。首先需要对数据进行划分：训练集和测试集
In [47]:
代码语言：javascript复制# 1、特征和标签数据
X = new_df.drop("Class", axis=1)
y = new_df["Class"]In [48]:
代码语言：javascript复制# 2、数据已经归一化，直接切分
from sklearn.model_selection import train_test_split
# 8-2的比例
X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.2,random_state=44)In [49]:
代码语言：javascript复制# 3、将数据转成数组，然后传给模型
X_train = X_train.values
X_test = X_test.values
y_train = y_train.values
y_test = y_test.valuesIn [50]:
代码语言：javascript复制# 4、创建4个模型
classifiers = {
"逻辑回归LogisiticRegression": LogisticRegression(),
"K近邻KNearest": KNeighborsClassifier(),
"支持向量机分类Support Vector Classifier": SVC(),
"决策树分类DecisionTreeClassifier": DecisionTreeClassifier()
}
for key, classifier in classifiers.items():
classifier.fit(X_train, y_train) # 模型训练
training_score = cross_val_score(classifier, # 模型
X_train, # 训练集数据
y_train,
cv=5) # 5折交叉验证
print("模型-", key,
"5次平均得分：", round(training_score.mean(), 2)*100)
模型- 逻辑回归LogisiticRegression 5次平均得分： 93.0
模型- K近邻KNearest 5次平均得分： 93.0
模型- 支持向量机分类Support Vector Classifier 5次平均得分： 93.0
模型- 决策树分类DecisionTreeClassifier 5次平均得分： 91.0网格搜索
针对不同测模型实施网格搜索，寻找最优参数
In [51]:
代码语言：javascript复制from sklearn.model_selection import GridSearchCV
# 逻辑回归
lr_params = {"penalty":["l1", "l2"],
"C": [0.001, 0.01, 0.1, 1, 10, 100, 1000]
}
grid_lr = GridSearchCV(LogisticRegression(), lr_params)
grid_lr.fit(X_train, y_train)
# 最好的参数组合
best_para_lr = grid_lr.best_estimator_
best_para_lrOut[51]:
代码语言：javascript复制LogisticRegression(C=0.1)In [52]:
代码语言：javascript复制# k近邻
knn_params = {"n_neighbors": list(range(2,5,1)),
"algorithm":["auto","ball_tree","kd_tree","brute"]
}
grid_knn = GridSearchCV(KNeighborsClassifier(), knn_params)
grid_knn.fit(X_train, y_train)
# 最好的参数组合
best_para_knn = grid_knn.best_estimator_
best_para_knnOut[52]:
代码语言：javascript复制KNeighborsClassifier(n_neighbors=2)In [53]:
代码语言：javascript复制# 支持向量机分类
svc_params = {"C":[0.5, 0.7, 0.9, 1],
"kernel":["rbf","poly","sigmoid","linear"]
}
grid_svc = GridSearchCV(SVC(), svc_params)
grid_svc.fit(X_train, y_train)
best_para_svc = grid_svc.best_estimator_
best_para_svcOut[53]:
代码语言：javascript复制SVC(C=0.9, kernel='linear')In [54]:
代码语言：javascript复制# 决策树
dt_params = {"criterion":["gini","entropy"],
"max_depth":list(range(2, 5, 1)),
"min_samples_leaf": list(range(5,7,1))
}
grid_dt = GridSearchCV(DecisionTreeClassifier(), dt_params)
grid_dt.fit(X_train, y_train)
best_para_dt = grid_dt.best_estimator_
best_para_dtOut[54]:
代码语言：javascript复制DecisionTreeClassifier(max_depth=3, min_samples_leaf=5)重新训练并评分
基于最优参数重新计算得分：
In [55]:
代码语言：javascript复制lr_score = cross_val_score(best_para_lr, X_train, y_train,cv=5)
print("逻辑回归交叉验证得分：", round(lr_score.mean() * 100, 2).astype(str) "%")
逻辑回归交叉验证得分： 93.63%In [56]:
代码语言：javascript复制knn_score = cross_val_score(best_para_knn, X_train, y_train,cv=5)
print("KNN交叉验证得分：", round(knn_score.mean() * 100, 2).astype(str) "%")
KNN交叉验证得分： 93.37%In [57]:
代码语言：javascript复制svc_score = cross_val_score(best_para_svc, X_train, y_train,cv=5)
print("SVC交叉验证得分：", round(svc_score.mean() * 100, 2).astype(str) "%")
SVC交叉验证得分： 93.5%In [58]:
代码语言：javascript复制dt_score = cross_val_score(best_para_dt, X_train, y_train,cv=5)
print("决策树交叉验证得分：", round(dt_score.mean() * 100, 2).astype(str) "%")
决策树交叉验证得分： 93.24%小结：通过不同模型的交叉验证得分我们发现，逻辑回归模型是最高的
基于欠采样数据的交叉验证
主要是基于Near-Miss算法来实现欠采样：
- Near-miss-1：选择到最近的三个样本平均距离最小的多数类样本
- Near-miss-2：选择到最远的三个样本平均距离最小的多数类样本
- Near-miss-3：为每个少数类样本选择给定数目的最近多数类样本
- 最远距离：选择到最近的三个样本平均距离最大的多样类样本
In [59]:
代码语言：javascript复制undersample_X = df.drop("Class", axis=1)
undersample_y = df["Class"]
sfk = StratifiedKFold(
n_splits=5, # 5次交叉验证
random_state=None,
shuffle=False)
for train_index , test_index in sfk.split(undersample_X,undersample_y):
# 每次随机生成的验证集和测试集的索引号
# print("Train: ", train_index)
# print("Test: ", test_index)
undersample_Xtrain = undersample_X.iloc[train_index] # 根据索引号来确定数据
undersample_Xtest = undersample_X.iloc[test_index]
undersample_ytrain = undersample_y.iloc[train_index]
undersample_ytest = undersample_y.iloc[test_index]
# 数据转成numpy数组
undersample_Xtrain = undersample_Xtrain.values
undersample_Xtest = undersample_Xtest.values
undersample_ytrain = undersample_ytrain.values
undersample_ytest = undersample_ytest.values
# 5个评价指标
undersample_accuracy = []
undersample_precision = []
undersample_recall = []
undersample_f1 = []
undersample_auc = []使用近邻缺失Near-Miss算法来查看数据分布：
In [60]:
代码语言：javascript复制undersample_X = df.drop("Class", axis=1)
undersample_y = df["Class"]
X_nearmiss, y_nearmiss = NearMiss().fit_resample(undersample_X.values, undersample_y.values)
print("NearMiss Label Distributions: {}", format(Counter(y_nearmiss)))
NearMiss Label Distributions: {} Counter({0: 492, 1: 492})以网格搜索过后的逻辑回归模型来实施交叉验证：
In [61]:
代码语言：javascript复制for train, test in sfk.split(undersample_Xtrain, undersample_ytrain):
undersample_pipeline = imbalanced_make_pipeline(NearMiss(sampling_strategy="majority"), best_para_lr)
# 模型训练：传入训练X-y
undersample_model = undersample_pipeline.fit(undersample_Xtrain[train], undersample_ytrain[train])
# 对测试集预测-Xtrain
undersample_prediction = undersample_model.predict(undersample_Xtrain[test])
# y_test真实值和预测值的评分
undersample_accuracy.append(undersample_pipeline.score(original_Xtrain[test], original_ytrain[test]))
undersample_precision.append(precision_score(original_ytrain[test], undersample_prediction))
undersample_recall.append(recall_score(original_ytrain[test], undersample_prediction))
undersample_f1.append(f1_score(original_ytrain[test], undersample_prediction))
undersample_auc.append(roc_auc_score(original_ytrain[test], undersample_prediction))绘制学习曲线
In [62]:
代码语言：javascript复制from sklearn.model_selection import ShuffleSplit, learning_curveIn [63]:
代码语言：javascript复制def plot_learning_curve(est1,est2,est3,est4,X,y,ylim=None,cv=None,n_jobs=1,train_sizes=np.linspace(0.1, 1, 5)):
f, ((ax1,ax2), (ax3,ax4)) = plt.subplots(2,2,figsize=(20,14), sharey=True)
if ylim is not None:
plt.ylim(*ylim)
# 模型1
train_sizes, train_scores, test_scores = learning_curve(
est1, X, y, cv=cv, n_jobs=n_jobs, train_sizes=train_sizes)
train_scores_mean = np.mean(train_scores, axis=1) # 训练集的均值和方差
train_scores_std = np.std(train_scores, axis=1)
test_scores_mean = np.mean(test_scores, axis=1) # 测试集的均值和方差
test_scores_std = np.std(test_scores, axis=1)
ax1.fill_between(train_sizes, train_scores_mean - train_scores_std, # 填充区域的设置
train_scores_mean train_scores_std, alpha=0.1,
color="#ff9124")
ax1.fill_between(train_sizes, test_scores_mean - test_scores_std,
test_scores_mean test_scores_std, alpha=0.1,
color="#2492ff")
ax1.plot(train_sizes, train_scores_mean,
'o-', color="#ff9124",
label="Training score") # 绘制训练集得分
ax1.plot(train_sizes, test_scores_mean,
'o-', color="#2492ff",
label="Cross-validation score") # 绘制交叉验证得分
ax1.set_title("逻辑回归学习曲线", fontsize=14)
ax1.set_xlabel('Training size (m)') # 两个轴的标题
ax1.set_ylabel('Score')
ax1.grid(True) # 网格显示
ax1.legend(loc="best") # 图例位置
# 模型2-knn
train_sizes, train_scores, test_scores = learning_curve(
est2, X, y, cv=cv, n_jobs=n_jobs, train_sizes=train_sizes)
train_scores_mean = np.mean(train_scores, axis=1)
train_scores_std = np.std(train_scores, axis=1)
test_scores_mean = np.mean(test_scores, axis=1)
test_scores_std = np.std(test_scores, axis=1)
ax2.fill_between(train_sizes, train_scores_mean - train_scores_std,
train_scores_mean train_scores_std, alpha=0.1,
color="#ff9124")
ax2.fill_between(train_sizes, test_scores_mean - test_scores_std,
test_scores_mean test_scores_std, alpha=0.1, color="#2492ff")
ax2.plot(train_sizes, train_scores_mean, 'o-', color="#ff9124",
label="Training score")
ax2.plot(train_sizes, test_scores_mean, 'o-', color="#2492ff",
label="Cross-validation score")
ax2.set_title("k近邻学习曲线", fontsize=14)
ax2.set_xlabel('Training size (m)')
ax2.set_ylabel('Score')
ax2.grid(True)
ax2.legend(loc="best")
# 模型3-支持向量机
train_sizes, train_scores, test_scores = learning_curve(
est3, X, y, cv=cv, n_jobs=n_jobs, train_sizes=train_sizes)
train_scores_mean = np.mean(train_scores, axis=1)
train_scores_std = np.std(train_scores, axis=1)
test_scores_mean = np.mean(test_scores, axis=1)
test_scores_std = np.std(test_scores, axis=1)
ax3.fill_between(train_sizes, train_scores_mean - train_scores_std,
train_scores_mean train_scores_std, alpha=0.1,
color="#ff9124")
ax3.fill_between(train_sizes, test_scores_mean - test_scores_std,
test_scores_mean test_scores_std, alpha=0.1, color="#2492ff")
ax3.plot(train_sizes, train_scores_mean, 'o-', color="#ff9124",
label="Training score")
ax3.plot(train_sizes, test_scores_mean, 'o-', color="#2492ff",
label="Cross-validation score")
ax3.set_title("支持向量机学习曲线", fontsize=14)
ax3.set_xlabel('Training size (m)')
ax3.set_ylabel('Score')
ax3.grid(True)
ax3.legend(loc="best")
# 模型4-决策树
train_sizes, train_scores, test_scores = learning_curve(
est4, X, y, cv=cv, n_jobs=n_jobs, train_sizes=train_sizes)
train_scores_mean = np.mean(train_scores, axis=1)
train_scores_std = np.std(train_scores, axis=1)
test_scores_mean = np.mean(test_scores, axis=1)
test_scores_std = np.std(test_scores, axis=1)
ax4.fill_between(train_sizes, train_scores_mean - train_scores_std,
train_scores_mean train_scores_std, alpha=0.1,
color="#ff9124")
ax4.fill_between(train_sizes, test_scores_mean - test_scores_std,
test_scores_mean test_scores_std, alpha=0.1, color="#2492ff")
ax4.plot(train_sizes, train_scores_mean, 'o-', color="#ff9124",
label="Training score")
ax4.plot(train_sizes, test_scores_mean, 'o-', color="#2492ff",
label="Cross-validation score")
ax4.set_title("决策树学习曲线", fontsize=14)
ax4.set_xlabel('Training size (m)')
ax4.set_ylabel('Score')
ax4.grid(True)
ax4.legend(loc="best")
return pltIn [64]:
代码语言：javascript复制cv = ShuffleSplit(n_splits=100,
test_size=0.2,
random_state=42
)
plot_learning_curve(best_para_lr, # 传入4种模型
best_para_knn,
best_para_svc,
best_para_dt,
X_train,
y_train,
(0.87,1.01),
cv=cv,
n_jobs=4
)
plt.showroc曲线
In [65]:
代码语言：javascript复制from sklearn.metrics import roc_curve, roc_auc_score
from sklearn.model_selection import cross_val_predictIn [66]:
代码语言：javascript复制lr_pred = cross_val_predict(best_para_lr,
X_train,
y_train,
cv=5,
# method="decision_function"
)
knn_pred = cross_val_predict(best_para_knn,
X_train,
y_train,
cv=5
)
svc_pred = cross_val_predict(best_para_svc,
X_train,
y_train,
cv=5
)
dt_pred = cross_val_predict(best_para_dt,
X_train,
y_train,
cv=5
)In [67]:
代码语言：javascript复制print('Logistic Regression: ', roc_auc_score(y_train, lr_pred))
print('KNears Neighbors: ', roc_auc_score(y_train, knn_pred))
print('Support Vector Classifier: ', roc_auc_score(y_train, svc_pred))
print('Decision Tree Classifier: ', roc_auc_score(y_train, dt_pred))
Logistic Regression: 0.934970120644943
KNears Neighbors: 0.9314677528469951
Support Vector Classifier: 0.9339060209719247
Decision Tree Classifier: 0.930932179501635In [68]:
代码语言：javascript复制log_fpr, log_tpr, log_thresold = roc_curve(y_train, lr_pred)
knear_fpr, knear_tpr, knear_threshold = roc_curve(y_train, knn_pred)
svc_fpr, svc_tpr, svc_threshold = roc_curve(y_train, svc_pred)
tree_fpr, tree_tpr, tree_threshold = roc_curve(y_train, dt_pred)
# 绘制tpr-fpr得分图
def graph_roc_curve_multiple(log_fpr, log_tpr, knear_fpr, knear_tpr, svc_fpr, svc_tpr, tree_fpr, tree_tpr):
plt.figure(figsize=(16,8))
plt.title('ROC Curve n Top 4 Classifiers', fontsize=18)
plt.plot(log_fpr, log_tpr, label='Logistic Regression Classifier Score: {:.4f}'.format(roc_auc_score(y_train, lr_pred)))
plt.plot(knear_fpr, knear_tpr, label='KNears Neighbors Classifier Score: {:.4f}'.format(roc_auc_score(y_train, knn_pred)))
plt.plot(svc_fpr, svc_tpr, label='Support Vector Classifier Score: {:.4f}'.format(roc_auc_score(y_train, svc_pred)))
plt.plot(tree_fpr, tree_tpr, label='Decision Tree Classifier Score: {:.4f}'.format(roc_auc_score(y_train, dt_pred)))
plt.plot([0, 1], [0, 1], 'k--')
plt.axis([-0.01, 1, 0, 1])
plt.xlabel('False Positive Rate', fontsize=16)
plt.ylabel('True Positive Rate', fontsize=16)
plt.annotate('Minimum ROC Score of 50% n (This is the minimum score to get)',
xy=(0.5, 0.5),
xytext=(0.6, 0.3),
arrowprops=dict(facecolor='#6E726D', shrink=0.05),
)
plt.legend()
graph_roc_curve_multiple(log_fpr, log_tpr, knear_fpr, knear_tpr, svc_fpr, svc_tpr, tree_fpr, tree_tpr)
plt.show()探索逻辑回归评价指标
探索在逻辑回归模型的分类评价指标：
In [69]:
代码语言：javascript复制def logistic_roc_curve(log_fpr, log_tpr):
plt.figure(figsize=(12,8))
plt.title('Logistic Regression ROC Curve', fontsize=16)
plt.plot(log_fpr, log_tpr, 'b-', linewidth=2)
plt.plot([0, 1], [0, 1], 'r--')
plt.xlabel('False Positive Rate', fontsize=16)
plt.ylabel('True Positive Rate', fontsize=16)
plt.axis([-0.01,1,0,1])
logistic_roc_curve(log_fpr, log_tpr)
plt.show()from sklearn.metrics import precision_recall_curve
precision, recall, threshold = precision_recall_curve(y_train, lr_pred)In [71]:
代码语言：javascript复制from sklearn.metrics import recall_score, precision_score, f1_score, accuracy_score
y_pred = best_para_lr.predict(X_train)
# Overfitting Case
print('---' * 20)
print('Recall Score: {:.2f}'.format(recall_score(y_train, y_pred)))
print('Precision Score: {:.2f}'.format(precision_score(y_train, y_pred)))
print('F1 Score: {:.2f}'.format(f1_score(y_train, y_pred)))
print('Accuracy Score: {:.2f}'.format(accuracy_score(y_train, y_pred)))
print('---' * 20)
print("Accuracy Score: {:.2f}".format(np.mean(undersample_accuracy)))
print("Precision Score: {:.2f}".format(np.mean(undersample_precision)))
print("Recall Score: {:.2f}".format(np.mean(undersample_recall)))
print("F1 Score: {:.2f}".format(np.mean(undersample_f1)))
print('---' * 20)
------------------------------------------------------------
# 基于原数据
Recall Score: 0.92
Precision Score: 0.79
F1 Score: 0.85
Accuracy Score: 0.84
------------------------------------------------------------
# 基于欠采样的数据
Accuracy Score: 0.75
Precision Score: 0.00
Recall Score: 0.24
F1 Score: 0.00
------------------------------------------------------------
