From 1c143f339923d1183cb75779ba429722c9ff2e32 Mon Sep 17 00:00:00 2001
From: Doubleking-1 <71910936+Doubleking-1@users.noreply.github.com>
Date: Wed, 3 Jan 2024 23:17:02 +0800
Subject: [PATCH] Wangzun

second homework of ML
---
 ...ork_credit_scoring_finetune_ensemble.ipynb | 2073 +++++++++++++++++
 1 file changed, 2073 insertions(+)
 create mode 100644 2023/homework/Zun_Wang/homework_credit_scoring_finetune_ensemble.ipynb

diff --git a/2023/homework/Zun_Wang/homework_credit_scoring_finetune_ensemble.ipynb b/2023/homework/Zun_Wang/homework_credit_scoring_finetune_ensemble.ipynb
new file mode 100644
index 00000000..87a7e7a5
--- /dev/null
+++ b/2023/homework/Zun_Wang/homework_credit_scoring_finetune_ensemble.ipynb
@@ -0,0 +1,2073 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 一起来打怪之 Credit Scoring 练习"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "-------\n",
+    "## >>>说明：\n",
+    "### 1. 答题步骤：\n",
+    "- 回答问题**请保留每一步**操作过程，请不要仅仅给出最后答案\n",
+    "- 请养成代码注释的好习惯\n",
+    "\n",
+    "### 2. 解题思路：\n",
+    "- 为方便大家准确理解题目，在习题实战中有所收获，本文档提供了解题思路提示\n",
+    "- 解题思路**仅供参考**，鼓励原创解题方法\n",
+    "- 为督促同学们自己思考，解题思路内容设置为**注释**，请注意查看\n",
+    "\n",
+    "### 3. 所用数据：\n",
+    "- 问题使用了多个数据库，请注意导入每个数据库后都先**查看和了解数据的基本性质**，后面的问题不再一一提醒"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "--------\n",
+    "## 操作题"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 信用卡欺诈项目"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    " #### 前期数据导入,预览及处理(此部分勿修改，涉及的数据文件无需复制移动)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>SeriousDlqin2yrs</th>\n",
+       "      <th>RevolvingUtilizationOfUnsecuredLines</th>\n",
+       "      <th>age</th>\n",
+       "      <th>NumberOfTime30-59DaysPastDueNotWorse</th>\n",
+       "      <th>DebtRatio</th>\n",
+       "      <th>MonthlyIncome</th>\n",
+       "      <th>NumberOfOpenCreditLinesAndLoans</th>\n",
+       "      <th>NumberOfTimes90DaysLate</th>\n",
+       "      <th>NumberRealEstateLoansOrLines</th>\n",
+       "      <th>NumberOfTime60-89DaysPastDueNotWorse</th>\n",
+       "      <th>NumberOfDependents</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0.766127</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.802982</td>\n",
+       "      <td>9120.0</td>\n",
+       "      <td>13.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>2.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.957151</td>\n",
+       "      <td>40.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.121876</td>\n",
+       "      <td>2600.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.658180</td>\n",
+       "      <td>38.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.085113</td>\n",
+       "      <td>3042.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.233810</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.036050</td>\n",
+       "      <td>3300.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.907239</td>\n",
+       "      <td>49.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.024926</td>\n",
+       "      <td>63588.0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   SeriousDlqin2yrs  RevolvingUtilizationOfUnsecuredLines   age  \\\n",
+       "0                 1                              0.766127  45.0   \n",
+       "1                 0                              0.957151  40.0   \n",
+       "2                 0                              0.658180  38.0   \n",
+       "3                 0                              0.233810  30.0   \n",
+       "4                 0                              0.907239  49.0   \n",
+       "\n",
+       "   NumberOfTime30-59DaysPastDueNotWorse  DebtRatio  MonthlyIncome  \\\n",
+       "0                                   2.0   0.802982         9120.0   \n",
+       "1                                   0.0   0.121876         2600.0   \n",
+       "2                                   1.0   0.085113         3042.0   \n",
+       "3                                   0.0   0.036050         3300.0   \n",
+       "4                                   1.0   0.024926        63588.0   \n",
+       "\n",
+       "   NumberOfOpenCreditLinesAndLoans  NumberOfTimes90DaysLate  \\\n",
+       "0                             13.0                      0.0   \n",
+       "1                              4.0                      0.0   \n",
+       "2                              2.0                      1.0   \n",
+       "3                              5.0                      0.0   \n",
+       "4                              7.0                      0.0   \n",
+       "\n",
+       "   NumberRealEstateLoansOrLines  NumberOfTime60-89DaysPastDueNotWorse  \\\n",
+       "0                           6.0                                   0.0   \n",
+       "1                           0.0                                   0.0   \n",
+       "2                           0.0                                   0.0   \n",
+       "3                           0.0                                   0.0   \n",
+       "4                           1.0                                   0.0   \n",
+       "\n",
+       "   NumberOfDependents  \n",
+       "0                 2.0  \n",
+       "1                 1.0  \n",
+       "2                 0.0  \n",
+       "3                 0.0  \n",
+       "4                 0.0  "
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "pd.set_option('display.max_columns', 500)\n",
+    "import zipfile\n",
+    "with zipfile.ZipFile('KaggleCredit2.csv.zip', 'r') as z:\n",
+    "    f = z.open('KaggleCredit2.csv')\n",
+    "    data = pd.read_csv(f, index_col=0)\n",
+    "data.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(112915, 11)"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# 检查数据维度\n",
+    "data.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "SeriousDlqin2yrs                           0\n",
+       "RevolvingUtilizationOfUnsecuredLines       0\n",
+       "age                                     4267\n",
+       "NumberOfTime30-59DaysPastDueNotWorse       0\n",
+       "DebtRatio                                  0\n",
+       "MonthlyIncome                              0\n",
+       "NumberOfOpenCreditLinesAndLoans            0\n",
+       "NumberOfTimes90DaysLate                    0\n",
+       "NumberRealEstateLoansOrLines               0\n",
+       "NumberOfTime60-89DaysPastDueNotWorse       0\n",
+       "NumberOfDependents                      4267\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# 查看数据缺失值情况\n",
+    "data.isnull().sum(axis=0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/tmp/ipykernel_253566/2980780030.py:3: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n",
+      "  data.shapey = data['SeriousDlqin2yrs']\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 清除缺失值\n",
+    "data.dropna(inplace=True)\n",
+    "data.shapey = data['SeriousDlqin2yrs']\n",
+    "X = data.drop('SeriousDlqin2yrs', axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.06742876076872101"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# 取出对应的X和y\n",
+    "y = data['SeriousDlqin2yrs']\n",
+    "X = data.drop('SeriousDlqin2yrs', axis=1)\n",
+    "# 查看平均的欺诈率\n",
+    "y.mean()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 以下为操作题"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 1.把数据切分成训练集和测试集"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "((76053, 10), (32595, 10), (76053,), (32595,), (108648, 10))"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# 提示：查看train_test_split函数\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "# 把数据切分成70%的训练集，30%的测试集\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)\n",
+    "\n",
+    "# 查看训练集跟测试集的维度---原来数据集的维度\n",
+    "X_train.shape, X_test.shape, y_train.shape, y_test.shape, X.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 0, 'Catalog')"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# 通过SeriousDlqin2yrs字段查看正负样本分布情况\n",
+    "# 提示：value_counts\n",
+    "data_counts = data['SeriousDlqin2yrs'].value_counts()\n",
+    "\n",
+    "# 绘制两种类别的柱状图\n",
+    "# 提示：dataframe可以直接plot(kind='bar')\n",
+    "import matplotlib.pyplot as plt\n",
+    "fig = plt.figure()\n",
+    "data_counts.plot(kind='bar')\n",
+    "plt.title('Positive and negative sample distribution')\n",
+    "plt.ylabel('Number')\n",
+    "plt.xlabel('Catalog')\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 2.数据预处理之离散化"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# 请对年龄按照3岁一个区间进行离散化\n",
+    "# 提示：可以先计算出分桶边界，再基于pandas的cut函数进行离散化(分箱、分桶)\n",
+    "low = min(data['age'])\n",
+    "high = max(data['age'])\n",
+    "\n",
+    "bins = list(range(int(low), int(high), 3))  # 以3岁为一个区间\n",
+    "\n",
+    "# 使用 cut 函数进行离散化\n",
+    "data['age_group'] = pd.cut(data['age'], bins)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 3.数据预处理之独热向量编码"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>SeriousDlqin2yrs</th>\n",
+       "      <th>RevolvingUtilizationOfUnsecuredLines</th>\n",
+       "      <th>age</th>\n",
+       "      <th>NumberOfTime30-59DaysPastDueNotWorse</th>\n",
+       "      <th>DebtRatio</th>\n",
+       "      <th>MonthlyIncome</th>\n",
+       "      <th>NumberOfOpenCreditLinesAndLoans</th>\n",
+       "      <th>NumberOfTimes90DaysLate</th>\n",
+       "      <th>NumberRealEstateLoansOrLines</th>\n",
+       "      <th>NumberOfTime60-89DaysPastDueNotWorse</th>\n",
+       "      <th>NumberOfDependents</th>\n",
+       "      <th>age_group</th>\n",
+       "      <th>age_group_(0, 3]</th>\n",
+       "      <th>age_group_(3, 6]</th>\n",
+       "      <th>age_group_(6, 9]</th>\n",
+       "      <th>age_group_(9, 12]</th>\n",
+       "      <th>age_group_(12, 15]</th>\n",
+       "      <th>age_group_(15, 18]</th>\n",
+       "      <th>age_group_(18, 21]</th>\n",
+       "      <th>age_group_(21, 24]</th>\n",
+       "      <th>age_group_(24, 27]</th>\n",
+       "      <th>age_group_(27, 30]</th>\n",
+       "      <th>age_group_(30, 33]</th>\n",
+       "      <th>age_group_(33, 36]</th>\n",
+       "      <th>age_group_(36, 39]</th>\n",
+       "      <th>age_group_(39, 42]</th>\n",
+       "      <th>age_group_(42, 45]</th>\n",
+       "      <th>age_group_(45, 48]</th>\n",
+       "      <th>age_group_(48, 51]</th>\n",
+       "      <th>age_group_(51, 54]</th>\n",
+       "      <th>age_group_(54, 57]</th>\n",
+       "      <th>age_group_(57, 60]</th>\n",
+       "      <th>age_group_(60, 63]</th>\n",
+       "      <th>age_group_(63, 66]</th>\n",
+       "      <th>age_group_(66, 69]</th>\n",
+       "      <th>age_group_(69, 72]</th>\n",
+       "      <th>age_group_(72, 75]</th>\n",
+       "      <th>age_group_(75, 78]</th>\n",
+       "      <th>age_group_(78, 81]</th>\n",
+       "      <th>age_group_(81, 84]</th>\n",
+       "      <th>age_group_(84, 87]</th>\n",
+       "      <th>age_group_(87, 90]</th>\n",
+       "      <th>age_group_(90, 93]</th>\n",
+       "      <th>age_group_(93, 96]</th>\n",
+       "      <th>age_group_(96, 99]</th>\n",
+       "      <th>age_group_(99, 102]</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0.766127</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.802982</td>\n",
+       "      <td>9120.0</td>\n",
+       "      <td>13.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>(42, 45]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.957151</td>\n",
+       "      <td>40.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.121876</td>\n",
+       "      <td>2600.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>(39, 42]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.658180</td>\n",
+       "      <td>38.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.085113</td>\n",
+       "      <td>3042.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(36, 39]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.233810</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.036050</td>\n",
+       "      <td>3300.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(27, 30]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.907239</td>\n",
+       "      <td>49.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.024926</td>\n",
+       "      <td>63588.0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(48, 51]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>112910</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.385742</td>\n",
+       "      <td>50.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.404293</td>\n",
+       "      <td>3400.0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(48, 51]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>112911</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.040674</td>\n",
+       "      <td>74.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.225131</td>\n",
+       "      <td>2100.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(72, 75]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>112912</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.299745</td>\n",
+       "      <td>44.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.716562</td>\n",
+       "      <td>5584.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>(42, 45]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>112913</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>5716.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(27, 30]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>112914</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0.850283</td>\n",
+       "      <td>64.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.249908</td>\n",
+       "      <td>8158.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>(63, 66]</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>108648 rows × 46 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        SeriousDlqin2yrs  RevolvingUtilizationOfUnsecuredLines   age  \\\n",
+       "0                      1                              0.766127  45.0   \n",
+       "1                      0                              0.957151  40.0   \n",
+       "2                      0                              0.658180  38.0   \n",
+       "3                      0                              0.233810  30.0   \n",
+       "4                      0                              0.907239  49.0   \n",
+       "...                  ...                                   ...   ...   \n",
+       "112910                 0                              0.385742  50.0   \n",
+       "112911                 0                              0.040674  74.0   \n",
+       "112912                 0                              0.299745  44.0   \n",
+       "112913                 0                              0.000000  30.0   \n",
+       "112914                 0                              0.850283  64.0   \n",
+       "\n",
+       "        NumberOfTime30-59DaysPastDueNotWorse  DebtRatio  MonthlyIncome  \\\n",
+       "0                                        2.0   0.802982         9120.0   \n",
+       "1                                        0.0   0.121876         2600.0   \n",
+       "2                                        1.0   0.085113         3042.0   \n",
+       "3                                        0.0   0.036050         3300.0   \n",
+       "4                                        1.0   0.024926        63588.0   \n",
+       "...                                      ...        ...            ...   \n",
+       "112910                                   0.0   0.404293         3400.0   \n",
+       "112911                                   0.0   0.225131         2100.0   \n",
+       "112912                                   0.0   0.716562         5584.0   \n",
+       "112913                                   0.0   0.000000         5716.0   \n",
+       "112914                                   0.0   0.249908         8158.0   \n",
+       "\n",
+       "        NumberOfOpenCreditLinesAndLoans  NumberOfTimes90DaysLate  \\\n",
+       "0                                  13.0                      0.0   \n",
+       "1                                   4.0                      0.0   \n",
+       "2                                   2.0                      1.0   \n",
+       "3                                   5.0                      0.0   \n",
+       "4                                   7.0                      0.0   \n",
+       "...                                 ...                      ...   \n",
+       "112910                              7.0                      0.0   \n",
+       "112911                              4.0                      0.0   \n",
+       "112912                              4.0                      0.0   \n",
+       "112913                              4.0                      0.0   \n",
+       "112914                              8.0                      0.0   \n",
+       "\n",
+       "        NumberRealEstateLoansOrLines  NumberOfTime60-89DaysPastDueNotWorse  \\\n",
+       "0                                6.0                                   0.0   \n",
+       "1                                0.0                                   0.0   \n",
+       "2                                0.0                                   0.0   \n",
+       "3                                0.0                                   0.0   \n",
+       "4                                1.0                                   0.0   \n",
+       "...                              ...                                   ...   \n",
+       "112910                           0.0                                   0.0   \n",
+       "112911                           1.0                                   0.0   \n",
+       "112912                           1.0                                   0.0   \n",
+       "112913                           0.0                                   0.0   \n",
+       "112914                           2.0                                   0.0   \n",
+       "\n",
+       "        NumberOfDependents age_group  age_group_(0, 3]  age_group_(3, 6]  \\\n",
+       "0                      2.0  (42, 45]                 0                 0   \n",
+       "1                      1.0  (39, 42]                 0                 0   \n",
+       "2                      0.0  (36, 39]                 0                 0   \n",
+       "3                      0.0  (27, 30]                 0                 0   \n",
+       "4                      0.0  (48, 51]                 0                 0   \n",
+       "...                    ...       ...               ...               ...   \n",
+       "112910                 0.0  (48, 51]                 0                 0   \n",
+       "112911                 0.0  (72, 75]                 0                 0   \n",
+       "112912                 2.0  (42, 45]                 0                 0   \n",
+       "112913                 0.0  (27, 30]                 0                 0   \n",
+       "112914                 0.0  (63, 66]                 0                 0   \n",
+       "\n",
+       "        age_group_(6, 9]  age_group_(9, 12]  age_group_(12, 15]  \\\n",
+       "0                      0                  0                   0   \n",
+       "1                      0                  0                   0   \n",
+       "2                      0                  0                   0   \n",
+       "3                      0                  0                   0   \n",
+       "4                      0                  0                   0   \n",
+       "...                  ...                ...                 ...   \n",
+       "112910                 0                  0                   0   \n",
+       "112911                 0                  0                   0   \n",
+       "112912                 0                  0                   0   \n",
+       "112913                 0                  0                   0   \n",
+       "112914                 0                  0                   0   \n",
+       "\n",
+       "        age_group_(15, 18]  age_group_(18, 21]  age_group_(21, 24]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(24, 27]  age_group_(27, 30]  age_group_(30, 33]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   1                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   1                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(33, 36]  age_group_(36, 39]  age_group_(39, 42]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   1   \n",
+       "2                        0                   1                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(42, 45]  age_group_(45, 48]  age_group_(48, 51]  \\\n",
+       "0                        1                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   1   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   1   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   1                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(51, 54]  age_group_(54, 57]  age_group_(57, 60]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(60, 63]  age_group_(63, 66]  age_group_(66, 69]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   1                   0   \n",
+       "\n",
+       "        age_group_(69, 72]  age_group_(72, 75]  age_group_(75, 78]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   1                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(78, 81]  age_group_(81, 84]  age_group_(84, 87]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(87, 90]  age_group_(90, 93]  age_group_(93, 96]  \\\n",
+       "0                        0                   0                   0   \n",
+       "1                        0                   0                   0   \n",
+       "2                        0                   0                   0   \n",
+       "3                        0                   0                   0   \n",
+       "4                        0                   0                   0   \n",
+       "...                    ...                 ...                 ...   \n",
+       "112910                   0                   0                   0   \n",
+       "112911                   0                   0                   0   \n",
+       "112912                   0                   0                   0   \n",
+       "112913                   0                   0                   0   \n",
+       "112914                   0                   0                   0   \n",
+       "\n",
+       "        age_group_(96, 99]  age_group_(99, 102]  \n",
+       "0                        0                    0  \n",
+       "1                        0                    0  \n",
+       "2                        0                    0  \n",
+       "3                        0                    0  \n",
+       "4                        0                    0  \n",
+       "...                    ...                  ...  \n",
+       "112910                   0                    0  \n",
+       "112911                   0                    0  \n",
+       "112912                   0                    0  \n",
+       "112913                   0                    0  \n",
+       "112914                   0                    0  \n",
+       "\n",
+       "[108648 rows x 46 columns]"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# 请对上述分箱后的年龄段进行独热向量编码\n",
+    "# 提示：使用pandas的get_dummies完成\n",
+    "# 对年龄段进行独热向量编码\n",
+    "one_hot_encoded = pd.get_dummies(data['age_group'], prefix='age_group')\n",
+    "\n",
+    "# 将独热编码结果与原始 DataFrame 进行合并\n",
+    "data = pd.concat([data, one_hot_encoded], axis=1)\n",
+    "\n",
+    "data\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 4.数据预处理之幅度缩放"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# 请对连续值特征进行幅度缩放\n",
+    "# 提示：可以使用StandardScaler等幅度缩放器进行处理\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "\n",
+    "sc = StandardScaler()\n",
+    "X_train_std = sc.fit_transform(X_train)\n",
+    "X_test_std = sc.fit_transform(X_test)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 5.使用logistic regression建模，并且输出一下系数，分析重要度。   "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "系数： [[-0.01427294 -0.36429202  1.72869067  0.31207913 -0.11534462 -0.09187206\n",
+      "   1.68994946 -0.19639706 -3.2487085   0.11638382]]\n",
+      "截距： [-2.85903863]\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 提示：fit建模，建完模之后可以取出coef属性\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "import seaborn as sns\n",
+    "\n",
+    "lr = LogisticRegression(C=1000.0,random_state=0, penalty='l2', solver='liblinear') \n",
+    "lr.fit(X_train_std, y_train)\n",
+    "# 输出系数\n",
+    "coefficients = lr.coef_\n",
+    "intercept = lr.intercept_\n",
+    "\n",
+    "print(\"系数：\", coefficients)\n",
+    "print(\"截距：\", intercept)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 6.使用网格搜索交叉验证进行调参\n",
+    "调整penalty和C参数，其中penalty候选为\"l1\"和\"l2\"，C的候选为[1,10,100,500]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "最佳参数： {'C': 1, 'penalty': 'l1'}\n",
+      "最佳模型： LogisticRegression(C=1, penalty='l1', random_state=0, solver='liblinear')\n",
+      "在测试集上的准确率： 0.9339162448228255\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 提示：先按照上面要求准备好网格字典，再使用GridSearchCV进行调参\n",
+    "# 设置参数候选\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "\n",
+    "param_grid = {\n",
+    "    'penalty': ['l1', 'l2'],\n",
+    "    'C': [1, 10, 100, 500]\n",
+    "}\n",
+    "\n",
+    "# 初始化 GridSearchCV\n",
+    "grid_search = GridSearchCV(lr, param_grid, cv=5)\n",
+    "\n",
+    "# 执行网格搜索交叉验证\n",
+    "grid_search.fit(X_train_std, y_train)\n",
+    "\n",
+    "# 输出最佳参数\n",
+    "print(\"最佳参数：\", grid_search.best_params_)\n",
+    "\n",
+    "# 输出最佳模型\n",
+    "best_model = grid_search.best_estimator_\n",
+    "print(\"最佳模型：\", best_model)\n",
+    "\n",
+    "# 在测试集上评估最佳模型\n",
+    "accuracy = best_model.score(X_test_std, y_test)\n",
+    "print(\"在测试集上的准确率：\", accuracy)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 7.在测试集上进行预测，计算 查准率/查全率/auc/混淆矩阵/f1值 等测试指标"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "准确率： 0.9339162448228255\n",
+      "查全率： 0.046061722708429294\n",
+      "AUC： 0.6979862258129022\n",
+      "F1值： 0.08496176720475784\n",
+      "混淆矩阵：\n",
+      "[[30341    83]\n",
+      " [ 2071   100]]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# 提示：在测试集上预测可以使用predict\n",
+    "# 提示：各种指标可以在sklearn.metrics中查到各种评估指标，分别是accuracy_score、recall_score、auc、confusion_matrix、f1_score\n",
+    "\n",
+    "from sklearn.metrics import accuracy_score, recall_score, roc_auc_score, confusion_matrix, f1_score,ConfusionMatrixDisplay\n",
+    "\n",
+    "# 在测试集上进行预测\n",
+    "y_pred = best_model.predict(X_test_std)\n",
+    "\n",
+    "# 计算准确率\n",
+    "accuracy = accuracy_score(y_test, y_pred)\n",
+    "print(\"准确率：\", accuracy)\n",
+    "\n",
+    "# 计算查全率\n",
+    "recall = recall_score(y_test, y_pred)\n",
+    "print(\"查全率：\", recall)\n",
+    "\n",
+    "# 计算AUC\n",
+    "y_pred_proba = best_model.predict_proba(X_test_std)[:, 1]  # 预测为正例的概率\n",
+    "auc = roc_auc_score(y_test, y_pred_proba)\n",
+    "print(\"AUC：\", auc)\n",
+    "\n",
+    "# 计算F1值\n",
+    "f1 = f1_score(y_test, y_pred)\n",
+    "print(\"F1值：\", f1)\n",
+    "\n",
+    "# 计算混淆矩阵\n",
+    "cm = confusion_matrix(y_test, y_pred)\n",
+    "print(\"混淆矩阵：\")\n",
+    "print(cm)\n",
+    "disp_lr = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=lr.classes_)\n",
+    "\n",
+    "disp_lr.plot()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 8.更多优化\n",
+    "银行通常会有更严格的要求，因为欺诈带来的后果通常比较严重，一般我们会调整模型的标准。   \n",
+    "\n",
+    "比如在logistic regression当中，一般我们的概率判定边界为0.5，但是我们可以把阈值设定低一些，来提高模型的“敏感度”   \n",
+    "试试看把阈值设定为0.3，再看看这个时候的混淆矩阵等评估指标。"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "阈值为 0.1 时的混淆矩阵：\n",
+      "[[26778  3646]\n",
+      " [ 1253   918]]\n",
+      "阈值为 0.1 时的准确率： 0.8497008743672343\n",
+      "阈值为 0.1 时的查全率： 0.42284661446338095\n",
+      "阈值为 0.1 时的AUC： 0.6515035070739202\n",
+      "阈值为 0.1 时的F1值： 0.2726057906458797\n",
+      "--------------------\n",
+      "阈值为 0.2 时的混淆矩阵：\n",
+      "[[29799   625]\n",
+      " [ 1754   417]]\n",
+      "阈值为 0.2 时的准确率： 0.9270133456051541\n",
+      "阈值为 0.2 时的查全率： 0.19207738369415017\n",
+      "阈值为 0.2 时的AUC： 0.5857671956598545\n",
+      "阈值为 0.2 时的F1值： 0.25957049486461253\n",
+      "--------------------\n",
+      "阈值为 0.3 时的混淆矩阵：\n",
+      "[[30178   246]\n",
+      " [ 1927   244]]\n",
+      "阈值为 0.3 时的准确率： 0.9333333333333333\n",
+      "阈值为 0.3 时的查全率： 0.11239060340856748\n",
+      "阈值为 0.3 时的AUC： 0.5521524408049937\n",
+      "阈值为 0.3 时的F1值： 0.18338970311912817\n",
+      "--------------------\n",
+      "阈值为 0.4 时的混淆矩阵：\n",
+      "[[30291   133]\n",
+      " [ 2018   153]]\n",
+      "阈值为 0.4 时的准确率： 0.9340082834790612\n",
+      "阈值为 0.4 时的查全率： 0.07047443574389682\n",
+      "阈值为 0.4 时的AUC： 0.5330514434833079\n",
+      "阈值为 0.4 时的F1值： 0.12454212454212456\n",
+      "--------------------\n",
+      "阈值为 0.5 时的混淆矩阵：\n",
+      "[[30341    83]\n",
+      " [ 2071   100]]\n",
+      "阈值为 0.5 时的准确率： 0.9339162448228255\n",
+      "阈值为 0.5 时的查全率： 0.046061722708429294\n",
+      "阈值为 0.5 时的AUC： 0.5216668066605518\n",
+      "阈值为 0.5 时的F1值： 0.08496176720475784\n",
+      "--------------------\n",
+      "阈值为 0.6 时的混淆矩阵：\n",
+      "[[30371    53]\n",
+      " [ 2104    67]]\n",
+      "阈值为 0.6 时的准确率： 0.93382420616659\n",
+      "阈值为 0.6 时的查全率： 0.030861354214647627\n",
+      "阈值为 0.6 时的AUC： 0.5145596542306475\n",
+      "阈值为 0.6 时的F1值： 0.058489742470536885\n",
+      "--------------------\n",
+      "阈值为 0.7 时的混淆矩阵：\n",
+      "[[30390    34]\n",
+      " [ 2128    43]]\n",
+      "阈值为 0.7 时的准确率： 0.9336708084061973\n",
+      "阈值为 0.7 时的查全率： 0.019806540764624597\n",
+      "阈值为 0.7 时的AUC： 0.5093445009897275\n",
+      "阈值为 0.7 时的F1值： 0.03825622775800712\n",
+      "--------------------\n",
+      "阈值为 0.8 时的混淆矩阵：\n",
+      "[[30403    21]\n",
+      " [ 2143    28]]\n",
+      "阈值为 0.8 时的准确率： 0.9336094493020402\n",
+      "阈值为 0.8 时的查全率： 0.012897282358360202\n",
+      "阈值为 0.8 时的AUC： 0.5061035189072896\n",
+      "阈值为 0.8 时的F1值： 0.025225225225225224\n",
+      "--------------------\n",
+      "阈值为 0.9 时的混淆矩阵：\n",
+      "[[30418     6]\n",
+      " [ 2167     4]]\n",
+      "阈值为 0.9 时的准确率： 0.9333333333333333\n",
+      "阈值为 0.9 时的查全率： 0.0018424689083371719\n",
+      "阈值为 0.9 时的AUC： 0.5008226280907713\n",
+      "阈值为 0.9 时的F1值： 0.003668042182485099\n",
+      "--------------------\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 提示：thresholds = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9]\n",
+    "# 根据predict_proba的结果和threshold的比较确定结果，再评估各种结果指标\n",
+    "import numpy as np\n",
+    "thresholds = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]\n",
+    "\n",
+    "for threshold in thresholds:\n",
+    "    # 根据阈值确定最终的分类结果\n",
+    "    y_pred_threshold = np.where(y_pred_proba > threshold, 1, 0)\n",
+    "\n",
+    "    # 计算混淆矩阵\n",
+    "    cm_threshold = confusion_matrix(y_test, y_pred_threshold)\n",
+    "    print(\"阈值为\", threshold, \"时的混淆矩阵：\")\n",
+    "    print(cm_threshold)\n",
+    "\n",
+    "    # 计算准确率\n",
+    "    accuracy_threshold = accuracy_score(y_test, y_pred_threshold)\n",
+    "    print(\"阈值为\", threshold, \"时的准确率：\", accuracy_threshold)\n",
+    "\n",
+    "    # 计算查全率\n",
+    "    recall_threshold = recall_score(y_test, y_pred_threshold)\n",
+    "    print(\"阈值为\", threshold, \"时的查全率：\", recall_threshold)\n",
+    "\n",
+    "    # 计算AUC\n",
+    "    auc_threshold = roc_auc_score(y_test, y_pred_threshold)\n",
+    "    print(\"阈值为\", threshold, \"时的AUC：\", auc_threshold)\n",
+    "\n",
+    "    # 计算F1值\n",
+    "    f1_threshold = f1_score(y_test, y_pred_threshold)\n",
+    "    print(\"阈值为\", threshold, \"时的F1值：\", f1_threshold)\n",
+    "\n",
+    "    print(\"--------------------\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 9.尝试对不同特征的重要度进行排序，通过特征选择的方式，对特征进行筛选。并重新建模，观察此时的模型准确率等评估指标。"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 720x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "feat_labels = data.columns[1:]\n",
+    "\n",
+    "# 获取特征重要性\n",
+    "feature_importance = lr.coef_[0]  # 假设model是你之前训练的逻辑回归模型\n",
+    "\n",
+    "# 将特征重要性进行排序\n",
+    "sorted_idx = feature_importance.argsort()\n",
+    "\n",
+    "# 创建水平条形图\n",
+    "plt.figure(figsize=(10, 8))\n",
+    "plt.barh(range(len(sorted_idx)), feature_importance[sorted_idx], align='center')\n",
+    "plt.yticks(range(len(sorted_idx)), np.array(feat_labels)[sorted_idx])\n",
+    "plt.xlabel('Feature Importance')\n",
+    "plt.ylabel('Features')\n",
+    "plt.title('Feature Importance Ranking')\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "特征选择后的模型准确率： 0.9339469243749041\n",
+      "特征选择后的模型查全率： 0.036849378166743434\n",
+      "特征选择后的模型AUC： 0.5174057566615994\n",
+      "特征选择后的模型F1值： 0.06917423259835712\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 可以根据逻辑回归的系数绝对值大小进行排序，也可以基于树模型的特征重要度进行排序\n",
+    "# 特征选择可以使用RFE或者selectFromModel\n",
+    "\n",
+    "from sklearn.feature_selection import SelectFromModel\n",
+    "\n",
+    "\n",
+    "# 使用 SelectFromModel 进行特征选择\n",
+    "selector = SelectFromModel(lr)\n",
+    "selector.fit(X_train_std, y_train)\n",
+    "\n",
+    "# 获取选择的特征索引\n",
+    "selected_features = selector.get_support(indices=True)\n",
+    "\n",
+    "# # 根据选择的特征重新构建训练集和测试集\n",
+    "X_train_selected = X_train_std[:, selected_features]\n",
+    "X_test_selected = X_test_std[:, selected_features]\n",
+    "\n",
+    "# 在选择的特征上重新训练模型\n",
+    "model_selected = LogisticRegression()\n",
+    "model_selected.fit(X_train_selected, y_train)\n",
+    "\n",
+    "# 在测试集上进行预测\n",
+    "y_pred_selected = model_selected.predict(X_test_selected)\n",
+    "\n",
+    "# 计算准确率\n",
+    "accuracy_selected = accuracy_score(y_test, y_pred_selected)\n",
+    "print(\"特征选择后的模型准确率：\", accuracy_selected)\n",
+    "\n",
+    "# 计算查全率\n",
+    "recall_selected = recall_score(y_test, y_pred_selected)\n",
+    "print(\"特征选择后的模型查全率：\", recall_selected)\n",
+    "\n",
+    "# 计算AUC\n",
+    "auc_selected = roc_auc_score(y_test, y_pred_selected)\n",
+    "print(\"特征选择后的模型AUC：\", auc_selected)\n",
+    "\n",
+    "# 计算F1值\n",
+    "f1_selected = f1_score(y_test, y_pred_selected)\n",
+    "print(\"特征选择后的模型F1值：\", f1_selected)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 10.其他模型算法尝试\n",
+    "使用RandomForestClassifier/SVM/KNN等sklearn分类算法进行分类，尝试上述超参数调优算法过程。"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# 随机森林\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "# 支持向量机\n",
+    "from sklearn.svm import SVC\n",
+    "# K最近邻\n",
+    "from sklearn.neighbors import KNeighborsClassifier\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n",
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Random Forest - Best Parameters: {'max_depth': 10, 'min_samples_split': 10, 'n_estimators': 200}\n",
+      "Random Forest - Best Score: 0.935492337006019\n",
+      "SVM - Best Parameters: {'C': 1, 'gamma': 1}\n",
+      "SVM - Best Score: 0.9344404476310031\n",
+      "KNN - Best Parameters: {'algorithm': 'ball_tree', 'n_neighbors': 7, 'weights': 'uniform'}\n",
+      "KNN - Best Score: 0.9333754150632625\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.model_selection import GridSearchCV\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.svm import SVC\n",
+    "from sklearn.neighbors import KNeighborsClassifier\n",
+    "\n",
+    "from sklearn.metrics import precision_score, recall_score, roc_auc_score, confusion_matrix, f1_score, roc_curve, auc\n",
+    "\n",
+    "# 定义参数网格\n",
+    "param_grid_rf = {\n",
+    "    'n_estimators': [100, 200],\n",
+    "    'max_depth': [5, 10],\n",
+    "    'min_samples_split': [5, 10]\n",
+    "}\n",
+    "\n",
+    "param_grid_svm = {\n",
+    "    'C': [0.1, 1],\n",
+    "    'gamma': [0.1,1]\n",
+    "}\n",
+    "\n",
+    "param_grid_knn = {\n",
+    "    'n_neighbors': [3, 5, 7],\n",
+    "    'weights': ['uniform', 'distance'],\n",
+    "    'algorithm': ['ball_tree', 'kd_tree']\n",
+    "}\n",
+    "\n",
+    "# 创建分类器对象\n",
+    "rf_classifier = RandomForestClassifier()\n",
+    "svm_classifier = SVC()\n",
+    "knn_classifier = KNeighborsClassifier()\n",
+    "\n",
+    "# 创建网格搜索对象\n",
+    "grid_search_rf = GridSearchCV(rf_classifier, param_grid_rf, cv=5)\n",
+    "grid_search_svm = GridSearchCV(svm_classifier, param_grid_svm, cv=5)\n",
+    "grid_search_knn = GridSearchCV(knn_classifier, param_grid_knn, cv=5)\n",
+    "\n",
+    "# 在训练集上进行网格搜索\n",
+    "grid_search_rf.fit(X_train_std, y_train)\n",
+    "grid_search_svm.fit(X_train_std, y_train)\n",
+    "grid_search_knn.fit(X_train_std, y_train)\n",
+    "\n",
+    "# 输出最佳参数组合和对应的得分\n",
+    "print(\"Random Forest - Best Parameters:\", grid_search_rf.best_params_)\n",
+    "print(\"Random Forest - Best Score:\", grid_search_rf.best_score_)\n",
+    "print(\"SVM - Best Parameters:\", grid_search_svm.best_params_)\n",
+    "print(\"SVM - Best Score:\", grid_search_svm.best_score_)\n",
+    "print(\"KNN - Best Parameters:\", grid_search_knn.best_params_)\n",
+    "print(\"KNN - Best Score:\", grid_search_knn.best_score_)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/dk/anaconda3/envs/igwn-py39/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:211: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
+      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Random Forest - Precision: 0.615071283095723\n",
+      "Random Forest - Recall: 0.13910640257945647\n",
+      "Random Forest - F1 Score: 0.22689706987227648\n",
+      "Random Forest - AUC: 0.8532556239086381\n",
+      "SVM - Precision: 0.6153846153846154\n",
+      "SVM - Recall: 0.07369875633348687\n",
+      "SVM - F1 Score: 0.1316330728095434\n",
+      "SVM - AUC: 0.6692492308612815\n",
+      "KNN - Precision: 0.5091324200913242\n",
+      "KNN - Recall: 0.10271764163979732\n",
+      "KNN - F1 Score: 0.1709467228823304\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 864x432 with 3 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "\n",
+    "\n",
+    "# 使用最佳参数进行训练\n",
+    "best_rf = grid_search_rf.best_estimator_\n",
+    "best_svm = grid_search_svm.best_estimator_\n",
+    "best_knn = grid_search_knn.best_estimator_\n",
+    "\n",
+    "best_rf.fit(X_train_std, y_train)\n",
+    "best_svm.fit(X_train_std, y_train)\n",
+    "best_knn.fit(X_train_std, y_train)\n",
+    "\n",
+    "# 预测\n",
+    "y_pred_rf = best_rf.predict(X_test_std)\n",
+    "y_pred_svm = best_svm.predict(X_test_std)\n",
+    "y_pred_knn = best_knn.predict(X_test_std)\n",
+    "\n",
+    "# 计算评估指标\n",
+    "precision_rf = precision_score(y_test, y_pred_rf)\n",
+    "precision_svm = precision_score(y_test, y_pred_svm)\n",
+    "precision_knn = precision_score(y_test, y_pred_knn)\n",
+    "\n",
+    "recall_rf = recall_score(y_test, y_pred_rf)\n",
+    "recall_svm = recall_score(y_test, y_pred_svm)\n",
+    "recall_knn = recall_score(y_test, y_pred_knn)\n",
+    "\n",
+    "f1_rf = f1_score(y_test, y_pred_rf)\n",
+    "f1_svm = f1_score(y_test, y_pred_svm)\n",
+    "f1_knn = f1_score(y_test, y_pred_knn)\n",
+    "\n",
+    "auc_rf = roc_auc_score(y_test, best_rf.predict_proba(X_test_std)[:, 1])\n",
+    "auc_svm = roc_auc_score(y_test, best_svm.decision_function(X_test_std))\n",
+    "fpr_rf, tpr_rf, thresholds_rf = roc_curve(y_test, best_rf.predict_proba(X_test_std)[:, 1])\n",
+    "fpr_svm, tpr_svm, thresholds_svm = roc_curve(y_test, best_svm.decision_function(X_test_std))\n",
+    "\n",
+    "conf_matrix_rf = confusion_matrix(y_test, y_pred_rf)\n",
+    "conf_matrix_svm = confusion_matrix(y_test, y_pred_svm)\n",
+    "conf_matrix_knn = confusion_matrix(y_test, y_pred_knn)\n",
+    "\n",
+    "# 输出评估指标\n",
+    "print(\"Random Forest - Precision:\", precision_rf)\n",
+    "print(\"Random Forest - Recall:\", recall_rf)\n",
+    "print(\"Random Forest - F1 Score:\", f1_rf)\n",
+    "print(\"Random Forest - AUC:\", auc_rf)\n",
+    "print(\"SVM - Precision:\", precision_svm)\n",
+    "print(\"SVM - Recall:\", recall_svm)\n",
+    "print(\"SVM - F1 Score:\", f1_svm)\n",
+    "print(\"SVM - AUC:\", auc_svm)\n",
+    "print(\"KNN - Precision:\", precision_knn)\n",
+    "print(\"KNN - Recall:\", recall_knn)\n",
+    "print(\"KNN - F1 Score:\", f1_knn)\n",
+    "\n",
+    "# 绘制混淆矩阵\n",
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "\n",
+    "plt.subplot(1, 3, 1)\n",
+    "sns.heatmap(conf_matrix_rf, annot=True, fmt=\"d\", cmap=\"YlGnBu\", cbar=False)\n",
+    "plt.title(\"Random Forest Confusion Matrix\")\n",
+    "plt.xlabel(\"Predicted\")\n",
+    "plt.ylabel(\"Actual\")\n",
+    "\n",
+    "plt.subplot(1, 3, 2)\n",
+    "sns.heatmap(conf_matrix_svm, annot=True, fmt=\"d\", cmap=\"YlGnBu\", cbar=False)\n",
+    "plt.title(\"SVM Confusion Matrix\")\n",
+    "plt.xlabel(\"Predicted\")\n",
+    "plt.ylabel(\"Actual\")\n",
+    "\n",
+    "plt.subplot(1, 3, 3)\n",
+    "sns.heatmap(conf_matrix_knn, annot=True, fmt=\"d\", cmap=\"YlGnBu\", cbar=False)\n",
+    "plt.title(\"KNN Confusion Matrix\")\n",
+    "plt.xlabel(\"Predicted\")\n",
+    "plt.ylabel(\"Actual\")\n",
+    "\n",
+    "plt.show()\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}