├── Train.csv
├── README.md
└── Feature Engineering.ipynb


/Train.csv:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/HanXiaoyang/Feature_Engineering_and_XGBoost_Parameter_Tuning/HEAD/Train.csv


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | ### [Data Hackathon 3.X](https://discuss.analyticsvidhya.com/t/data-hackathon-online-3-x-5th-6th-september-2015/2899)的特征工程和Xgboost模型调参例子。
2 | 为了便于阅读和交互式学习，这里用的ipython notebook格式，csv文件是此比赛的数据集。欢迎大家下载自己运行。<br>
3 | 有问题可以联系[@寒小阳](http://blog.csdn.net/han_xiaoyang)<br>
4 | 邮箱：hanxiaoyang.ml@gmail.com
5 | 


--------------------------------------------------------------------------------
/Feature Engineering.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {},
   6 |    "source": [
   7 |     "# 特征工程\n",
   8 |     "数据集来源于Data Hackathon 3.x，所有的特征处理也只做最基本的参考，可自行尝试更多的特征工程工作，参考github里Feature engineering和Kaggle Titanic的案例。"
   9 |    ]
  10 |   },
  11 |   {
  12 |    "cell_type": "markdown",
  13 |    "metadata": {},
  14 |    "source": [
  15 |     "### 加载需要的库:"
  16 |    ]
  17 |   },
  18 |   {
  19 |    "cell_type": "code",
  20 |    "execution_count": 1,
  21 |    "metadata": {
  22 |     "collapsed": true
  23 |    },
  24 |    "outputs": [],
  25 |    "source": [
  26 |     "import pandas as pd\n",
  27 |     "import numpy as np\n",
  28 |     "%matplotlib inline"
  29 |    ]
  30 |   },
  31 |   {
  32 |    "cell_type": "code",
  33 |    "execution_count": 2,
  34 |    "metadata": {
  35 |     "collapsed": false
  36 |    },
  37 |    "outputs": [],
  38 |    "source": [
  39 |     "#载入数据:\n",
  40 |     "train = pd.read_csv('Train.csv')\n",
  41 |     "test = pd.read_csv('Test.csv')"
  42 |    ]
  43 |   },
  44 |   {
  45 |    "cell_type": "code",
  46 |    "execution_count": 3,
  47 |    "metadata": {
  48 |     "collapsed": false
  49 |    },
  50 |    "outputs": [
  51 |     {
  52 |      "data": {
  53 |       "text/plain": [
  54 |        "((87020, 26), (37717, 24))"
  55 |       ]
  56 |      },
  57 |      "execution_count": 3,
  58 |      "metadata": {},
  59 |      "output_type": "execute_result"
  60 |     }
  61 |    ],
  62 |    "source": [
  63 |     "train.shape, test.shape"
  64 |    ]
  65 |   },
  66 |   {
  67 |    "cell_type": "markdown",
  68 |    "metadata": {},
  69 |    "source": [
  70 |     "### 看看数据的基本情况"
  71 |    ]
  72 |   },
  73 |   {
  74 |    "cell_type": "code",
  75 |    "execution_count": 4,
  76 |    "metadata": {
  77 |     "collapsed": false
  78 |    },
  79 |    "outputs": [
  80 |     {
  81 |      "data": {
  82 |       "text/plain": [
  83 |        "ID                        object\n",
  84 |        "Gender                    object\n",
  85 |        "City                      object\n",
  86 |        "Monthly_Income             int64\n",
  87 |        "DOB                       object\n",
  88 |        "Lead_Creation_Date        object\n",
  89 |        "Loan_Amount_Applied      float64\n",
  90 |        "Loan_Tenure_Applied      float64\n",
  91 |        "Existing_EMI             float64\n",
  92 |        "Employer_Name             object\n",
  93 |        "Salary_Account            object\n",
  94 |        "Mobile_Verified           object\n",
  95 |        "Var5                       int64\n",
  96 |        "Var1                      object\n",
  97 |        "Loan_Amount_Submitted    float64\n",
  98 |        "Loan_Tenure_Submitted    float64\n",
  99 |        "Interest_Rate            float64\n",
 100 |        "Processing_Fee           float64\n",
 101 |        "EMI_Loan_Submitted       float64\n",
 102 |        "Filled_Form               object\n",
 103 |        "Device_Type               object\n",
 104 |        "Var2                      object\n",
 105 |        "Source                    object\n",
 106 |        "Var4                       int64\n",
 107 |        "LoggedIn                   int64\n",
 108 |        "Disbursed                  int64\n",
 109 |        "dtype: object"
 110 |       ]
 111 |      },
 112 |      "execution_count": 4,
 113 |      "metadata": {},
 114 |      "output_type": "execute_result"
 115 |     }
 116 |    ],
 117 |    "source": [
 118 |     "train.dtypes"
 119 |    ]
 120 |   },
 121 |   {
 122 |    "cell_type": "markdown",
 123 |    "metadata": {},
 124 |    "source": [
 125 |     "### 拿前5条出来看看"
 126 |    ]
 127 |   },
 128 |   {
 129 |    "cell_type": "code",
 130 |    "execution_count": 5,
 131 |    "metadata": {
 132 |     "collapsed": false
 133 |    },
 134 |    "outputs": [
 135 |     {
 136 |      "data": {
 137 |       "text/html": [
 138 |        "<div>\n",
 139 |        "<table border=\"1\" class=\"dataframe\">\n",
 140 |        "  <thead>\n",
 141 |        "    <tr style=\"text-align: right;\">\n",
 142 |        "      <th></th>\n",
 143 |        "      <th>ID</th>\n",
 144 |        "      <th>Gender</th>\n",
 145 |        "      <th>City</th>\n",
 146 |        "      <th>Monthly_Income</th>\n",
 147 |        "      <th>DOB</th>\n",
 148 |        "      <th>Lead_Creation_Date</th>\n",
 149 |        "      <th>Loan_Amount_Applied</th>\n",
 150 |        "      <th>Loan_Tenure_Applied</th>\n",
 151 |        "      <th>Existing_EMI</th>\n",
 152 |        "      <th>Employer_Name</th>\n",
 153 |        "      <th>...</th>\n",
 154 |        "      <th>Interest_Rate</th>\n",
 155 |        "      <th>Processing_Fee</th>\n",
 156 |        "      <th>EMI_Loan_Submitted</th>\n",
 157 |        "      <th>Filled_Form</th>\n",
 158 |        "      <th>Device_Type</th>\n",
 159 |        "      <th>Var2</th>\n",
 160 |        "      <th>Source</th>\n",
 161 |        "      <th>Var4</th>\n",
 162 |        "      <th>LoggedIn</th>\n",
 163 |        "      <th>Disbursed</th>\n",
 164 |        "    </tr>\n",
 165 |        "  </thead>\n",
 166 |        "  <tbody>\n",
 167 |        "    <tr>\n",
 168 |        "      <th>0</th>\n",
 169 |        "      <td>ID000002C20</td>\n",
 170 |        "      <td>Female</td>\n",
 171 |        "      <td>Delhi</td>\n",
 172 |        "      <td>20000</td>\n",
 173 |        "      <td>23-May-78</td>\n",
 174 |        "      <td>15-May-15</td>\n",
 175 |        "      <td>300000.0</td>\n",
 176 |        "      <td>5.0</td>\n",
 177 |        "      <td>0.0</td>\n",
 178 |        "      <td>CYBOSOL</td>\n",
 179 |        "      <td>...</td>\n",
 180 |        "      <td>NaN</td>\n",
 181 |        "      <td>NaN</td>\n",
 182 |        "      <td>NaN</td>\n",
 183 |        "      <td>N</td>\n",
 184 |        "      <td>Web-browser</td>\n",
 185 |        "      <td>G</td>\n",
 186 |        "      <td>S122</td>\n",
 187 |        "      <td>1</td>\n",
 188 |        "      <td>0</td>\n",
 189 |        "      <td>0</td>\n",
 190 |        "    </tr>\n",
 191 |        "    <tr>\n",
 192 |        "      <th>1</th>\n",
 193 |        "      <td>ID000004E40</td>\n",
 194 |        "      <td>Male</td>\n",
 195 |        "      <td>Mumbai</td>\n",
 196 |        "      <td>35000</td>\n",
 197 |        "      <td>07-Oct-85</td>\n",
 198 |        "      <td>04-May-15</td>\n",
 199 |        "      <td>200000.0</td>\n",
 200 |        "      <td>2.0</td>\n",
 201 |        "      <td>0.0</td>\n",
 202 |        "      <td>TATA CONSULTANCY SERVICES LTD (TCS)</td>\n",
 203 |        "      <td>...</td>\n",
 204 |        "      <td>13.25</td>\n",
 205 |        "      <td>NaN</td>\n",
 206 |        "      <td>6762.9</td>\n",
 207 |        "      <td>N</td>\n",
 208 |        "      <td>Web-browser</td>\n",
 209 |        "      <td>G</td>\n",
 210 |        "      <td>S122</td>\n",
 211 |        "      <td>3</td>\n",
 212 |        "      <td>0</td>\n",
 213 |        "      <td>0</td>\n",
 214 |        "    </tr>\n",
 215 |        "    <tr>\n",
 216 |        "      <th>2</th>\n",
 217 |        "      <td>ID000007H20</td>\n",
 218 |        "      <td>Male</td>\n",
 219 |        "      <td>Panchkula</td>\n",
 220 |        "      <td>22500</td>\n",
 221 |        "      <td>10-Oct-81</td>\n",
 222 |        "      <td>19-May-15</td>\n",
 223 |        "      <td>600000.0</td>\n",
 224 |        "      <td>4.0</td>\n",
 225 |        "      <td>0.0</td>\n",
 226 |        "      <td>ALCHEMIST HOSPITALS LTD</td>\n",
 227 |        "      <td>...</td>\n",
 228 |        "      <td>NaN</td>\n",
 229 |        "      <td>NaN</td>\n",
 230 |        "      <td>NaN</td>\n",
 231 |        "      <td>N</td>\n",
 232 |        "      <td>Web-browser</td>\n",
 233 |        "      <td>B</td>\n",
 234 |        "      <td>S143</td>\n",
 235 |        "      <td>1</td>\n",
 236 |        "      <td>0</td>\n",
 237 |        "      <td>0</td>\n",
 238 |        "    </tr>\n",
 239 |        "    <tr>\n",
 240 |        "      <th>3</th>\n",
 241 |        "      <td>ID000008I30</td>\n",
 242 |        "      <td>Male</td>\n",
 243 |        "      <td>Saharsa</td>\n",
 244 |        "      <td>35000</td>\n",
 245 |        "      <td>30-Nov-87</td>\n",
 246 |        "      <td>09-May-15</td>\n",
 247 |        "      <td>1000000.0</td>\n",
 248 |        "      <td>5.0</td>\n",
 249 |        "      <td>0.0</td>\n",
 250 |        "      <td>BIHAR GOVERNMENT</td>\n",
 251 |        "      <td>...</td>\n",
 252 |        "      <td>NaN</td>\n",
 253 |        "      <td>NaN</td>\n",
 254 |        "      <td>NaN</td>\n",
 255 |        "      <td>N</td>\n",
 256 |        "      <td>Web-browser</td>\n",
 257 |        "      <td>B</td>\n",
 258 |        "      <td>S143</td>\n",
 259 |        "      <td>3</td>\n",
 260 |        "      <td>0</td>\n",
 261 |        "      <td>0</td>\n",
 262 |        "    </tr>\n",
 263 |        "    <tr>\n",
 264 |        "      <th>4</th>\n",
 265 |        "      <td>ID000009J40</td>\n",
 266 |        "      <td>Male</td>\n",
 267 |        "      <td>Bengaluru</td>\n",
 268 |        "      <td>100000</td>\n",
 269 |        "      <td>17-Feb-84</td>\n",
 270 |        "      <td>20-May-15</td>\n",
 271 |        "      <td>500000.0</td>\n",
 272 |        "      <td>2.0</td>\n",
 273 |        "      <td>25000.0</td>\n",
 274 |        "      <td>GLOBAL EDGE SOFTWARE</td>\n",
 275 |        "      <td>...</td>\n",
 276 |        "      <td>NaN</td>\n",
 277 |        "      <td>NaN</td>\n",
 278 |        "      <td>NaN</td>\n",
 279 |        "      <td>N</td>\n",
 280 |        "      <td>Web-browser</td>\n",
 281 |        "      <td>B</td>\n",
 282 |        "      <td>S134</td>\n",
 283 |        "      <td>3</td>\n",
 284 |        "      <td>1</td>\n",
 285 |        "      <td>0</td>\n",
 286 |        "    </tr>\n",
 287 |        "  </tbody>\n",
 288 |        "</table>\n",
 289 |        "<p>5 rows × 26 columns</p>\n",
 290 |        "</div>"
 291 |       ],
 292 |       "text/plain": [
 293 |        "            ID  Gender       City  Monthly_Income        DOB  \\\n",
 294 |        "0  ID000002C20  Female      Delhi           20000  23-May-78   \n",
 295 |        "1  ID000004E40    Male     Mumbai           35000  07-Oct-85   \n",
 296 |        "2  ID000007H20    Male  Panchkula           22500  10-Oct-81   \n",
 297 |        "3  ID000008I30    Male    Saharsa           35000  30-Nov-87   \n",
 298 |        "4  ID000009J40    Male  Bengaluru          100000  17-Feb-84   \n",
 299 |        "\n",
 300 |        "  Lead_Creation_Date  Loan_Amount_Applied  Loan_Tenure_Applied  Existing_EMI  \\\n",
 301 |        "0          15-May-15             300000.0                  5.0           0.0   \n",
 302 |        "1          04-May-15             200000.0                  2.0           0.0   \n",
 303 |        "2          19-May-15             600000.0                  4.0           0.0   \n",
 304 |        "3          09-May-15            1000000.0                  5.0           0.0   \n",
 305 |        "4          20-May-15             500000.0                  2.0       25000.0   \n",
 306 |        "\n",
 307 |        "                         Employer_Name    ...    Interest_Rate Processing_Fee  \\\n",
 308 |        "0                              CYBOSOL    ...              NaN            NaN   \n",
 309 |        "1  TATA CONSULTANCY SERVICES LTD (TCS)    ...            13.25            NaN   \n",
 310 |        "2              ALCHEMIST HOSPITALS LTD    ...              NaN            NaN   \n",
 311 |        "3                     BIHAR GOVERNMENT    ...              NaN            NaN   \n",
 312 |        "4                 GLOBAL EDGE SOFTWARE    ...              NaN            NaN   \n",
 313 |        "\n",
 314 |        "   EMI_Loan_Submitted Filled_Form  Device_Type  Var2  Source  Var4  LoggedIn  \\\n",
 315 |        "0                 NaN           N  Web-browser     G    S122     1         0   \n",
 316 |        "1              6762.9           N  Web-browser     G    S122     3         0   \n",
 317 |        "2                 NaN           N  Web-browser     B    S143     1         0   \n",
 318 |        "3                 NaN           N  Web-browser     B    S143     3         0   \n",
 319 |        "4                 NaN           N  Web-browser     B    S134     3         1   \n",
 320 |        "\n",
 321 |        "  Disbursed  \n",
 322 |        "0         0  \n",
 323 |        "1         0  \n",
 324 |        "2         0  \n",
 325 |        "3         0  \n",
 326 |        "4         0  \n",
 327 |        "\n",
 328 |        "[5 rows x 26 columns]"
 329 |       ]
 330 |      },
 331 |      "execution_count": 5,
 332 |      "metadata": {},
 333 |      "output_type": "execute_result"
 334 |     }
 335 |    ],
 336 |    "source": [
 337 |     "train.head(5)"
 338 |    ]
 339 |   },
 340 |   {
 341 |    "cell_type": "code",
 342 |    "execution_count": 6,
 343 |    "metadata": {
 344 |     "collapsed": false
 345 |    },
 346 |    "outputs": [
 347 |     {
 348 |      "data": {
 349 |       "text/plain": [
 350 |        "(124737, 27)"
 351 |       ]
 352 |      },
 353 |      "execution_count": 6,
 354 |      "metadata": {},
 355 |      "output_type": "execute_result"
 356 |     }
 357 |    ],
 358 |    "source": [
 359 |     "#合成一个总的data\n",
 360 |     "train['source']= 'train'\n",
 361 |     "test['source'] = 'test'\n",
 362 |     "data=pd.concat([train, test],ignore_index=True)\n",
 363 |     "data.shape"
 364 |    ]
 365 |   },
 366 |   {
 367 |    "cell_type": "markdown",
 368 |    "metadata": {},
 369 |    "source": [
 370 |     "### 数据应用/建模一个很重要的工作是，你要看看异常点，比如说缺省值"
 371 |    ]
 372 |   },
 373 |   {
 374 |    "cell_type": "code",
 375 |    "execution_count": 7,
 376 |    "metadata": {
 377 |     "collapsed": false,
 378 |     "scrolled": false
 379 |    },
 380 |    "outputs": [
 381 |     {
 382 |      "data": {
 383 |       "text/plain": [
 384 |        "City                      1401\n",
 385 |        "DOB                          0\n",
 386 |        "Device_Type                  0\n",
 387 |        "Disbursed                37717\n",
 388 |        "EMI_Loan_Submitted       84901\n",
 389 |        "Employer_Name              113\n",
 390 |        "Existing_EMI               111\n",
 391 |        "Filled_Form                  0\n",
 392 |        "Gender                       0\n",
 393 |        "ID                           0\n",
 394 |        "Interest_Rate            84901\n",
 395 |        "Lead_Creation_Date           0\n",
 396 |        "Loan_Amount_Applied        111\n",
 397 |        "Loan_Amount_Submitted    49535\n",
 398 |        "Loan_Tenure_Applied        111\n",
 399 |        "Loan_Tenure_Submitted    49535\n",
 400 |        "LoggedIn                 37717\n",
 401 |        "Mobile_Verified              0\n",
 402 |        "Monthly_Income               0\n",
 403 |        "Processing_Fee           85346\n",
 404 |        "Salary_Account           16801\n",
 405 |        "Source                       0\n",
 406 |        "Var1                         0\n",
 407 |        "Var2                         0\n",
 408 |        "Var4                         0\n",
 409 |        "Var5                         0\n",
 410 |        "source                       0\n",
 411 |        "dtype: int64"
 412 |       ]
 413 |      },
 414 |      "execution_count": 7,
 415 |      "metadata": {},
 416 |      "output_type": "execute_result"
 417 |     }
 418 |    ],
 419 |    "source": [
 420 |     "data.apply(lambda x: sum(x.isnull()))"
 421 |    ]
 422 |   },
 423 |   {
 424 |    "cell_type": "markdown",
 425 |    "metadata": {},
 426 |    "source": [
 427 |     "### 要对数据有更深的认识，比如说，咱们看看这些字段，分别有多少种取值(甚至你可以看看分布)"
 428 |    ]
 429 |   },
 430 |   {
 431 |    "cell_type": "code",
 432 |    "execution_count": 8,
 433 |    "metadata": {
 434 |     "collapsed": false,
 435 |     "scrolled": false
 436 |    },
 437 |    "outputs": [
 438 |     {
 439 |      "name": "stdout",
 440 |      "output_type": "stream",
 441 |      "text": [
 442 |       "\n",
 443 |       "Gender这一列数据的不同取值和出现的次数\n",
 444 |       "\n",
 445 |       "Male      71398\n",
 446 |       "Female    53339\n",
 447 |       "Name: Gender, dtype: int64\n",
 448 |       "\n",
 449 |       "Salary_Account这一列数据的不同取值和出现的次数\n",
 450 |       "\n",
 451 |       "HDFC Bank                                          25180\n",
 452 |       "ICICI Bank                                         19547\n",
 453 |       "State Bank of India                                17110\n",
 454 |       "Axis Bank                                          12590\n",
 455 |       "Citibank                                            3398\n",
 456 |       "Kotak Bank                                          2955\n",
 457 |       "IDBI Bank                                           2213\n",
 458 |       "Punjab National Bank                                1747\n",
 459 |       "Bank of India                                       1713\n",
 460 |       "Bank of Baroda                                      1675\n",
 461 |       "Standard Chartered Bank                             1434\n",
 462 |       "Canara Bank                                         1385\n",
 463 |       "Union Bank of India                                 1330\n",
 464 |       "Yes Bank                                            1120\n",
 465 |       "ING Vysya                                            996\n",
 466 |       "Corporation bank                                     948\n",
 467 |       "Indian Overseas Bank                                 901\n",
 468 |       "State Bank of Hyderabad                              854\n",
 469 |       "Indian Bank                                          773\n",
 470 |       "Oriental Bank of Commerce                            761\n",
 471 |       "IndusInd Bank                                        711\n",
 472 |       "Andhra Bank                                          706\n",
 473 |       "Central Bank of India                                648\n",
 474 |       "Syndicate Bank                                       614\n",
 475 |       "Bank of Maharasthra                                  576\n",
 476 |       "HSBC                                                 474\n",
 477 |       "State Bank of Bikaner & Jaipur                       448\n",
 478 |       "Karur Vysya Bank                                     435\n",
 479 |       "State Bank of Mysore                                 385\n",
 480 |       "Federal Bank                                         377\n",
 481 |       "Vijaya Bank                                          354\n",
 482 |       "Allahabad Bank                                       345\n",
 483 |       "UCO Bank                                             344\n",
 484 |       "State Bank of Travancore                             333\n",
 485 |       "Karnataka Bank                                       279\n",
 486 |       "United Bank of India                                 276\n",
 487 |       "Dena Bank                                            268\n",
 488 |       "Saraswat Bank                                        265\n",
 489 |       "State Bank of Patiala                                263\n",
 490 |       "South Indian Bank                                    223\n",
 491 |       "Deutsche Bank                                        176\n",
 492 |       "Abhyuday Co-op Bank Ltd                              161\n",
 493 |       "The Ratnakar Bank Ltd                                113\n",
 494 |       "Tamil Nadu Mercantile Bank                           103\n",
 495 |       "Punjab & Sind bank                                    84\n",
 496 |       "J&K Bank                                              78\n",
 497 |       "Lakshmi Vilas bank                                    69\n",
 498 |       "Dhanalakshmi Bank Ltd                                 66\n",
 499 |       "State Bank of Indore                                  32\n",
 500 |       "Catholic Syrian Bank                                  27\n",
 501 |       "India Bulls                                           21\n",
 502 |       "B N P Paribas                                         15\n",
 503 |       "Firstrand Bank Limited                                11\n",
 504 |       "GIC Housing Finance Ltd                               10\n",
 505 |       "Bank of Rajasthan                                      8\n",
 506 |       "Kerala Gramin Bank                                     4\n",
 507 |       "Industrial And Commercial Bank Of China Limited        3\n",
 508 |       "Ahmedabad Mercantile Cooperative Bank                  1\n",
 509 |       "Name: Salary_Account, dtype: int64\n",
 510 |       "\n",
 511 |       "Mobile_Verified这一列数据的不同取值和出现的次数\n",
 512 |       "\n",
 513 |       "Y    80928\n",
 514 |       "N    43809\n",
 515 |       "Name: Mobile_Verified, dtype: int64\n",
 516 |       "\n",
 517 |       "Var1这一列数据的不同取值和出现的次数\n",
 518 |       "\n",
 519 |       "HBXX    84901\n",
 520 |       "HBXC    12952\n",
 521 |       "HBXB     6502\n",
 522 |       "HAXA     4214\n",
 523 |       "HBXA     3042\n",
 524 |       "HAXB     2879\n",
 525 |       "HBXD     2818\n",
 526 |       "HAXC     2171\n",
 527 |       "HBXH     1387\n",
 528 |       "HCXF      990\n",
 529 |       "HAYT      710\n",
 530 |       "HAVC      570\n",
 531 |       "HAXM      386\n",
 532 |       "HCXD      348\n",
 533 |       "HCYS      318\n",
 534 |       "HVYS      252\n",
 535 |       "HAZD      161\n",
 536 |       "HCXG      114\n",
 537 |       "HAXF       22\n",
 538 |       "Name: Var1, dtype: int64\n",
 539 |       "\n",
 540 |       "Filled_Form这一列数据的不同取值和出现的次数\n",
 541 |       "\n",
 542 |       "N    96740\n",
 543 |       "Y    27997\n",
 544 |       "Name: Filled_Form, dtype: int64\n",
 545 |       "\n",
 546 |       "Device_Type这一列数据的不同取值和出现的次数\n",
 547 |       "\n",
 548 |       "Web-browser    92105\n",
 549 |       "Mobile         32632\n",
 550 |       "Name: Device_Type, dtype: int64\n",
 551 |       "\n",
 552 |       "Var2这一列数据的不同取值和出现的次数\n",
 553 |       "\n",
 554 |       "B    53481\n",
 555 |       "G    47338\n",
 556 |       "C    20366\n",
 557 |       "E     1855\n",
 558 |       "D      918\n",
 559 |       "F      770\n",
 560 |       "A        9\n",
 561 |       "Name: Var2, dtype: int64\n",
 562 |       "\n",
 563 |       "Source这一列数据的不同取值和出现的次数\n",
 564 |       "\n",
 565 |       "S122    55249\n",
 566 |       "S133    42900\n",
 567 |       "S159     7999\n",
 568 |       "S143     6140\n",
 569 |       "S127     2804\n",
 570 |       "S137     2450\n",
 571 |       "S134     1900\n",
 572 |       "S161     1109\n",
 573 |       "S151     1018\n",
 574 |       "S157      929\n",
 575 |       "S153      705\n",
 576 |       "S144      447\n",
 577 |       "S156      432\n",
 578 |       "S158      294\n",
 579 |       "S123      112\n",
 580 |       "S141       83\n",
 581 |       "S162       60\n",
 582 |       "S124       43\n",
 583 |       "S150       19\n",
 584 |       "S160       11\n",
 585 |       "S136        5\n",
 586 |       "S138        5\n",
 587 |       "S155        5\n",
 588 |       "S139        4\n",
 589 |       "S129        4\n",
 590 |       "S135        2\n",
 591 |       "S131        1\n",
 592 |       "S130        1\n",
 593 |       "S132        1\n",
 594 |       "S125        1\n",
 595 |       "S140        1\n",
 596 |       "S142        1\n",
 597 |       "S126        1\n",
 598 |       "S154        1\n",
 599 |       "Name: Source, dtype: int64\n"
 600 |      ]
 601 |     }
 602 |    ],
 603 |    "source": [
 604 |     "var = ['Gender','Salary_Account','Mobile_Verified','Var1','Filled_Form','Device_Type','Var2','Source']\n",
 605 |     "for v in var:\n",
 606 |     "    print '\\n%s这一列数据的不同取值和出现的次数\\n'%v\n",
 607 |     "    print data[v].value_counts()"
 608 |    ]
 609 |   },
 610 |   {
 611 |    "cell_type": "markdown",
 612 |    "metadata": {},
 613 |    "source": [
 614 |     "### 紧接着你就可以开始处理你的字段(特征)了\n",
 615 |     "我这里只做了一些简单的处理，你大可在我的基础上做更复杂的特征处理"
 616 |    ]
 617 |   },
 618 |   {
 619 |    "cell_type": "markdown",
 620 |    "metadata": {},
 621 |    "source": [
 622 |     "### City字段处理"
 623 |    ]
 624 |   },
 625 |   {
 626 |    "cell_type": "code",
 627 |    "execution_count": 9,
 628 |    "metadata": {
 629 |     "collapsed": false
 630 |    },
 631 |    "outputs": [
 632 |     {
 633 |      "data": {
 634 |       "text/plain": [
 635 |        "724"
 636 |       ]
 637 |      },
 638 |      "execution_count": 9,
 639 |      "metadata": {},
 640 |      "output_type": "execute_result"
 641 |     }
 642 |    ],
 643 |    "source": [
 644 |     "len(data['City'].unique())"
 645 |    ]
 646 |   },
 647 |   {
 648 |    "cell_type": "markdown",
 649 |    "metadata": {},
 650 |    "source": [
 651 |     "#### 好像city的类型好多，粗暴一点，这个字段咱们不要了"
 652 |    ]
 653 |   },
 654 |   {
 655 |    "cell_type": "code",
 656 |    "execution_count": 10,
 657 |    "metadata": {
 658 |     "collapsed": false,
 659 |     "scrolled": false
 660 |    },
 661 |    "outputs": [],
 662 |    "source": [
 663 |     "data.drop('City',axis=1,inplace=True)"
 664 |    ]
 665 |   },
 666 |   {
 667 |    "cell_type": "markdown",
 668 |    "metadata": {},
 669 |    "source": [
 670 |     "### DOB字段处理\n",
 671 |     "DOB是出生的具体日期，咱们要具体日期作用没那么大，年龄段可能对我们有用，所有算一下年龄好了"
 672 |    ]
 673 |   },
 674 |   {
 675 |    "cell_type": "code",
 676 |    "execution_count": 11,
 677 |    "metadata": {
 678 |     "collapsed": false
 679 |    },
 680 |    "outputs": [
 681 |     {
 682 |      "data": {
 683 |       "text/plain": [
 684 |        "0    23-May-78\n",
 685 |        "1    07-Oct-85\n",
 686 |        "2    10-Oct-81\n",
 687 |        "3    30-Nov-87\n",
 688 |        "4    17-Feb-84\n",
 689 |        "Name: DOB, dtype: object"
 690 |       ]
 691 |      },
 692 |      "execution_count": 11,
 693 |      "metadata": {},
 694 |      "output_type": "execute_result"
 695 |     }
 696 |    ],
 697 |    "source": [
 698 |     "data['DOB'].head()"
 699 |    ]
 700 |   },
 701 |   {
 702 |    "cell_type": "code",
 703 |    "execution_count": 12,
 704 |    "metadata": {
 705 |     "collapsed": false
 706 |    },
 707 |    "outputs": [
 708 |     {
 709 |      "data": {
 710 |       "text/plain": [
 711 |        "0    37\n",
 712 |        "1    30\n",
 713 |        "2    34\n",
 714 |        "3    28\n",
 715 |        "4    31\n",
 716 |        "Name: Age, dtype: int64"
 717 |       ]
 718 |      },
 719 |      "execution_count": 12,
 720 |      "metadata": {},
 721 |      "output_type": "execute_result"
 722 |     }
 723 |    ],
 724 |    "source": [
 725 |     "#创建一个年龄的字段Age\n",
 726 |     "data['Age'] = data['DOB'].apply(lambda x: 115 - int(x[-2:]))\n",
 727 |     "data['Age'].head()"
 728 |    ]
 729 |   },
 730 |   {
 731 |    "cell_type": "code",
 732 |    "execution_count": 13,
 733 |    "metadata": {
 734 |     "collapsed": true
 735 |    },
 736 |    "outputs": [],
 737 |    "source": [
 738 |     "#把原始的DOB字段去掉:\n",
 739 |     "data.drop('DOB',axis=1,inplace=True)"
 740 |    ]
 741 |   },
 742 |   {
 743 |    "cell_type": "markdown",
 744 |    "metadata": {},
 745 |    "source": [
 746 |     "### EMI_Load_Submitted字段处理"
 747 |    ]
 748 |   },
 749 |   {
 750 |    "cell_type": "code",
 751 |    "execution_count": 14,
 752 |    "metadata": {
 753 |     "collapsed": false
 754 |    },
 755 |    "outputs": [
 756 |     {
 757 |      "data": {
 758 |       "text/plain": [
 759 |        "<matplotlib.axes._subplots.AxesSubplot at 0x10a00d550>"
 760 |       ]
 761 |      },
 762 |      "execution_count": 14,
 763 |      "metadata": {},
 764 |      "output_type": "execute_result"
 765 |     },
 766 |     {
 767 |      "data": {
 768 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEBCAYAAACT92m7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njIODmZmVcXAwM7MyDg5mZlbm/wPLSNHVButiugAAAABJRU5ErkJggg==\n",
 769 |       "text/plain": [
 770 |        "<matplotlib.figure.Figure at 0x109fe5c50>"
 771 |       ]
 772 |      },
 773 |      "metadata": {},
 774 |      "output_type": "display_data"
 775 |     }
 776 |    ],
 777 |    "source": [
 778 |     "data.boxplot(column=['EMI_Loan_Submitted'],return_type='axes')"
 779 |    ]
 780 |   },
 781 |   {
 782 |    "cell_type": "code",
 783 |    "execution_count": 15,
 784 |    "metadata": {
 785 |     "collapsed": false
 786 |    },
 787 |    "outputs": [
 788 |     {
 789 |      "data": {
 790 |       "text/html": [
 791 |        "<div>\n",
 792 |        "<table border=\"1\" class=\"dataframe\">\n",
 793 |        "  <thead>\n",
 794 |        "    <tr style=\"text-align: right;\">\n",
 795 |        "      <th></th>\n",
 796 |        "      <th>EMI_Loan_Submitted</th>\n",
 797 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
 798 |        "    </tr>\n",
 799 |        "  </thead>\n",
 800 |        "  <tbody>\n",
 801 |        "    <tr>\n",
 802 |        "      <th>0</th>\n",
 803 |        "      <td>NaN</td>\n",
 804 |        "      <td>1</td>\n",
 805 |        "    </tr>\n",
 806 |        "    <tr>\n",
 807 |        "      <th>1</th>\n",
 808 |        "      <td>6762.90</td>\n",
 809 |        "      <td>0</td>\n",
 810 |        "    </tr>\n",
 811 |        "    <tr>\n",
 812 |        "      <th>2</th>\n",
 813 |        "      <td>NaN</td>\n",
 814 |        "      <td>1</td>\n",
 815 |        "    </tr>\n",
 816 |        "    <tr>\n",
 817 |        "      <th>3</th>\n",
 818 |        "      <td>NaN</td>\n",
 819 |        "      <td>1</td>\n",
 820 |        "    </tr>\n",
 821 |        "    <tr>\n",
 822 |        "      <th>4</th>\n",
 823 |        "      <td>NaN</td>\n",
 824 |        "      <td>1</td>\n",
 825 |        "    </tr>\n",
 826 |        "    <tr>\n",
 827 |        "      <th>5</th>\n",
 828 |        "      <td>6978.92</td>\n",
 829 |        "      <td>0</td>\n",
 830 |        "    </tr>\n",
 831 |        "    <tr>\n",
 832 |        "      <th>6</th>\n",
 833 |        "      <td>NaN</td>\n",
 834 |        "      <td>1</td>\n",
 835 |        "    </tr>\n",
 836 |        "    <tr>\n",
 837 |        "      <th>7</th>\n",
 838 |        "      <td>NaN</td>\n",
 839 |        "      <td>1</td>\n",
 840 |        "    </tr>\n",
 841 |        "    <tr>\n",
 842 |        "      <th>8</th>\n",
 843 |        "      <td>30824.65</td>\n",
 844 |        "      <td>0</td>\n",
 845 |        "    </tr>\n",
 846 |        "    <tr>\n",
 847 |        "      <th>9</th>\n",
 848 |        "      <td>10883.38</td>\n",
 849 |        "      <td>0</td>\n",
 850 |        "    </tr>\n",
 851 |        "  </tbody>\n",
 852 |        "</table>\n",
 853 |        "</div>"
 854 |       ],
 855 |       "text/plain": [
 856 |        "   EMI_Loan_Submitted  EMI_Loan_Submitted_Missing\n",
 857 |        "0                 NaN                           1\n",
 858 |        "1             6762.90                           0\n",
 859 |        "2                 NaN                           1\n",
 860 |        "3                 NaN                           1\n",
 861 |        "4                 NaN                           1\n",
 862 |        "5             6978.92                           0\n",
 863 |        "6                 NaN                           1\n",
 864 |        "7                 NaN                           1\n",
 865 |        "8            30824.65                           0\n",
 866 |        "9            10883.38                           0"
 867 |       ]
 868 |      },
 869 |      "execution_count": 15,
 870 |      "metadata": {},
 871 |      "output_type": "execute_result"
 872 |     }
 873 |    ],
 874 |    "source": [
 875 |     "#好像缺失值比较多，干脆就开一个新的字段，表明是缺失值还是不是缺失值\n",
 876 |     "data['EMI_Loan_Submitted_Missing'] = data['EMI_Loan_Submitted'].apply(lambda x: 1 if pd.isnull(x) else 0)\n",
 877 |     "data[['EMI_Loan_Submitted','EMI_Loan_Submitted_Missing']].head(10)"
 878 |    ]
 879 |   },
 880 |   {
 881 |    "cell_type": "code",
 882 |    "execution_count": 16,
 883 |    "metadata": {
 884 |     "collapsed": true
 885 |    },
 886 |    "outputs": [],
 887 |    "source": [
 888 |     "#原始那一列就可以不要了\n",
 889 |     "data.drop('EMI_Loan_Submitted',axis=1,inplace=True)"
 890 |    ]
 891 |   },
 892 |   {
 893 |    "cell_type": "markdown",
 894 |    "metadata": {},
 895 |    "source": [
 896 |     "### Employer Name字段处理"
 897 |    ]
 898 |   },
 899 |   {
 900 |    "cell_type": "markdown",
 901 |    "metadata": {},
 902 |    "source": [
 903 |     "#### 看看个数"
 904 |    ]
 905 |   },
 906 |   {
 907 |    "cell_type": "code",
 908 |    "execution_count": 17,
 909 |    "metadata": {
 910 |     "collapsed": false
 911 |    },
 912 |    "outputs": [
 913 |     {
 914 |      "data": {
 915 |       "text/plain": [
 916 |        "57193"
 917 |       ]
 918 |      },
 919 |      "execution_count": 17,
 920 |      "metadata": {},
 921 |      "output_type": "execute_result"
 922 |     }
 923 |    ],
 924 |    "source": [
 925 |     "len(data['Employer_Name'].value_counts())"
 926 |    ]
 927 |   },
 928 |   {
 929 |    "cell_type": "markdown",
 930 |    "metadata": {},
 931 |    "source": [
 932 |     "#### 不看也知道，每个人都有一个名字，太多了，懒癌晚期的同学直接drop掉了"
 933 |    ]
 934 |   },
 935 |   {
 936 |    "cell_type": "code",
 937 |    "execution_count": 18,
 938 |    "metadata": {
 939 |     "collapsed": true
 940 |    },
 941 |    "outputs": [],
 942 |    "source": [
 943 |     "#丢掉\n",
 944 |     "data.drop('Employer_Name',axis=1,inplace=True)"
 945 |    ]
 946 |   },
 947 |   {
 948 |    "cell_type": "markdown",
 949 |    "metadata": {},
 950 |    "source": [
 951 |     "### Existing_EMI字段"
 952 |    ]
 953 |   },
 954 |   {
 955 |    "cell_type": "code",
 956 |    "execution_count": 19,
 957 |    "metadata": {
 958 |     "collapsed": false
 959 |    },
 960 |    "outputs": [
 961 |     {
 962 |      "data": {
 963 |       "text/plain": [
 964 |        "<matplotlib.axes._subplots.AxesSubplot at 0x109a2c310>"
 965 |       ]
 966 |      },
 967 |      "execution_count": 19,
 968 |      "metadata": {},
 969 |      "output_type": "execute_result"
 970 |     },
 971 |     {
 972 |      "data": {
 973 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEHCAYAAABY/HZ4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEA9JREFUeJzt3H2MZXV9x/H3B1ax9WHHh0QqCNOqhEbFKSmENLa9SCuL\nNoq1FpbGdKg2m1ZMaf8QTGxgExtLbdqJ2JasXdnaBCH1KVgRaehejPUBWl3EujysWoWVWq2ydhHt\ndv32jzkze5mdhzu7s3vPuft+JRvu75zfnPlBdj6c+ZyHVBWSpG46btQLkCQdOkNckjrMEJekDjPE\nJanDDHFJ6jBDXJI67KiHeJKtSb6V5ItDzP2LJF9I8vkk9yX57tFYoyR1RY72feJJXgrsBd5XVWes\n4usuA6aq6o1HbHGS1DFH/Uy8qj4FfG9wW5KfSfLxJHcluSPJaYt86Ubg/UdlkZLUEetGvYDGFmBT\nVX0lydnA3wDnze1McgowCfzzaJYnSe008hBP8mTgF4B/SJJm8xMWTLsY+ED5jgBJepyRhzizlc73\nqurMZeZcDPz+UVqPJHXGip34MHeTJHlXkgeS7EgyNcT3TfOHqvof4GtJfmPgeGcMfD4dmKiqzw5x\nXEk6pgxzYfN64Pyldia5AHheVb0A2ARct9zBktwAfBo4Lck3klwK/BbwhuZ/Al8CXjXwJRcBNw6x\nTkk65gx1i2GSU4GPLnZLYJLrgO1VdVMz3gn0qupba71YSdLjrcUthicBDw6MdzfbJElHmI/dS1KH\nrcXdKbuB5w6MT262HSSJtwhK0iGoqiy2fdgQn7+bZBE3A28CbkpyDvDIcn24t3qrrXq9Hv1+f9TL\nkA5y4BGag60Y4s3dJD3gmUm+AVwFPBGoqtpSVbckeUWSXcCjwKVrsmrpKHvkkUdGvQRp1VYM8aq6\nZIg5l63NcqTROfHEE0e9BGnVvLApNa688spRL0FataP6Ktokvv5EklYpyZIXNj0Tlxpe1FQXGeKS\n1GHWKZLUctYpkjSmDHGpYSeuLjLEJanD7MQlqeXsxCVpTBniUsNOXF1kiEtSh9mJS1LL2YlL0pgy\nxKWGnbi6yBCXpA6zE5eklrMTl6QxZYhLDTtxdZEhLkkdZicuSS1nJy5JY8oQlxp24uoiQ1ySOsxO\nXJJazk5cksaUIS417MTVRYa4JHWYnbgktZyduCSNKUNcatiJq4sMcUnqMDtxSWo5O3FJGlOGuNSw\nE1cXGeKS1GFDhXiSDUnuTXJ/kisW2f/MJB9PsiPJPUmm13yl0hHW6/VGvQRp1Va8sJnkOOB+4Dzg\nm8BdwMVVde/AnKuAJ1XVW5M8C7gPeHZV/d+CY3lhU5JW6XAvbJ4NPFBVX6+qfcCNwKsXzPlP4KnN\n56cC/70wwKW2sxNXF60bYs5JwIMD44eYDfZB7wFuT/JN4CnARWuzPEnScoYJ8WG8Fbi7qs5N8jzg\nn5KcUVV7F06cnp5mcnISgImJCaampua7yLkzIceORzGe29aW9Tg+dsf9fp9t27YBzOflUobpxM8B\nrq6qDc34SqCq6pqBObcAf1JV/9KMbweuqKp/XXAsO3FJWqXD7cTvAp6f5NQkTwQuBm5eMGcn8CvN\nN3s2cBrw1UNfsnT0zZ0JSV2yYp1SVfuTXAbcxmzob62qnUk2ze6uLcA7gOuT3A0EeEtVffdILlyS\n5LtTJKn1fHeKJI0pQ1xq2ImriwxxSeowO3FJajk7cUkaU4a41LATVxcZ4pLUYXbiktRyduKSNKYM\ncalhJ64uMsQlqcPsxCWp5ezEJWlMGeJSw05cXWSIS1KH2YlLUsvZiUvSmDLEpYaduLrIEJekDrMT\nl6SWsxOXpDFliEsNO3F1kSEuSR1mJy5JLWcnLkljyhCXGnbi6iJDXJI6zE5cklrOTlySxpQhLjXs\nxNVFhrgkdZiduCS1nJ24JI0pQ1xq2ImriwxxSeqwoTrxJBuAGWZDf2tVXbPInB7wl8ATgG9X1bmL\nzLETl6RVWq4TXzHEkxwH3A+cB3wTuAu4uKruHZizHvg08PKq2p3kWVX1nUWOZYhL0iod7oXNs4EH\nqurrVbUPuBF49YI5lwAfrKrdAIsFuNR2duLqomFC/CTgwYHxQ822QacBz0iyPcldSV6/VguUJC1t\n3Roe50zgZcCTgc8k+UxV7Vqj40tHXK/XG/USpFUbJsR3A6cMjE9utg16CPhOVf0Q+GGSTwIvAQ4K\n8enpaSYnJwGYmJhgampq/odn7tdZx44dOz6Wx/1+n23btgHM5+VShrmweTxwH7MXNh8G7gQ2VtXO\ngTmnA9cCG4ATgM8BF1XVlxccywubaq1+vz//AyW1yXIXNlc8E6+q/UkuA27jwC2GO5Nsmt1dW6rq\n3iSfAL4I7Ae2LAxwSdLa890pktRyvjtFksaUIS415i4sSV1iiEtSh9mJS1LL2YlL0pgyxKXGzMzM\nqJcgrZohLjV27Ngx6iVIq2aIS42VHm+W2mitXoAldVK/35+/tXDz5s3z23u9no/gqxO8O0VqTE9P\nz790SGoT706RpDFliEuN6enpUS9BWjXrFElqOesUaQi+O0VdZIhLUodZp0hSy1mnSNKYMsSlhp24\nusgQl6QOsxOXpJazE5ekMWWISw07cXWRIS5JHWYnLkktZycuSWPKEJcaduLqIkNckjrMTlySWs5O\nXJLGlCEuNezE1UWGuCR1mJ24JLWcnbgkjSlDXGrYiauLDHFJ6jA7cUlqucPuxJNsSHJvkvuTXLHM\nvLOS7Evy64e6WEnS8FYM8STHAe8GzgdeCGxMcvoS8/4U+MRaL1I6GuzE1UXDnImfDTxQVV+vqn3A\njcCrF5n3ZuADwH+t4fokScsYJsRPAh4cGD/UbJuX5DnAhVX1N8CivY3Udr1eb9RLkFZtre5OmQEG\nu3KDXJKOgnVDzNkNnDIwPrnZNujngRuTBHgWcEGSfVV188KDTU9PMzk5CcDExARTU1PzZ0BznaRj\nx6MYz8zM+PfRcSvG/X6fbdu2Aczn5VJWvMUwyfHAfcB5wMPAncDGqtq5xPzrgY9W1YcW2ecthmqt\nfr8//wMltclytxiueCZeVfuTXAbcxmz9srWqdibZNLu7tiz8ksNesTQCBri6yId9JKnlfAGWNIS5\nTlLqEkNckjrMOkWSWs46RZLGlCEuNezE1UWGuCR1mJ24JLWcnbgkjSlDXGrYiauLDHFJ6jA7cUlq\nOTtxSRpThrjUsBNXFxniktRhduKS1HJ24pI0pgxxqWEnri4yxCWpw+zEJanl7MQlaUwZ4lLDTlxd\nZIhLUofZiUtSy9mJS9KYMsSlhp24usgQl6QOsxOXpJazE5ekMWWISw07cXWRIS5JHWYnLkktZycu\nSWPKEJcaduLqIkNckjrMTlySWs5OXJLG1FAhnmRDknuT3J/kikX2X5Lk7ubPp5K8eO2XKh1ZduLq\nohVDPMlxwLuB84EXAhuTnL5g2leBX6qqlwBvB96z1guVJB1sxU48yTnAVVV1QTO+EqiqumaJ+RPA\nPVX13EX22YlL0iodbid+EvDgwPihZttS3gh8fPjlSZIO1bq1PFiSc4FLgZcuNWd6eprJyUkAJiYm\nmJqaotfrAQc6SceORzGemZnx76PjVoz7/T7btm0DmM/LpQxbp1xdVRua8aJ1SpIzgA8CG6rqK0sc\nyzpFrdXv9+d/oKQ2Wa5OGSbEjwfuA84DHgbuBDZW1c6BOacAtwOvr6rPLnMsQ1ySVmm5EF+xTqmq\n/UkuA25jtkPfWlU7k2ya3V1bgD8GngH8dZIA+6rq7LX7V5AkLcYnNqWGdYrayic2JWlMeSYuSS3n\nmbgkjSlDXGrM3acrdYkhLkkdZicuSS1nJy5JY8oQlxp24uoiQ1ySOsxOXJJazk5cksaUIS417MTV\nRYa4JHWYnbgktZyduCSNKUNcatiJq4sMcUnqMDtxSWo5O3FJGlOGuNSwE1cXGeKS1GF24pLUcnbi\nkjSmDHGpMTMzM+olSKtmiEuNHTt2jHoJ0qoZ4lJjcnJy1EuQVm3dqBcgjVK/35+/tXDz5s3z23u9\nHr1ebzSLklbBu1OkxoYNG7j11ltHvQzpIMvdnWKIS43169ezZ8+eUS9DOoi3GEpDeOyxx0a9BGnV\n7MR1TBvsxPft28fVV18N2ImrO6xTdEx7zWtew/bt2wHYs2cP69evB+Dcc8/lwx/+8CiXJs2zE5eG\n0PygjHoZ0kGWC3HrFB3TZmZm+MhHPjI/nqtQLrzwQi6//PIRrUoaniGuY9odd9zxuCc15z4//elP\nN8TVCUPdnZJkQ5J7k9yf5Iol5rwryQNJdiSZWttlSkfGrl272Lt3L3v37gWY/7xr164Rr0wazooh\nnuQ44N3A+cALgY1JTl8w5wLgeVX1AmATcN0RWKu05tavX8+6detYt272l9K5z3MXOKW2W/HCZpJz\ngKuq6oJmfCVQVXXNwJzrgO1VdVMz3gn0qupbC47lhU21SrLotSIAL3KqNQ73YZ+TgAcHxg8125ab\ns3uROZKkNeYTm5LUYcPcnbIbOGVgfHKzbeGc564wB1j+11epTfy7qi4YJsTvAp6f5FTgYeBiYOOC\nOTcDbwJuajr0Rxb24cCSnY4k6dCsGOJVtT/JZcBtzNYvW6tqZ5JNs7trS1XdkuQVSXYBjwKXHtll\nS5LgKD92L0laW17YlKQOM8TVKkn2J/l8ki80/3zLCvP/McnTltn/B0meNOz8Q1jvLyd5ZMGaX9bs\n+3GS9w3MPT7Jt5Pc3Ix/O8m1a7UWHZt8d4ra5tGqOnPYyVX1aytMuRz4e+CHQ84/FJ+sqlctsv1R\n4EVJTqiqHwG/yuOfpwCwz9Rh8UxcbXPQHUxJnta8u+cFzfiGJG9oPn8tyTOS/GRzlv2FJF9M8rok\nbwaeA2xPcvuC+acm+XKSLUm+lOTWJCc0c85KcndzVv1nSe5Z7ZoH3AK8svm8EXj/qv5rSCswxNU2\nP7GgmnhdVX2f2VtY/y7JRcBEVW1t5s+dyW4AdlfVz1XVGcCtVXUts88r9KrqvAXzAZ4PXFtVLwL2\nAK9ttr8X+N3mN4L9rHy2/IsL1vzTA9/rRmbfN3QCcAbwudX/J5GWZp2itvnBYnVKVd2e5DeBvwJe\nPLBr7iz4HuDPk7wD+FhVfWpgfxaZD/C1qpo7y/43YDLJeuApVXVns/0GDpxJL2WpOoWq+lKSSWbP\nwj/G8mft0qp5Jq5OyOzjkz/LbM/8zIX7q+oB4Exmw/ztSd42xGF/NPB5PwdOatY6aG8G3olVio4A\nQ1xts1SA/hHwZeAS4Pokxz/ui5KfAh6rqhuYDcy5s/nvA0vdjXLQ96qqPcD3k5zVbLr4MNY8t/29\nwOaq+vchjiWtinWK2uZJST7PbAAWcCuwDfgd4Kyq+kGSO4C3AZs50Fe/GHhnkh8D/wv8XrP9PcCt\nSXY3vfhgv71U1/1G4G+T7AfuYLYvX85LF6z57VX1obnjV9VuZt/JL605n9iUFkjy5Kp6tPl8BXBi\nVf3hiJclLcozcelgr0zyVmZ/Pv4DmB7paqRleCYuDSHJy4FrOFDBBPhqVb126a+SjjxDXJI6zLtT\nJKnDDHFJ6jBDXJI6zBCXpA4zxCWpw/4fn10SfexcgdUAAAAASUVORK5CYII=\n",
 974 |       "text/plain": [
 975 |        "<matplotlib.figure.Figure at 0x108aa9750>"
 976 |       ]
 977 |      },
 978 |      "metadata": {},
 979 |      "output_type": "display_data"
 980 |     }
 981 |    ],
 982 |    "source": [
 983 |     "data.boxplot(column='Existing_EMI',return_type='axes')"
 984 |    ]
 985 |   },
 986 |   {
 987 |    "cell_type": "code",
 988 |    "execution_count": 21,
 989 |    "metadata": {
 990 |     "collapsed": false
 991 |    },
 992 |    "outputs": [
 993 |     {
 994 |      "data": {
 995 |       "text/plain": [
 996 |        "count    1.246260e+05\n",
 997 |        "mean     3.636342e+03\n",
 998 |        "std      3.369124e+04\n",
 999 |        "min      0.000000e+00\n",
1000 |        "25%               NaN\n",
1001 |        "50%               NaN\n",
1002 |        "75%               NaN\n",
1003 |        "max      1.000000e+07\n",
1004 |        "Name: Existing_EMI, dtype: float64"
1005 |       ]
1006 |      },
1007 |      "execution_count": 21,
1008 |      "metadata": {},
1009 |      "output_type": "execute_result"
1010 |     }
1011 |    ],
1012 |    "source": [
1013 |     "data['Existing_EMI'].describe()"
1014 |    ]
1015 |   },
1016 |   {
1017 |    "cell_type": "code",
1018 |    "execution_count": 22,
1019 |    "metadata": {
1020 |     "collapsed": false
1021 |    },
1022 |    "outputs": [],
1023 |    "source": [
1024 |     "#缺省值不多，用均值代替\n",
1025 |     "data['Existing_EMI'].fillna(0, inplace=True)"
1026 |    ]
1027 |   },
1028 |   {
1029 |    "cell_type": "markdown",
1030 |    "metadata": {},
1031 |    "source": [
1032 |     "### Interest_Rate字段:"
1033 |    ]
1034 |   },
1035 |   {
1036 |    "cell_type": "code",
1037 |    "execution_count": 23,
1038 |    "metadata": {
1039 |     "collapsed": false
1040 |    },
1041 |    "outputs": [
1042 |     {
1043 |      "data": {
1044 |       "text/plain": [
1045 |        "<matplotlib.axes._subplots.AxesSubplot at 0x10a05f250>"
1046 |       ]
1047 |      },
1048 |      "execution_count": 23,
1049 |      "metadata": {},
1050 |      "output_type": "execute_result"
1051 |     },
1052 |     {
1053 |      "data": {
1054 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW0AAAEBCAYAAACzN/QDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEJhJREFUeJzt3W+MZXV9x/H3h8VQKGVnq8LSQp1gWqJJdbTSP5KWaauV\nWgPUKG3kwY7yQNuoiNaKTZtdTWMUGxrbxCeizkhqFUlTQaMQYS6JEKnCrlAgPJE1Me2u/7CiNI2B\nbx/M2WWYndm5d3Zm7jn3vF/JZc7v3HPu/d5l+XLmc8753VQVkqRuOGncBUiShmfTlqQOsWlLUofY\ntCWpQ2zaktQhNm1J6pChm3aSk5Lcl+TmZrwryW1JHklya5KdW1emJAlGO9K+Cnho2fga4CtVdT5w\nB/DezSxMknSsoZp2knOAVwPXL1t9KbDQLC8Al21uaZKklYY90v5H4N3A8tsnz6qqwwBVdQg4c5Nr\nkyStsG7TTvInwOGqOgDkOJt6P7wkbbGTh9jmQuCSJK8GTgV+IckNwKEkZ1XV4SS7ge+utnMSm7kk\nbUBVHXOgnFEmjEpyEfCuqrokybXAD6rqQ0neA+yqqmtW2aeclEpttG/fPvbt2zfuMqRVJVm1aZ/I\nddofBF6Z5BHgD5ux1BkHDx4cdwnSyIaJR46qqjuBO5vlHwKv2IqiJEmr845I9dbc3Ny4S5BGNlKm\nvaE3MNOWpJFtRaYtddpgMBh3CdLIbNqS1CHGI5LUQsYjkjQBbNrqLTNtdZFNW5I6xExbklrITFuS\nJoBNW71lpq0usmlLUoeYaUtSC5lpS9IEsGmrt8y01UXGI+qt5tfPcZchrcp4RJImgE1bkjpkpK8b\nk7ouyZpjoxJ1gU1bvbK8MZtpq4uMRySpQ2zaktQhNm311uLi4rhLkEZm05akDvHmGklqIW+ukaQJ\nYNNWb628ZlvqApu2JHXIuk07ySlJ7kmyP8mDST7QrN+b5DtJ7mseF299uZLUb0OdiExyWlU9kWQH\ncBfwLuAVwONVdd06+3oiUq1xvEjEv6dqk7VORA51G3tVPdEsnsLS0fljR153c8qTtoe3savrhsq0\nk5yUZD9wCBhU1UPNU29NciDJ9Ul2blmVkiRgyKZdVU9V1UuAc4DfS3IR8FHgvKqaYamZHzcmkSSd\nuJFm+auqHyf5IvCyqrpz2VMfA25Za7+5uTmmp6cBmJqaYmZmhtnZWeDpr3xy7Hi7x1XVqnoc93s8\nGAyYn58HONovV7PuicgkzwF+VlX/k+RU4FbgfcCDVXWo2eZq4IKqesMq+3siUq00GAyO/scjtc2J\nnIg8G1jI0mn3k4Abqur2JJ9KMgM8BRwE3ryZBUuSjuXcI5LUQs49IkkTwKat3jpyEkjqEpu2JHWI\nmbYktZCZtiRNAJu2estMW11k05akDjHTlqQWMtOWVvDrxtRFNm1J6hCbtiR1yEhTs0pdtzISWT72\n3Iu6wKatXvHrxtR1xiOS1CEeaatXjEfUdTZt9YrxiLrOeESSOsSmLUkdYtNWby0uLo67BGlkzj2i\n3jLTVps594gkTQCbtiR1iJf8qVe8TltdZ9NWr3idtrrOeESSOsSmLUkdYjyiXjHTVtfZtNUrZtrq\nOuMRSeqQdZt2klOS3JNkf5IHk3ygWb8ryW1JHklya5KdW1+udGKSHH2sNpbabqjb2JOcVlVPJNkB\n3AW8C7gE+EFVXZvkPcCuqrpmlX29jV2tZDyiNjuh29ir6olm8ZRmn8eAS4GFZv0CcNkm1ClJOo6h\nmnaSk5LsBw4Bg6p6CDirqg4DVNUh4MytK1OSBENePVJVTwEvSXIGcGuSWWDl75Vr/p45NzfH9PQ0\nAFNTU8zMzDA7OwvAYDAAcOx428eLi4utqsdxv8eDwYD5+XmAo/1yNSNPzZrk74D/Ba4EZqvqcJLd\nwGJVvWCV7c201Upm2mqzDWfaSZ5z5MqQJKcCrwT2AzcDc81me4DPb1q1kqRVDROPnA0sZOmaqJOA\nG6rq9ibjvjHJm4BvA5dvYZ2SJPzmGvXM8a7H9u+p2mSteMTb2NUr3saurvM2dknqEJu2JHWI8Yh6\nxalZ1XU2bfWKmba6znhEkjrEI231ivGIus6mrV4xHlHXGY9IUofYtCWpQ2za6pUdO3as+nVjO3bs\nGHNl0nDMtNUrTz755NFlM211kUfaktQhNm1J6hDjEfWK12mr62za6hWv01bXGY9IUod4pK2Jcbxv\npdnMfTw61zjZtDUxRm2mCdh/1TXGI+qxwbgLkEZm05akDrFpq7f27p0ddwnSyLLVJ1WSlCduJGk0\nzSWpx5wp90hbvTUYDMZdgjQym7YkdYjxiCS1kPGIJE0Am7Z6a25uMO4SpJGt27STnJPkjiQPJnkg\nydua9XuTfCfJfc3j4q0vV9o8CwvjrkAa3bqZdpLdwO6qOpDkdOBe4FLgz4DHq+q6dfY301YreRu7\n2mytTHvduUeq6hBwqFn+SZKHgV8+8rqbWqUk6bhGyrSTTAMzwD3NqrcmOZDk+iQ7N7k2aYsNxl2A\nNLKhZ/lropGbgKuaI+6PAu+vqkry98B1wJWr7Ts3N8f09DQAU1NTzMzMMDs7Czx9g4Njx44d93k8\nGAyYn58HONovVzPUddpJTga+AHypqj6yyvPPA26pqhet8pyZtlpp376lh9RGa2XawzbtTwHfr6p3\nLlu3u8m7SXI1cEFVvWGVfW3akjSiDd9ck+RC4ArgD5LsX3Z537VJ7k9yALgIuHrTq5a20JFfTaUu\nGebqkbuAHas89eXNL0eSdDzOPSJJLeTcI5I0AWza6i3nHlEXGY+ot5IBVbPjLkNa1Qld8neCb2zT\nVis594jazExbkiaATVs9Nhh3AdLIbNqS1CE2bfXW3r2z4y5BGpknIiWphTwRKa3g3CPqIpu2JHWI\n8YgktZDxiCRNAJu2esu5R9RFxiPqLeceUZs594i0gnOPqM3MtCVpAti01WODcRcgjcymLUkdYtNW\nbzn3iLrIE5GS1EKeiJRWcO4RdZFNW5I6xHhEklrIeESSJoBNW73l3CPqIuMR9ZZzj6jNNhyPJDkn\nyR1JHkzyQJK3N+t3JbktySNJbk2ycysKl7bO7LgLkEa27pF2kt3A7qo6kOR04F7gUuCNwA+q6tok\n7wF2VdU1q+zvkbZayQmj1GYbPtKuqkNVdaBZ/gnwMHAOS417odlsAbhs88qVtsNg3AVIIxvpRGSS\naWAG+BpwVlUdhqXGDpy52cVJkp7p5GE3bKKRm4CrquonSVb+YrnmL5pzc3NMT08DMDU1xczMDLOz\ns8DTd6U5drx8/NrXzvLYY/D00fBs83Mzx7MkW/n6cPrpA265Zfx/no7bPx4MBszPzwMc7ZerGerq\nkSQnA18AvlRVH2nWPQzMVtXhJvderKoXrLKvmbZGNil586R8Dm2/E7255hPAQ0caduNmYK5Z3gN8\n/oQqlLbZkaMcqUvWjUeSXAhcATyQZD9LMcjfAB8CbkzyJuDbwOVbWagkyZtr1FKTEitMyufQ9nPu\nEUmaADZt9ZaZtrrIpi1JHWKmrVaalCx4Uj6Htp+ZtiRNAJu2estMW11k05akDjHTVitNShY8KZ9D\n289MW5ImgE1bvWWmrS4aempWaTsVgWN+MeyeWvZPaTOYaauVJiULnpTPoe1npi1JE8Cmrd4y01YX\n2bQlqUPMtNVKk5IFT8rn0PYz05akCWDTVm+ZaauLbNqS1CFm2mqlScmCJ+VzaPuZaUvSBLBpq7fM\ntNVFNm1J6hAzbbXSpGTBk/I5tP3MtCVpAti01Vtm2uoim7YkdYiZtlppUrLgSfkc2n4bzrSTfDzJ\n4ST3L1u3N8l3ktzXPC7e7IIlSccaJh75JPCqVdZfV1UvbR5f3uS6pC1npq0uWrdpV9VXgcdWeWoC\nvsFPkrrlRE5EvjXJgSTXJ9m5aRVJ22R2dnbcJUgj22jT/ihwXlXNAIeA6zavJGlJ0v3Hrl3j/lPU\npDl5IztV1feWDT8G3HK87efm5piengZgamqKmZmZo0c5R3JFx46Xj6u2/v2SAYuLbPnnGQzG/+fp\nuP3jwWDA/Pw8wNF+uZqhLvlLMg3cUlW/3ox3V9WhZvlq4IKqesMa+3rJn1opGRz9n4PUNmtd8rdu\n007yaWAWeDZwGNgL/D4wAzwFHATeXFWH19jfpq1W8hpqtdmGm/YmvLFNW61k01abOWGUdIzBuAuQ\nRmbTVm/t2TPuCqTRGY9IUgsZj0jSBLBpq7eOXCMrdYlNW5I6xExbklrITFtaYd++cVcgjc4jbfWW\nt7GrzTzSlqQJ4JG2esvb2NVmHmlL0gSwaavHBuMuQBqZTVu95dwj6iIzbUlqITNtSZoANm31lnOP\nqIts2pLUIRv6NnapjZJj4r8t4TkajZNNWxPDZqo+MB5Rb5lpq4ts2pLUIV6nLUkt5HXakjQBbNrq\nLTNtdZFNW5I6xExbklrITFuSJsC6TTvJx5McTnL/snW7ktyW5JEktybZubVlSpvPTFtdNMyR9ieB\nV61Ydw3wlao6H7gDeO9mFyZttQMHDoy7BGlk6zbtqvoq8NiK1ZcCC83yAnDZJtclbbkf/ehH4y5B\nGtlGM+0zq+owQFUdAs7cvJIkSWvZrBORXh6izjl48OC4S5BGttFZ/g4nOauqDifZDXz3eBtv15SZ\n0qgWFhbW30hqkWGbdprHETcDc8CHgD3A59facbXrDCVJG7PuzTVJPg3MAs8GDgN7gX8HPgecC3wb\nuLyqPKsjSVtsy++IlCRtHu+IlKQOsWlrbJI8PsQ2VyX5uS2u48VJ/nidbfYk+W6S+5I8mOQvh3jd\ni5L8zuZVKtm0NV7DZHPvAE4b5UWTjPr3egZ49RDbfaaqXgr8LrA3yXPX2X4WePmItUjHZdPW2DVH\npItJPpfk4SQ3NOvfBvwSsJjk9mbdHyW5O8k3knw2yWnN+keTfDDJN4DXJTkvyZeSfD3JnUl+rdnu\n9UkeSLI/ySDJs4D3A5c3R9GvX6/eqvoh8C1gunnN1yT5WpJ7mzl5npvkecBbgHc0r3thkuckuSnJ\nPc3Dhq7RVZUPH2N5AD9ufl7E0lQJZ7N0aendwMub574F7GqWnw3cCZzajP8a+Ntm+VHgr5a99leA\n5zfLvwnc3izfD5zdLJ/R/NwD/NM6tR7dBvgV4BDwi81457LtrgQ+3CzvBd657Ll/Wfa5zgUeGve/\nAx/de2z05hpps/1HVf03QJIDLB3F3s0z7xH4beCFwF1ZumPrWc02R3y22f/nWYolPpen7+x6VvPz\nLmAhyY3Av41Y458nuQg4H3h3LR1xA5zbvN7Zzfs8usb+rwBesKym05OcVlVPjFiHesymrbb4v2XL\nT7L6380At1XVFWu8xk+bnycBj9VS/vwMVfUXSS4AXgPcm+SYbY7jM1X19iS/AdyY5BNV9VPgn4F/\nqKovNk197xr7B/itqvrZCO8pPYOZtsZpmLtlfwyc0Sx/DbgwyfMBkpyW5FdX7lBVjwOPJnnd0TdK\nXtT8PK+qvl5Ve1mafuFc4PFl77GuqrqXpbuCr2pWnQH8V7O8Z9mmK1/3tmX7kOTFw76ndIRNW+O0\n1tUjy9d/DPhyktur6vvAG4F/TfJNlqKR89d4rSuAK5McSPKfwCXN+g8nub/5Uo+7q+p+YBF44bAn\nIhvXAm9pToS+D7gpydeB7y3b5hbgT4+ciATeDrwsyTebmt485HtJR3lHpCR1iEfaktQhnoiUlkky\nx1LuvPxX0Luq6m3jqUh6JuMRSeoQ4xFJ6hCbtiR1iE1bkjrEpi1JHWLTlqQO+X+2nl4VRBTAYgAA\nAABJRU5ErkJggg==\n",
1055 |       "text/plain": [
1056 |        "<matplotlib.figure.Figure at 0x10a053d10>"
1057 |       ]
1058 |      },
1059 |      "metadata": {},
1060 |      "output_type": "display_data"
1061 |     }
1062 |    ],
1063 |    "source": [
1064 |     "data.boxplot(column=['Interest_Rate'],return_type='axes')"
1065 |    ]
1066 |   },
1067 |   {
1068 |    "cell_type": "code",
1069 |    "execution_count": 24,
1070 |    "metadata": {
1071 |     "collapsed": false
1072 |    },
1073 |    "outputs": [
1074 |     {
1075 |      "name": "stdout",
1076 |      "output_type": "stream",
1077 |      "text": [
1078 |       "   Interest_Rate  Interest_Rate_Missing\n",
1079 |       "0            NaN                      1\n",
1080 |       "1          13.25                      0\n",
1081 |       "2            NaN                      1\n",
1082 |       "3            NaN                      1\n",
1083 |       "4            NaN                      1\n",
1084 |       "5          13.99                      0\n",
1085 |       "6            NaN                      1\n",
1086 |       "7            NaN                      1\n",
1087 |       "8          14.85                      0\n",
1088 |       "9          18.25                      0\n"
1089 |      ]
1090 |     }
1091 |    ],
1092 |    "source": [
1093 |     "#缺省值太多，也造一个字段，表示有无\n",
1094 |     "data['Interest_Rate_Missing'] = data['Interest_Rate'].apply(lambda x: 1 if pd.isnull(x) else 0)\n",
1095 |     "print data[['Interest_Rate','Interest_Rate_Missing']].head(10)"
1096 |    ]
1097 |   },
1098 |   {
1099 |    "cell_type": "code",
1100 |    "execution_count": 25,
1101 |    "metadata": {
1102 |     "collapsed": true
1103 |    },
1104 |    "outputs": [],
1105 |    "source": [
1106 |     "data.drop('Interest_Rate',axis=1,inplace=True)"
1107 |    ]
1108 |   },
1109 |   {
1110 |    "cell_type": "markdown",
1111 |    "metadata": {},
1112 |    "source": [
1113 |     "### Lead Creation Date字段"
1114 |    ]
1115 |   },
1116 |   {
1117 |    "cell_type": "code",
1118 |    "execution_count": 26,
1119 |    "metadata": {
1120 |     "collapsed": false
1121 |    },
1122 |    "outputs": [
1123 |     {
1124 |      "data": {
1125 |       "text/html": [
1126 |        "<div>\n",
1127 |        "<table border=\"1\" class=\"dataframe\">\n",
1128 |        "  <thead>\n",
1129 |        "    <tr style=\"text-align: right;\">\n",
1130 |        "      <th></th>\n",
1131 |        "      <th>Device_Type</th>\n",
1132 |        "      <th>Disbursed</th>\n",
1133 |        "      <th>Existing_EMI</th>\n",
1134 |        "      <th>Filled_Form</th>\n",
1135 |        "      <th>Gender</th>\n",
1136 |        "      <th>ID</th>\n",
1137 |        "      <th>Loan_Amount_Applied</th>\n",
1138 |        "      <th>Loan_Amount_Submitted</th>\n",
1139 |        "      <th>Loan_Tenure_Applied</th>\n",
1140 |        "      <th>Loan_Tenure_Submitted</th>\n",
1141 |        "      <th>...</th>\n",
1142 |        "      <th>Salary_Account</th>\n",
1143 |        "      <th>Source</th>\n",
1144 |        "      <th>Var1</th>\n",
1145 |        "      <th>Var2</th>\n",
1146 |        "      <th>Var4</th>\n",
1147 |        "      <th>Var5</th>\n",
1148 |        "      <th>source</th>\n",
1149 |        "      <th>Age</th>\n",
1150 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
1151 |        "      <th>Interest_Rate_Missing</th>\n",
1152 |        "    </tr>\n",
1153 |        "  </thead>\n",
1154 |        "  <tbody>\n",
1155 |        "    <tr>\n",
1156 |        "      <th>0</th>\n",
1157 |        "      <td>Web-browser</td>\n",
1158 |        "      <td>0.0</td>\n",
1159 |        "      <td>0.0</td>\n",
1160 |        "      <td>N</td>\n",
1161 |        "      <td>Female</td>\n",
1162 |        "      <td>ID000002C20</td>\n",
1163 |        "      <td>300000.0</td>\n",
1164 |        "      <td>NaN</td>\n",
1165 |        "      <td>5.0</td>\n",
1166 |        "      <td>NaN</td>\n",
1167 |        "      <td>...</td>\n",
1168 |        "      <td>HDFC Bank</td>\n",
1169 |        "      <td>S122</td>\n",
1170 |        "      <td>HBXX</td>\n",
1171 |        "      <td>G</td>\n",
1172 |        "      <td>1</td>\n",
1173 |        "      <td>0</td>\n",
1174 |        "      <td>train</td>\n",
1175 |        "      <td>37</td>\n",
1176 |        "      <td>1</td>\n",
1177 |        "      <td>1</td>\n",
1178 |        "    </tr>\n",
1179 |        "    <tr>\n",
1180 |        "      <th>1</th>\n",
1181 |        "      <td>Web-browser</td>\n",
1182 |        "      <td>0.0</td>\n",
1183 |        "      <td>0.0</td>\n",
1184 |        "      <td>N</td>\n",
1185 |        "      <td>Male</td>\n",
1186 |        "      <td>ID000004E40</td>\n",
1187 |        "      <td>200000.0</td>\n",
1188 |        "      <td>200000.0</td>\n",
1189 |        "      <td>2.0</td>\n",
1190 |        "      <td>2.0</td>\n",
1191 |        "      <td>...</td>\n",
1192 |        "      <td>ICICI Bank</td>\n",
1193 |        "      <td>S122</td>\n",
1194 |        "      <td>HBXA</td>\n",
1195 |        "      <td>G</td>\n",
1196 |        "      <td>3</td>\n",
1197 |        "      <td>13</td>\n",
1198 |        "      <td>train</td>\n",
1199 |        "      <td>30</td>\n",
1200 |        "      <td>0</td>\n",
1201 |        "      <td>0</td>\n",
1202 |        "    </tr>\n",
1203 |        "    <tr>\n",
1204 |        "      <th>2</th>\n",
1205 |        "      <td>Web-browser</td>\n",
1206 |        "      <td>0.0</td>\n",
1207 |        "      <td>0.0</td>\n",
1208 |        "      <td>N</td>\n",
1209 |        "      <td>Male</td>\n",
1210 |        "      <td>ID000007H20</td>\n",
1211 |        "      <td>600000.0</td>\n",
1212 |        "      <td>450000.0</td>\n",
1213 |        "      <td>4.0</td>\n",
1214 |        "      <td>4.0</td>\n",
1215 |        "      <td>...</td>\n",
1216 |        "      <td>State Bank of India</td>\n",
1217 |        "      <td>S143</td>\n",
1218 |        "      <td>HBXX</td>\n",
1219 |        "      <td>B</td>\n",
1220 |        "      <td>1</td>\n",
1221 |        "      <td>0</td>\n",
1222 |        "      <td>train</td>\n",
1223 |        "      <td>34</td>\n",
1224 |        "      <td>1</td>\n",
1225 |        "      <td>1</td>\n",
1226 |        "    </tr>\n",
1227 |        "    <tr>\n",
1228 |        "      <th>3</th>\n",
1229 |        "      <td>Web-browser</td>\n",
1230 |        "      <td>0.0</td>\n",
1231 |        "      <td>0.0</td>\n",
1232 |        "      <td>N</td>\n",
1233 |        "      <td>Male</td>\n",
1234 |        "      <td>ID000008I30</td>\n",
1235 |        "      <td>1000000.0</td>\n",
1236 |        "      <td>920000.0</td>\n",
1237 |        "      <td>5.0</td>\n",
1238 |        "      <td>5.0</td>\n",
1239 |        "      <td>...</td>\n",
1240 |        "      <td>State Bank of India</td>\n",
1241 |        "      <td>S143</td>\n",
1242 |        "      <td>HBXX</td>\n",
1243 |        "      <td>B</td>\n",
1244 |        "      <td>3</td>\n",
1245 |        "      <td>10</td>\n",
1246 |        "      <td>train</td>\n",
1247 |        "      <td>28</td>\n",
1248 |        "      <td>1</td>\n",
1249 |        "      <td>1</td>\n",
1250 |        "    </tr>\n",
1251 |        "    <tr>\n",
1252 |        "      <th>4</th>\n",
1253 |        "      <td>Web-browser</td>\n",
1254 |        "      <td>0.0</td>\n",
1255 |        "      <td>25000.0</td>\n",
1256 |        "      <td>N</td>\n",
1257 |        "      <td>Male</td>\n",
1258 |        "      <td>ID000009J40</td>\n",
1259 |        "      <td>500000.0</td>\n",
1260 |        "      <td>500000.0</td>\n",
1261 |        "      <td>2.0</td>\n",
1262 |        "      <td>2.0</td>\n",
1263 |        "      <td>...</td>\n",
1264 |        "      <td>HDFC Bank</td>\n",
1265 |        "      <td>S134</td>\n",
1266 |        "      <td>HBXX</td>\n",
1267 |        "      <td>B</td>\n",
1268 |        "      <td>3</td>\n",
1269 |        "      <td>17</td>\n",
1270 |        "      <td>train</td>\n",
1271 |        "      <td>31</td>\n",
1272 |        "      <td>1</td>\n",
1273 |        "      <td>1</td>\n",
1274 |        "    </tr>\n",
1275 |        "  </tbody>\n",
1276 |        "</table>\n",
1277 |        "<p>5 rows × 24 columns</p>\n",
1278 |        "</div>"
1279 |       ],
1280 |       "text/plain": [
1281 |        "   Device_Type  Disbursed  Existing_EMI Filled_Form  Gender           ID  \\\n",
1282 |        "0  Web-browser        0.0           0.0           N  Female  ID000002C20   \n",
1283 |        "1  Web-browser        0.0           0.0           N    Male  ID000004E40   \n",
1284 |        "2  Web-browser        0.0           0.0           N    Male  ID000007H20   \n",
1285 |        "3  Web-browser        0.0           0.0           N    Male  ID000008I30   \n",
1286 |        "4  Web-browser        0.0       25000.0           N    Male  ID000009J40   \n",
1287 |        "\n",
1288 |        "   Loan_Amount_Applied  Loan_Amount_Submitted  Loan_Tenure_Applied  \\\n",
1289 |        "0             300000.0                    NaN                  5.0   \n",
1290 |        "1             200000.0               200000.0                  2.0   \n",
1291 |        "2             600000.0               450000.0                  4.0   \n",
1292 |        "3            1000000.0               920000.0                  5.0   \n",
1293 |        "4             500000.0               500000.0                  2.0   \n",
1294 |        "\n",
1295 |        "   Loan_Tenure_Submitted          ...                 Salary_Account Source  \\\n",
1296 |        "0                    NaN          ...                      HDFC Bank   S122   \n",
1297 |        "1                    2.0          ...                     ICICI Bank   S122   \n",
1298 |        "2                    4.0          ...            State Bank of India   S143   \n",
1299 |        "3                    5.0          ...            State Bank of India   S143   \n",
1300 |        "4                    2.0          ...                      HDFC Bank   S134   \n",
1301 |        "\n",
1302 |        "   Var1  Var2 Var4 Var5 source Age  EMI_Loan_Submitted_Missing  \\\n",
1303 |        "0  HBXX     G    1    0  train  37                           1   \n",
1304 |        "1  HBXA     G    3   13  train  30                           0   \n",
1305 |        "2  HBXX     B    1    0  train  34                           1   \n",
1306 |        "3  HBXX     B    3   10  train  28                           1   \n",
1307 |        "4  HBXX     B    3   17  train  31                           1   \n",
1308 |        "\n",
1309 |        "   Interest_Rate_Missing  \n",
1310 |        "0                      1  \n",
1311 |        "1                      0  \n",
1312 |        "2                      1  \n",
1313 |        "3                      1  \n",
1314 |        "4                      1  \n",
1315 |        "\n",
1316 |        "[5 rows x 24 columns]"
1317 |       ]
1318 |      },
1319 |      "execution_count": 26,
1320 |      "metadata": {},
1321 |      "output_type": "execute_result"
1322 |     }
1323 |    ],
1324 |    "source": [
1325 |     "#不！要！了！，是的，不要了！！！\n",
1326 |     "data.drop('Lead_Creation_Date',axis=1,inplace=True)\n",
1327 |     "data.head()"
1328 |    ]
1329 |   },
1330 |   {
1331 |    "cell_type": "markdown",
1332 |    "metadata": {},
1333 |    "source": [
1334 |     "### Loan Amount and Tenure applied字段"
1335 |    ]
1336 |   },
1337 |   {
1338 |    "cell_type": "code",
1339 |    "execution_count": 27,
1340 |    "metadata": {
1341 |     "collapsed": true
1342 |    },
1343 |    "outputs": [],
1344 |    "source": [
1345 |     "#找中位数去填补缺省值（因为缺省的不多）\n",
1346 |     "data['Loan_Amount_Applied'].fillna(data['Loan_Amount_Applied'].median(),inplace=True)\n",
1347 |     "data['Loan_Tenure_Applied'].fillna(data['Loan_Tenure_Applied'].median(),inplace=True)"
1348 |    ]
1349 |   },
1350 |   {
1351 |    "cell_type": "code",
1352 |    "execution_count": 28,
1353 |    "metadata": {
1354 |     "collapsed": false
1355 |    },
1356 |    "outputs": [
1357 |     {
1358 |      "data": {
1359 |       "text/html": [
1360 |        "<div>\n",
1361 |        "<table border=\"1\" class=\"dataframe\">\n",
1362 |        "  <thead>\n",
1363 |        "    <tr style=\"text-align: right;\">\n",
1364 |        "      <th></th>\n",
1365 |        "      <th>Device_Type</th>\n",
1366 |        "      <th>Disbursed</th>\n",
1367 |        "      <th>Existing_EMI</th>\n",
1368 |        "      <th>Filled_Form</th>\n",
1369 |        "      <th>Gender</th>\n",
1370 |        "      <th>ID</th>\n",
1371 |        "      <th>Loan_Amount_Applied</th>\n",
1372 |        "      <th>Loan_Amount_Submitted</th>\n",
1373 |        "      <th>Loan_Tenure_Applied</th>\n",
1374 |        "      <th>Loan_Tenure_Submitted</th>\n",
1375 |        "      <th>...</th>\n",
1376 |        "      <th>Salary_Account</th>\n",
1377 |        "      <th>Source</th>\n",
1378 |        "      <th>Var1</th>\n",
1379 |        "      <th>Var2</th>\n",
1380 |        "      <th>Var4</th>\n",
1381 |        "      <th>Var5</th>\n",
1382 |        "      <th>source</th>\n",
1383 |        "      <th>Age</th>\n",
1384 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
1385 |        "      <th>Interest_Rate_Missing</th>\n",
1386 |        "    </tr>\n",
1387 |        "  </thead>\n",
1388 |        "  <tbody>\n",
1389 |        "    <tr>\n",
1390 |        "      <th>0</th>\n",
1391 |        "      <td>Web-browser</td>\n",
1392 |        "      <td>0.0</td>\n",
1393 |        "      <td>0.0</td>\n",
1394 |        "      <td>N</td>\n",
1395 |        "      <td>Female</td>\n",
1396 |        "      <td>ID000002C20</td>\n",
1397 |        "      <td>300000.0</td>\n",
1398 |        "      <td>NaN</td>\n",
1399 |        "      <td>5.0</td>\n",
1400 |        "      <td>NaN</td>\n",
1401 |        "      <td>...</td>\n",
1402 |        "      <td>HDFC Bank</td>\n",
1403 |        "      <td>S122</td>\n",
1404 |        "      <td>HBXX</td>\n",
1405 |        "      <td>G</td>\n",
1406 |        "      <td>1</td>\n",
1407 |        "      <td>0</td>\n",
1408 |        "      <td>train</td>\n",
1409 |        "      <td>37</td>\n",
1410 |        "      <td>1</td>\n",
1411 |        "      <td>1</td>\n",
1412 |        "    </tr>\n",
1413 |        "    <tr>\n",
1414 |        "      <th>1</th>\n",
1415 |        "      <td>Web-browser</td>\n",
1416 |        "      <td>0.0</td>\n",
1417 |        "      <td>0.0</td>\n",
1418 |        "      <td>N</td>\n",
1419 |        "      <td>Male</td>\n",
1420 |        "      <td>ID000004E40</td>\n",
1421 |        "      <td>200000.0</td>\n",
1422 |        "      <td>200000.0</td>\n",
1423 |        "      <td>2.0</td>\n",
1424 |        "      <td>2.0</td>\n",
1425 |        "      <td>...</td>\n",
1426 |        "      <td>ICICI Bank</td>\n",
1427 |        "      <td>S122</td>\n",
1428 |        "      <td>HBXA</td>\n",
1429 |        "      <td>G</td>\n",
1430 |        "      <td>3</td>\n",
1431 |        "      <td>13</td>\n",
1432 |        "      <td>train</td>\n",
1433 |        "      <td>30</td>\n",
1434 |        "      <td>0</td>\n",
1435 |        "      <td>0</td>\n",
1436 |        "    </tr>\n",
1437 |        "    <tr>\n",
1438 |        "      <th>2</th>\n",
1439 |        "      <td>Web-browser</td>\n",
1440 |        "      <td>0.0</td>\n",
1441 |        "      <td>0.0</td>\n",
1442 |        "      <td>N</td>\n",
1443 |        "      <td>Male</td>\n",
1444 |        "      <td>ID000007H20</td>\n",
1445 |        "      <td>600000.0</td>\n",
1446 |        "      <td>450000.0</td>\n",
1447 |        "      <td>4.0</td>\n",
1448 |        "      <td>4.0</td>\n",
1449 |        "      <td>...</td>\n",
1450 |        "      <td>State Bank of India</td>\n",
1451 |        "      <td>S143</td>\n",
1452 |        "      <td>HBXX</td>\n",
1453 |        "      <td>B</td>\n",
1454 |        "      <td>1</td>\n",
1455 |        "      <td>0</td>\n",
1456 |        "      <td>train</td>\n",
1457 |        "      <td>34</td>\n",
1458 |        "      <td>1</td>\n",
1459 |        "      <td>1</td>\n",
1460 |        "    </tr>\n",
1461 |        "    <tr>\n",
1462 |        "      <th>3</th>\n",
1463 |        "      <td>Web-browser</td>\n",
1464 |        "      <td>0.0</td>\n",
1465 |        "      <td>0.0</td>\n",
1466 |        "      <td>N</td>\n",
1467 |        "      <td>Male</td>\n",
1468 |        "      <td>ID000008I30</td>\n",
1469 |        "      <td>1000000.0</td>\n",
1470 |        "      <td>920000.0</td>\n",
1471 |        "      <td>5.0</td>\n",
1472 |        "      <td>5.0</td>\n",
1473 |        "      <td>...</td>\n",
1474 |        "      <td>State Bank of India</td>\n",
1475 |        "      <td>S143</td>\n",
1476 |        "      <td>HBXX</td>\n",
1477 |        "      <td>B</td>\n",
1478 |        "      <td>3</td>\n",
1479 |        "      <td>10</td>\n",
1480 |        "      <td>train</td>\n",
1481 |        "      <td>28</td>\n",
1482 |        "      <td>1</td>\n",
1483 |        "      <td>1</td>\n",
1484 |        "    </tr>\n",
1485 |        "    <tr>\n",
1486 |        "      <th>4</th>\n",
1487 |        "      <td>Web-browser</td>\n",
1488 |        "      <td>0.0</td>\n",
1489 |        "      <td>25000.0</td>\n",
1490 |        "      <td>N</td>\n",
1491 |        "      <td>Male</td>\n",
1492 |        "      <td>ID000009J40</td>\n",
1493 |        "      <td>500000.0</td>\n",
1494 |        "      <td>500000.0</td>\n",
1495 |        "      <td>2.0</td>\n",
1496 |        "      <td>2.0</td>\n",
1497 |        "      <td>...</td>\n",
1498 |        "      <td>HDFC Bank</td>\n",
1499 |        "      <td>S134</td>\n",
1500 |        "      <td>HBXX</td>\n",
1501 |        "      <td>B</td>\n",
1502 |        "      <td>3</td>\n",
1503 |        "      <td>17</td>\n",
1504 |        "      <td>train</td>\n",
1505 |        "      <td>31</td>\n",
1506 |        "      <td>1</td>\n",
1507 |        "      <td>1</td>\n",
1508 |        "    </tr>\n",
1509 |        "  </tbody>\n",
1510 |        "</table>\n",
1511 |        "<p>5 rows × 24 columns</p>\n",
1512 |        "</div>"
1513 |       ],
1514 |       "text/plain": [
1515 |        "   Device_Type  Disbursed  Existing_EMI Filled_Form  Gender           ID  \\\n",
1516 |        "0  Web-browser        0.0           0.0           N  Female  ID000002C20   \n",
1517 |        "1  Web-browser        0.0           0.0           N    Male  ID000004E40   \n",
1518 |        "2  Web-browser        0.0           0.0           N    Male  ID000007H20   \n",
1519 |        "3  Web-browser        0.0           0.0           N    Male  ID000008I30   \n",
1520 |        "4  Web-browser        0.0       25000.0           N    Male  ID000009J40   \n",
1521 |        "\n",
1522 |        "   Loan_Amount_Applied  Loan_Amount_Submitted  Loan_Tenure_Applied  \\\n",
1523 |        "0             300000.0                    NaN                  5.0   \n",
1524 |        "1             200000.0               200000.0                  2.0   \n",
1525 |        "2             600000.0               450000.0                  4.0   \n",
1526 |        "3            1000000.0               920000.0                  5.0   \n",
1527 |        "4             500000.0               500000.0                  2.0   \n",
1528 |        "\n",
1529 |        "   Loan_Tenure_Submitted          ...                 Salary_Account Source  \\\n",
1530 |        "0                    NaN          ...                      HDFC Bank   S122   \n",
1531 |        "1                    2.0          ...                     ICICI Bank   S122   \n",
1532 |        "2                    4.0          ...            State Bank of India   S143   \n",
1533 |        "3                    5.0          ...            State Bank of India   S143   \n",
1534 |        "4                    2.0          ...                      HDFC Bank   S134   \n",
1535 |        "\n",
1536 |        "   Var1  Var2 Var4 Var5 source Age  EMI_Loan_Submitted_Missing  \\\n",
1537 |        "0  HBXX     G    1    0  train  37                           1   \n",
1538 |        "1  HBXA     G    3   13  train  30                           0   \n",
1539 |        "2  HBXX     B    1    0  train  34                           1   \n",
1540 |        "3  HBXX     B    3   10  train  28                           1   \n",
1541 |        "4  HBXX     B    3   17  train  31                           1   \n",
1542 |        "\n",
1543 |        "   Interest_Rate_Missing  \n",
1544 |        "0                      1  \n",
1545 |        "1                      0  \n",
1546 |        "2                      1  \n",
1547 |        "3                      1  \n",
1548 |        "4                      1  \n",
1549 |        "\n",
1550 |        "[5 rows x 24 columns]"
1551 |       ]
1552 |      },
1553 |      "execution_count": 28,
1554 |      "metadata": {},
1555 |      "output_type": "execute_result"
1556 |     }
1557 |    ],
1558 |    "source": [
1559 |     "data.head()"
1560 |    ]
1561 |   },
1562 |   {
1563 |    "cell_type": "markdown",
1564 |    "metadata": {},
1565 |    "source": [
1566 |     "### Loan Amount and Tenure selected"
1567 |    ]
1568 |   },
1569 |   {
1570 |    "cell_type": "code",
1571 |    "execution_count": 29,
1572 |    "metadata": {
1573 |     "collapsed": true
1574 |    },
1575 |    "outputs": [],
1576 |    "source": [
1577 |     "# 缺省值太多。。。是否缺省。。。\n",
1578 |     "data['Loan_Amount_Submitted_Missing'] = data['Loan_Amount_Submitted'].apply(lambda x: 1 if pd.isnull(x) else 0)\n",
1579 |     "data['Loan_Tenure_Submitted_Missing'] = data['Loan_Tenure_Submitted'].apply(lambda x: 1 if pd.isnull(x) else 0)"
1580 |    ]
1581 |   },
1582 |   {
1583 |    "cell_type": "code",
1584 |    "execution_count": 45,
1585 |    "metadata": {
1586 |     "collapsed": false
1587 |    },
1588 |    "outputs": [
1589 |     {
1590 |      "data": {
1591 |       "text/html": [
1592 |        "<div>\n",
1593 |        "<table border=\"1\" class=\"dataframe\">\n",
1594 |        "  <thead>\n",
1595 |        "    <tr style=\"text-align: right;\">\n",
1596 |        "      <th></th>\n",
1597 |        "      <th>Device_Type</th>\n",
1598 |        "      <th>Disbursed</th>\n",
1599 |        "      <th>Employer_Name</th>\n",
1600 |        "      <th>Existing_EMI</th>\n",
1601 |        "      <th>Filled_Form</th>\n",
1602 |        "      <th>Gender</th>\n",
1603 |        "      <th>ID</th>\n",
1604 |        "      <th>Loan_Amount_Applied</th>\n",
1605 |        "      <th>Loan_Amount_Submitted</th>\n",
1606 |        "      <th>Loan_Tenure_Applied</th>\n",
1607 |        "      <th>...</th>\n",
1608 |        "      <th>Var1</th>\n",
1609 |        "      <th>Var2</th>\n",
1610 |        "      <th>Var4</th>\n",
1611 |        "      <th>Var5</th>\n",
1612 |        "      <th>source</th>\n",
1613 |        "      <th>Age</th>\n",
1614 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
1615 |        "      <th>Interest_Rate_Missing</th>\n",
1616 |        "      <th>Loan_Amount_Submitted_Missing</th>\n",
1617 |        "      <th>Loan_Tenure_Submitted_Missing</th>\n",
1618 |        "    </tr>\n",
1619 |        "  </thead>\n",
1620 |        "  <tbody>\n",
1621 |        "    <tr>\n",
1622 |        "      <th>0</th>\n",
1623 |        "      <td>Web-browser</td>\n",
1624 |        "      <td>0.0</td>\n",
1625 |        "      <td>CYBOSOL</td>\n",
1626 |        "      <td>0.0</td>\n",
1627 |        "      <td>N</td>\n",
1628 |        "      <td>Female</td>\n",
1629 |        "      <td>ID000002C20</td>\n",
1630 |        "      <td>300000.0</td>\n",
1631 |        "      <td>NaN</td>\n",
1632 |        "      <td>5.0</td>\n",
1633 |        "      <td>...</td>\n",
1634 |        "      <td>HBXX</td>\n",
1635 |        "      <td>G</td>\n",
1636 |        "      <td>1</td>\n",
1637 |        "      <td>0</td>\n",
1638 |        "      <td>train</td>\n",
1639 |        "      <td>37</td>\n",
1640 |        "      <td>1</td>\n",
1641 |        "      <td>1</td>\n",
1642 |        "      <td>1</td>\n",
1643 |        "      <td>1</td>\n",
1644 |        "    </tr>\n",
1645 |        "    <tr>\n",
1646 |        "      <th>1</th>\n",
1647 |        "      <td>Web-browser</td>\n",
1648 |        "      <td>0.0</td>\n",
1649 |        "      <td>TATA CONSULTANCY SERVICES LTD (TCS)</td>\n",
1650 |        "      <td>0.0</td>\n",
1651 |        "      <td>N</td>\n",
1652 |        "      <td>Male</td>\n",
1653 |        "      <td>ID000004E40</td>\n",
1654 |        "      <td>200000.0</td>\n",
1655 |        "      <td>200000.0</td>\n",
1656 |        "      <td>2.0</td>\n",
1657 |        "      <td>...</td>\n",
1658 |        "      <td>HBXA</td>\n",
1659 |        "      <td>G</td>\n",
1660 |        "      <td>3</td>\n",
1661 |        "      <td>13</td>\n",
1662 |        "      <td>train</td>\n",
1663 |        "      <td>30</td>\n",
1664 |        "      <td>0</td>\n",
1665 |        "      <td>0</td>\n",
1666 |        "      <td>0</td>\n",
1667 |        "      <td>0</td>\n",
1668 |        "    </tr>\n",
1669 |        "    <tr>\n",
1670 |        "      <th>2</th>\n",
1671 |        "      <td>Web-browser</td>\n",
1672 |        "      <td>0.0</td>\n",
1673 |        "      <td>ALCHEMIST HOSPITALS LTD</td>\n",
1674 |        "      <td>0.0</td>\n",
1675 |        "      <td>N</td>\n",
1676 |        "      <td>Male</td>\n",
1677 |        "      <td>ID000007H20</td>\n",
1678 |        "      <td>600000.0</td>\n",
1679 |        "      <td>450000.0</td>\n",
1680 |        "      <td>4.0</td>\n",
1681 |        "      <td>...</td>\n",
1682 |        "      <td>HBXX</td>\n",
1683 |        "      <td>B</td>\n",
1684 |        "      <td>1</td>\n",
1685 |        "      <td>0</td>\n",
1686 |        "      <td>train</td>\n",
1687 |        "      <td>34</td>\n",
1688 |        "      <td>1</td>\n",
1689 |        "      <td>1</td>\n",
1690 |        "      <td>0</td>\n",
1691 |        "      <td>0</td>\n",
1692 |        "    </tr>\n",
1693 |        "    <tr>\n",
1694 |        "      <th>3</th>\n",
1695 |        "      <td>Web-browser</td>\n",
1696 |        "      <td>0.0</td>\n",
1697 |        "      <td>BIHAR GOVERNMENT</td>\n",
1698 |        "      <td>0.0</td>\n",
1699 |        "      <td>N</td>\n",
1700 |        "      <td>Male</td>\n",
1701 |        "      <td>ID000008I30</td>\n",
1702 |        "      <td>1000000.0</td>\n",
1703 |        "      <td>920000.0</td>\n",
1704 |        "      <td>5.0</td>\n",
1705 |        "      <td>...</td>\n",
1706 |        "      <td>HBXX</td>\n",
1707 |        "      <td>B</td>\n",
1708 |        "      <td>3</td>\n",
1709 |        "      <td>10</td>\n",
1710 |        "      <td>train</td>\n",
1711 |        "      <td>28</td>\n",
1712 |        "      <td>1</td>\n",
1713 |        "      <td>1</td>\n",
1714 |        "      <td>0</td>\n",
1715 |        "      <td>0</td>\n",
1716 |        "    </tr>\n",
1717 |        "    <tr>\n",
1718 |        "      <th>4</th>\n",
1719 |        "      <td>Web-browser</td>\n",
1720 |        "      <td>0.0</td>\n",
1721 |        "      <td>GLOBAL EDGE SOFTWARE</td>\n",
1722 |        "      <td>25000.0</td>\n",
1723 |        "      <td>N</td>\n",
1724 |        "      <td>Male</td>\n",
1725 |        "      <td>ID000009J40</td>\n",
1726 |        "      <td>500000.0</td>\n",
1727 |        "      <td>500000.0</td>\n",
1728 |        "      <td>2.0</td>\n",
1729 |        "      <td>...</td>\n",
1730 |        "      <td>HBXX</td>\n",
1731 |        "      <td>B</td>\n",
1732 |        "      <td>3</td>\n",
1733 |        "      <td>17</td>\n",
1734 |        "      <td>train</td>\n",
1735 |        "      <td>31</td>\n",
1736 |        "      <td>1</td>\n",
1737 |        "      <td>1</td>\n",
1738 |        "      <td>0</td>\n",
1739 |        "      <td>0</td>\n",
1740 |        "    </tr>\n",
1741 |        "  </tbody>\n",
1742 |        "</table>\n",
1743 |        "<p>5 rows × 27 columns</p>\n",
1744 |        "</div>"
1745 |       ],
1746 |       "text/plain": [
1747 |        "   Device_Type  Disbursed                        Employer_Name  Existing_EMI  \\\n",
1748 |        "0  Web-browser        0.0                              CYBOSOL           0.0   \n",
1749 |        "1  Web-browser        0.0  TATA CONSULTANCY SERVICES LTD (TCS)           0.0   \n",
1750 |        "2  Web-browser        0.0              ALCHEMIST HOSPITALS LTD           0.0   \n",
1751 |        "3  Web-browser        0.0                     BIHAR GOVERNMENT           0.0   \n",
1752 |        "4  Web-browser        0.0                 GLOBAL EDGE SOFTWARE       25000.0   \n",
1753 |        "\n",
1754 |        "  Filled_Form  Gender           ID  Loan_Amount_Applied  \\\n",
1755 |        "0           N  Female  ID000002C20             300000.0   \n",
1756 |        "1           N    Male  ID000004E40             200000.0   \n",
1757 |        "2           N    Male  ID000007H20             600000.0   \n",
1758 |        "3           N    Male  ID000008I30            1000000.0   \n",
1759 |        "4           N    Male  ID000009J40             500000.0   \n",
1760 |        "\n",
1761 |        "   Loan_Amount_Submitted  Loan_Tenure_Applied              ...                \\\n",
1762 |        "0                    NaN                  5.0              ...                 \n",
1763 |        "1               200000.0                  2.0              ...                 \n",
1764 |        "2               450000.0                  4.0              ...                 \n",
1765 |        "3               920000.0                  5.0              ...                 \n",
1766 |        "4               500000.0                  2.0              ...                 \n",
1767 |        "\n",
1768 |        "   Var1  Var2 Var4  Var5  source Age EMI_Loan_Submitted_Missing  \\\n",
1769 |        "0  HBXX     G    1     0   train  37                          1   \n",
1770 |        "1  HBXA     G    3    13   train  30                          0   \n",
1771 |        "2  HBXX     B    1     0   train  34                          1   \n",
1772 |        "3  HBXX     B    3    10   train  28                          1   \n",
1773 |        "4  HBXX     B    3    17   train  31                          1   \n",
1774 |        "\n",
1775 |        "  Interest_Rate_Missing Loan_Amount_Submitted_Missing  \\\n",
1776 |        "0                     1                             1   \n",
1777 |        "1                     0                             0   \n",
1778 |        "2                     1                             0   \n",
1779 |        "3                     1                             0   \n",
1780 |        "4                     1                             0   \n",
1781 |        "\n",
1782 |        "   Loan_Tenure_Submitted_Missing  \n",
1783 |        "0                              1  \n",
1784 |        "1                              0  \n",
1785 |        "2                              0  \n",
1786 |        "3                              0  \n",
1787 |        "4                              0  \n",
1788 |        "\n",
1789 |        "[5 rows x 27 columns]"
1790 |       ]
1791 |      },
1792 |      "execution_count": 45,
1793 |      "metadata": {},
1794 |      "output_type": "execute_result"
1795 |     }
1796 |    ],
1797 |    "source": [
1798 |     "data.head()"
1799 |    ]
1800 |   },
1801 |   {
1802 |    "cell_type": "code",
1803 |    "execution_count": 30,
1804 |    "metadata": {
1805 |     "collapsed": true
1806 |    },
1807 |    "outputs": [],
1808 |    "source": [
1809 |     "#原来的字段就没用了\n",
1810 |     "data.drop(['Loan_Amount_Submitted','Loan_Tenure_Submitted'],axis=1,inplace=True)"
1811 |    ]
1812 |   },
1813 |   {
1814 |    "cell_type": "markdown",
1815 |    "metadata": {},
1816 |    "source": [
1817 |     "### LoggedIn"
1818 |    ]
1819 |   },
1820 |   {
1821 |    "cell_type": "code",
1822 |    "execution_count": 31,
1823 |    "metadata": {
1824 |     "collapsed": true
1825 |    },
1826 |    "outputs": [],
1827 |    "source": [
1828 |     "#没想好怎么用。。。不要了。。。\n",
1829 |     "data.drop('LoggedIn',axis=1,inplace=True)"
1830 |    ]
1831 |   },
1832 |   {
1833 |    "cell_type": "markdown",
1834 |    "metadata": {},
1835 |    "source": [
1836 |     "### salary account"
1837 |    ]
1838 |   },
1839 |   {
1840 |    "cell_type": "code",
1841 |    "execution_count": 32,
1842 |    "metadata": {
1843 |     "collapsed": true
1844 |    },
1845 |    "outputs": [],
1846 |    "source": [
1847 |     "# 可能对接多个银行，所以也不要了\n",
1848 |     "data.drop('Salary_Account',axis=1,inplace=True)"
1849 |    ]
1850 |   },
1851 |   {
1852 |    "cell_type": "markdown",
1853 |    "metadata": {},
1854 |    "source": [
1855 |     "### Processing_Fee"
1856 |    ]
1857 |   },
1858 |   {
1859 |    "cell_type": "code",
1860 |    "execution_count": 33,
1861 |    "metadata": {
1862 |     "collapsed": true
1863 |    },
1864 |    "outputs": [],
1865 |    "source": [
1866 |     "#和之前一样的处理，有或者没有\n",
1867 |     "data['Processing_Fee_Missing'] = data['Processing_Fee'].apply(lambda x: 1 if pd.isnull(x) else 0)\n",
1868 |     "#旧的字段不要了\n",
1869 |     "data.drop('Processing_Fee',axis=1,inplace=True)"
1870 |    ]
1871 |   },
1872 |   {
1873 |    "cell_type": "markdown",
1874 |    "metadata": {},
1875 |    "source": [
1876 |     "### Source"
1877 |    ]
1878 |   },
1879 |   {
1880 |    "cell_type": "code",
1881 |    "execution_count": 34,
1882 |    "metadata": {
1883 |     "collapsed": false
1884 |    },
1885 |    "outputs": [
1886 |     {
1887 |      "data": {
1888 |       "text/plain": [
1889 |        "S122      55249\n",
1890 |        "S133      42900\n",
1891 |        "others    26588\n",
1892 |        "Name: Source, dtype: int64"
1893 |       ]
1894 |      },
1895 |      "execution_count": 34,
1896 |      "metadata": {},
1897 |      "output_type": "execute_result"
1898 |     }
1899 |    ],
1900 |    "source": [
1901 |     "data['Source'] = data['Source'].apply(lambda x: 'others' if x not in ['S122','S133'] else x)\n",
1902 |     "data['Source'].value_counts()"
1903 |    ]
1904 |   },
1905 |   {
1906 |    "cell_type": "markdown",
1907 |    "metadata": {},
1908 |    "source": [
1909 |     "## 最终的数据样式"
1910 |    ]
1911 |   },
1912 |   {
1913 |    "cell_type": "code",
1914 |    "execution_count": 35,
1915 |    "metadata": {
1916 |     "collapsed": false
1917 |    },
1918 |    "outputs": [
1919 |     {
1920 |      "data": {
1921 |       "text/html": [
1922 |        "<div>\n",
1923 |        "<table border=\"1\" class=\"dataframe\">\n",
1924 |        "  <thead>\n",
1925 |        "    <tr style=\"text-align: right;\">\n",
1926 |        "      <th></th>\n",
1927 |        "      <th>Device_Type</th>\n",
1928 |        "      <th>Disbursed</th>\n",
1929 |        "      <th>Existing_EMI</th>\n",
1930 |        "      <th>Filled_Form</th>\n",
1931 |        "      <th>Gender</th>\n",
1932 |        "      <th>ID</th>\n",
1933 |        "      <th>Loan_Amount_Applied</th>\n",
1934 |        "      <th>Loan_Tenure_Applied</th>\n",
1935 |        "      <th>Mobile_Verified</th>\n",
1936 |        "      <th>Monthly_Income</th>\n",
1937 |        "      <th>...</th>\n",
1938 |        "      <th>Var2</th>\n",
1939 |        "      <th>Var4</th>\n",
1940 |        "      <th>Var5</th>\n",
1941 |        "      <th>source</th>\n",
1942 |        "      <th>Age</th>\n",
1943 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
1944 |        "      <th>Interest_Rate_Missing</th>\n",
1945 |        "      <th>Loan_Amount_Submitted_Missing</th>\n",
1946 |        "      <th>Loan_Tenure_Submitted_Missing</th>\n",
1947 |        "      <th>Processing_Fee_Missing</th>\n",
1948 |        "    </tr>\n",
1949 |        "  </thead>\n",
1950 |        "  <tbody>\n",
1951 |        "    <tr>\n",
1952 |        "      <th>0</th>\n",
1953 |        "      <td>Web-browser</td>\n",
1954 |        "      <td>0.0</td>\n",
1955 |        "      <td>0.0</td>\n",
1956 |        "      <td>N</td>\n",
1957 |        "      <td>Female</td>\n",
1958 |        "      <td>ID000002C20</td>\n",
1959 |        "      <td>300000.0</td>\n",
1960 |        "      <td>5.0</td>\n",
1961 |        "      <td>N</td>\n",
1962 |        "      <td>20000</td>\n",
1963 |        "      <td>...</td>\n",
1964 |        "      <td>G</td>\n",
1965 |        "      <td>1</td>\n",
1966 |        "      <td>0</td>\n",
1967 |        "      <td>train</td>\n",
1968 |        "      <td>37</td>\n",
1969 |        "      <td>1</td>\n",
1970 |        "      <td>1</td>\n",
1971 |        "      <td>1</td>\n",
1972 |        "      <td>1</td>\n",
1973 |        "      <td>1</td>\n",
1974 |        "    </tr>\n",
1975 |        "    <tr>\n",
1976 |        "      <th>1</th>\n",
1977 |        "      <td>Web-browser</td>\n",
1978 |        "      <td>0.0</td>\n",
1979 |        "      <td>0.0</td>\n",
1980 |        "      <td>N</td>\n",
1981 |        "      <td>Male</td>\n",
1982 |        "      <td>ID000004E40</td>\n",
1983 |        "      <td>200000.0</td>\n",
1984 |        "      <td>2.0</td>\n",
1985 |        "      <td>Y</td>\n",
1986 |        "      <td>35000</td>\n",
1987 |        "      <td>...</td>\n",
1988 |        "      <td>G</td>\n",
1989 |        "      <td>3</td>\n",
1990 |        "      <td>13</td>\n",
1991 |        "      <td>train</td>\n",
1992 |        "      <td>30</td>\n",
1993 |        "      <td>0</td>\n",
1994 |        "      <td>0</td>\n",
1995 |        "      <td>0</td>\n",
1996 |        "      <td>0</td>\n",
1997 |        "      <td>1</td>\n",
1998 |        "    </tr>\n",
1999 |        "    <tr>\n",
2000 |        "      <th>2</th>\n",
2001 |        "      <td>Web-browser</td>\n",
2002 |        "      <td>0.0</td>\n",
2003 |        "      <td>0.0</td>\n",
2004 |        "      <td>N</td>\n",
2005 |        "      <td>Male</td>\n",
2006 |        "      <td>ID000007H20</td>\n",
2007 |        "      <td>600000.0</td>\n",
2008 |        "      <td>4.0</td>\n",
2009 |        "      <td>Y</td>\n",
2010 |        "      <td>22500</td>\n",
2011 |        "      <td>...</td>\n",
2012 |        "      <td>B</td>\n",
2013 |        "      <td>1</td>\n",
2014 |        "      <td>0</td>\n",
2015 |        "      <td>train</td>\n",
2016 |        "      <td>34</td>\n",
2017 |        "      <td>1</td>\n",
2018 |        "      <td>1</td>\n",
2019 |        "      <td>0</td>\n",
2020 |        "      <td>0</td>\n",
2021 |        "      <td>1</td>\n",
2022 |        "    </tr>\n",
2023 |        "    <tr>\n",
2024 |        "      <th>3</th>\n",
2025 |        "      <td>Web-browser</td>\n",
2026 |        "      <td>0.0</td>\n",
2027 |        "      <td>0.0</td>\n",
2028 |        "      <td>N</td>\n",
2029 |        "      <td>Male</td>\n",
2030 |        "      <td>ID000008I30</td>\n",
2031 |        "      <td>1000000.0</td>\n",
2032 |        "      <td>5.0</td>\n",
2033 |        "      <td>Y</td>\n",
2034 |        "      <td>35000</td>\n",
2035 |        "      <td>...</td>\n",
2036 |        "      <td>B</td>\n",
2037 |        "      <td>3</td>\n",
2038 |        "      <td>10</td>\n",
2039 |        "      <td>train</td>\n",
2040 |        "      <td>28</td>\n",
2041 |        "      <td>1</td>\n",
2042 |        "      <td>1</td>\n",
2043 |        "      <td>0</td>\n",
2044 |        "      <td>0</td>\n",
2045 |        "      <td>1</td>\n",
2046 |        "    </tr>\n",
2047 |        "    <tr>\n",
2048 |        "      <th>4</th>\n",
2049 |        "      <td>Web-browser</td>\n",
2050 |        "      <td>0.0</td>\n",
2051 |        "      <td>25000.0</td>\n",
2052 |        "      <td>N</td>\n",
2053 |        "      <td>Male</td>\n",
2054 |        "      <td>ID000009J40</td>\n",
2055 |        "      <td>500000.0</td>\n",
2056 |        "      <td>2.0</td>\n",
2057 |        "      <td>Y</td>\n",
2058 |        "      <td>100000</td>\n",
2059 |        "      <td>...</td>\n",
2060 |        "      <td>B</td>\n",
2061 |        "      <td>3</td>\n",
2062 |        "      <td>17</td>\n",
2063 |        "      <td>train</td>\n",
2064 |        "      <td>31</td>\n",
2065 |        "      <td>1</td>\n",
2066 |        "      <td>1</td>\n",
2067 |        "      <td>0</td>\n",
2068 |        "      <td>0</td>\n",
2069 |        "      <td>1</td>\n",
2070 |        "    </tr>\n",
2071 |        "  </tbody>\n",
2072 |        "</table>\n",
2073 |        "<p>5 rows × 22 columns</p>\n",
2074 |        "</div>"
2075 |       ],
2076 |       "text/plain": [
2077 |        "   Device_Type  Disbursed  Existing_EMI Filled_Form  Gender           ID  \\\n",
2078 |        "0  Web-browser        0.0           0.0           N  Female  ID000002C20   \n",
2079 |        "1  Web-browser        0.0           0.0           N    Male  ID000004E40   \n",
2080 |        "2  Web-browser        0.0           0.0           N    Male  ID000007H20   \n",
2081 |        "3  Web-browser        0.0           0.0           N    Male  ID000008I30   \n",
2082 |        "4  Web-browser        0.0       25000.0           N    Male  ID000009J40   \n",
2083 |        "\n",
2084 |        "   Loan_Amount_Applied  Loan_Tenure_Applied Mobile_Verified  Monthly_Income  \\\n",
2085 |        "0             300000.0                  5.0               N           20000   \n",
2086 |        "1             200000.0                  2.0               Y           35000   \n",
2087 |        "2             600000.0                  4.0               Y           22500   \n",
2088 |        "3            1000000.0                  5.0               Y           35000   \n",
2089 |        "4             500000.0                  2.0               Y          100000   \n",
2090 |        "\n",
2091 |        "            ...           Var2 Var4 Var5  source  Age  \\\n",
2092 |        "0           ...              G    1    0   train   37   \n",
2093 |        "1           ...              G    3   13   train   30   \n",
2094 |        "2           ...              B    1    0   train   34   \n",
2095 |        "3           ...              B    3   10   train   28   \n",
2096 |        "4           ...              B    3   17   train   31   \n",
2097 |        "\n",
2098 |        "  EMI_Loan_Submitted_Missing  Interest_Rate_Missing  \\\n",
2099 |        "0                          1                      1   \n",
2100 |        "1                          0                      0   \n",
2101 |        "2                          1                      1   \n",
2102 |        "3                          1                      1   \n",
2103 |        "4                          1                      1   \n",
2104 |        "\n",
2105 |        "   Loan_Amount_Submitted_Missing  Loan_Tenure_Submitted_Missing  \\\n",
2106 |        "0                              1                              1   \n",
2107 |        "1                              0                              0   \n",
2108 |        "2                              0                              0   \n",
2109 |        "3                              0                              0   \n",
2110 |        "4                              0                              0   \n",
2111 |        "\n",
2112 |        "   Processing_Fee_Missing  \n",
2113 |        "0                       1  \n",
2114 |        "1                       1  \n",
2115 |        "2                       1  \n",
2116 |        "3                       1  \n",
2117 |        "4                       1  \n",
2118 |        "\n",
2119 |        "[5 rows x 22 columns]"
2120 |       ]
2121 |      },
2122 |      "execution_count": 35,
2123 |      "metadata": {},
2124 |      "output_type": "execute_result"
2125 |     }
2126 |    ],
2127 |    "source": [
2128 |     "data.head()"
2129 |    ]
2130 |   },
2131 |   {
2132 |    "cell_type": "code",
2133 |    "execution_count": 36,
2134 |    "metadata": {
2135 |     "collapsed": false
2136 |    },
2137 |    "outputs": [
2138 |     {
2139 |      "data": {
2140 |       "text/html": [
2141 |        "<div>\n",
2142 |        "<table border=\"1\" class=\"dataframe\">\n",
2143 |        "  <thead>\n",
2144 |        "    <tr style=\"text-align: right;\">\n",
2145 |        "      <th></th>\n",
2146 |        "      <th>Disbursed</th>\n",
2147 |        "      <th>Existing_EMI</th>\n",
2148 |        "      <th>Loan_Amount_Applied</th>\n",
2149 |        "      <th>Loan_Tenure_Applied</th>\n",
2150 |        "      <th>Monthly_Income</th>\n",
2151 |        "      <th>Var4</th>\n",
2152 |        "      <th>Var5</th>\n",
2153 |        "      <th>Age</th>\n",
2154 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
2155 |        "      <th>Interest_Rate_Missing</th>\n",
2156 |        "      <th>Loan_Amount_Submitted_Missing</th>\n",
2157 |        "      <th>Loan_Tenure_Submitted_Missing</th>\n",
2158 |        "      <th>Processing_Fee_Missing</th>\n",
2159 |        "    </tr>\n",
2160 |        "  </thead>\n",
2161 |        "  <tbody>\n",
2162 |        "    <tr>\n",
2163 |        "      <th>count</th>\n",
2164 |        "      <td>87020.000000</td>\n",
2165 |        "      <td>1.247370e+05</td>\n",
2166 |        "      <td>1.247370e+05</td>\n",
2167 |        "      <td>124737.000000</td>\n",
2168 |        "      <td>1.247370e+05</td>\n",
2169 |        "      <td>124737.000000</td>\n",
2170 |        "      <td>124737.000000</td>\n",
2171 |        "      <td>124737.000000</td>\n",
2172 |        "      <td>124737.000000</td>\n",
2173 |        "      <td>124737.000000</td>\n",
2174 |        "      <td>124737.000000</td>\n",
2175 |        "      <td>124737.000000</td>\n",
2176 |        "      <td>124737.000000</td>\n",
2177 |        "    </tr>\n",
2178 |        "    <tr>\n",
2179 |        "      <th>mean</th>\n",
2180 |        "      <td>0.014629</td>\n",
2181 |        "      <td>3.633107e+03</td>\n",
2182 |        "      <td>2.298744e+05</td>\n",
2183 |        "      <td>2.138075</td>\n",
2184 |        "      <td>5.309073e+04</td>\n",
2185 |        "      <td>2.950560</td>\n",
2186 |        "      <td>4.964774</td>\n",
2187 |        "      <td>30.906996</td>\n",
2188 |        "      <td>0.680640</td>\n",
2189 |        "      <td>0.680640</td>\n",
2190 |        "      <td>0.397116</td>\n",
2191 |        "      <td>0.397116</td>\n",
2192 |        "      <td>0.684208</td>\n",
2193 |        "    </tr>\n",
2194 |        "    <tr>\n",
2195 |        "      <th>std</th>\n",
2196 |        "      <td>0.120062</td>\n",
2197 |        "      <td>3.367642e+04</td>\n",
2198 |        "      <td>3.539938e+05</td>\n",
2199 |        "      <td>2.014874</td>\n",
2200 |        "      <td>1.823394e+06</td>\n",
2201 |        "      <td>1.695261</td>\n",
2202 |        "      <td>5.669784</td>\n",
2203 |        "      <td>7.137860</td>\n",
2204 |        "      <td>0.466231</td>\n",
2205 |        "      <td>0.466231</td>\n",
2206 |        "      <td>0.489302</td>\n",
2207 |        "      <td>0.489302</td>\n",
2208 |        "      <td>0.464833</td>\n",
2209 |        "    </tr>\n",
2210 |        "    <tr>\n",
2211 |        "      <th>min</th>\n",
2212 |        "      <td>0.000000</td>\n",
2213 |        "      <td>0.000000e+00</td>\n",
2214 |        "      <td>0.000000e+00</td>\n",
2215 |        "      <td>0.000000</td>\n",
2216 |        "      <td>0.000000e+00</td>\n",
2217 |        "      <td>0.000000</td>\n",
2218 |        "      <td>0.000000</td>\n",
2219 |        "      <td>18.000000</td>\n",
2220 |        "      <td>0.000000</td>\n",
2221 |        "      <td>0.000000</td>\n",
2222 |        "      <td>0.000000</td>\n",
2223 |        "      <td>0.000000</td>\n",
2224 |        "      <td>0.000000</td>\n",
2225 |        "    </tr>\n",
2226 |        "    <tr>\n",
2227 |        "      <th>25%</th>\n",
2228 |        "      <td>NaN</td>\n",
2229 |        "      <td>0.000000e+00</td>\n",
2230 |        "      <td>0.000000e+00</td>\n",
2231 |        "      <td>0.000000</td>\n",
2232 |        "      <td>1.650000e+04</td>\n",
2233 |        "      <td>1.000000</td>\n",
2234 |        "      <td>0.000000</td>\n",
2235 |        "      <td>26.000000</td>\n",
2236 |        "      <td>0.000000</td>\n",
2237 |        "      <td>0.000000</td>\n",
2238 |        "      <td>0.000000</td>\n",
2239 |        "      <td>0.000000</td>\n",
2240 |        "      <td>0.000000</td>\n",
2241 |        "    </tr>\n",
2242 |        "    <tr>\n",
2243 |        "      <th>50%</th>\n",
2244 |        "      <td>NaN</td>\n",
2245 |        "      <td>0.000000e+00</td>\n",
2246 |        "      <td>1.000000e+05</td>\n",
2247 |        "      <td>2.000000</td>\n",
2248 |        "      <td>2.500000e+04</td>\n",
2249 |        "      <td>3.000000</td>\n",
2250 |        "      <td>2.000000</td>\n",
2251 |        "      <td>29.000000</td>\n",
2252 |        "      <td>1.000000</td>\n",
2253 |        "      <td>1.000000</td>\n",
2254 |        "      <td>0.000000</td>\n",
2255 |        "      <td>0.000000</td>\n",
2256 |        "      <td>1.000000</td>\n",
2257 |        "    </tr>\n",
2258 |        "    <tr>\n",
2259 |        "      <th>75%</th>\n",
2260 |        "      <td>NaN</td>\n",
2261 |        "      <td>3.500000e+03</td>\n",
2262 |        "      <td>3.000000e+05</td>\n",
2263 |        "      <td>4.000000</td>\n",
2264 |        "      <td>4.000000e+04</td>\n",
2265 |        "      <td>5.000000</td>\n",
2266 |        "      <td>11.000000</td>\n",
2267 |        "      <td>34.000000</td>\n",
2268 |        "      <td>1.000000</td>\n",
2269 |        "      <td>1.000000</td>\n",
2270 |        "      <td>1.000000</td>\n",
2271 |        "      <td>1.000000</td>\n",
2272 |        "      <td>1.000000</td>\n",
2273 |        "    </tr>\n",
2274 |        "    <tr>\n",
2275 |        "      <th>max</th>\n",
2276 |        "      <td>1.000000</td>\n",
2277 |        "      <td>1.000000e+07</td>\n",
2278 |        "      <td>1.500000e+07</td>\n",
2279 |        "      <td>10.000000</td>\n",
2280 |        "      <td>4.445544e+08</td>\n",
2281 |        "      <td>7.000000</td>\n",
2282 |        "      <td>18.000000</td>\n",
2283 |        "      <td>100.000000</td>\n",
2284 |        "      <td>1.000000</td>\n",
2285 |        "      <td>1.000000</td>\n",
2286 |        "      <td>1.000000</td>\n",
2287 |        "      <td>1.000000</td>\n",
2288 |        "      <td>1.000000</td>\n",
2289 |        "    </tr>\n",
2290 |        "  </tbody>\n",
2291 |        "</table>\n",
2292 |        "</div>"
2293 |       ],
2294 |       "text/plain": [
2295 |        "          Disbursed  Existing_EMI  Loan_Amount_Applied  Loan_Tenure_Applied  \\\n",
2296 |        "count  87020.000000  1.247370e+05         1.247370e+05        124737.000000   \n",
2297 |        "mean       0.014629  3.633107e+03         2.298744e+05             2.138075   \n",
2298 |        "std        0.120062  3.367642e+04         3.539938e+05             2.014874   \n",
2299 |        "min        0.000000  0.000000e+00         0.000000e+00             0.000000   \n",
2300 |        "25%             NaN  0.000000e+00         0.000000e+00             0.000000   \n",
2301 |        "50%             NaN  0.000000e+00         1.000000e+05             2.000000   \n",
2302 |        "75%             NaN  3.500000e+03         3.000000e+05             4.000000   \n",
2303 |        "max        1.000000  1.000000e+07         1.500000e+07            10.000000   \n",
2304 |        "\n",
2305 |        "       Monthly_Income           Var4           Var5            Age  \\\n",
2306 |        "count    1.247370e+05  124737.000000  124737.000000  124737.000000   \n",
2307 |        "mean     5.309073e+04       2.950560       4.964774      30.906996   \n",
2308 |        "std      1.823394e+06       1.695261       5.669784       7.137860   \n",
2309 |        "min      0.000000e+00       0.000000       0.000000      18.000000   \n",
2310 |        "25%      1.650000e+04       1.000000       0.000000      26.000000   \n",
2311 |        "50%      2.500000e+04       3.000000       2.000000      29.000000   \n",
2312 |        "75%      4.000000e+04       5.000000      11.000000      34.000000   \n",
2313 |        "max      4.445544e+08       7.000000      18.000000     100.000000   \n",
2314 |        "\n",
2315 |        "       EMI_Loan_Submitted_Missing  Interest_Rate_Missing  \\\n",
2316 |        "count               124737.000000          124737.000000   \n",
2317 |        "mean                     0.680640               0.680640   \n",
2318 |        "std                      0.466231               0.466231   \n",
2319 |        "min                      0.000000               0.000000   \n",
2320 |        "25%                      0.000000               0.000000   \n",
2321 |        "50%                      1.000000               1.000000   \n",
2322 |        "75%                      1.000000               1.000000   \n",
2323 |        "max                      1.000000               1.000000   \n",
2324 |        "\n",
2325 |        "       Loan_Amount_Submitted_Missing  Loan_Tenure_Submitted_Missing  \\\n",
2326 |        "count                  124737.000000                  124737.000000   \n",
2327 |        "mean                        0.397116                       0.397116   \n",
2328 |        "std                         0.489302                       0.489302   \n",
2329 |        "min                         0.000000                       0.000000   \n",
2330 |        "25%                         0.000000                       0.000000   \n",
2331 |        "50%                         0.000000                       0.000000   \n",
2332 |        "75%                         1.000000                       1.000000   \n",
2333 |        "max                         1.000000                       1.000000   \n",
2334 |        "\n",
2335 |        "       Processing_Fee_Missing  \n",
2336 |        "count           124737.000000  \n",
2337 |        "mean                 0.684208  \n",
2338 |        "std                  0.464833  \n",
2339 |        "min                  0.000000  \n",
2340 |        "25%                  0.000000  \n",
2341 |        "50%                  1.000000  \n",
2342 |        "75%                  1.000000  \n",
2343 |        "max                  1.000000  "
2344 |       ]
2345 |      },
2346 |      "execution_count": 36,
2347 |      "metadata": {},
2348 |      "output_type": "execute_result"
2349 |     }
2350 |    ],
2351 |    "source": [
2352 |     "data.describe()"
2353 |    ]
2354 |   },
2355 |   {
2356 |    "cell_type": "code",
2357 |    "execution_count": 37,
2358 |    "metadata": {
2359 |     "collapsed": false
2360 |    },
2361 |    "outputs": [
2362 |     {
2363 |      "data": {
2364 |       "text/plain": [
2365 |        "Device_Type                          0\n",
2366 |        "Disbursed                        37717\n",
2367 |        "Existing_EMI                         0\n",
2368 |        "Filled_Form                          0\n",
2369 |        "Gender                               0\n",
2370 |        "ID                                   0\n",
2371 |        "Loan_Amount_Applied                  0\n",
2372 |        "Loan_Tenure_Applied                  0\n",
2373 |        "Mobile_Verified                      0\n",
2374 |        "Monthly_Income                       0\n",
2375 |        "Source                               0\n",
2376 |        "Var1                                 0\n",
2377 |        "Var2                                 0\n",
2378 |        "Var4                                 0\n",
2379 |        "Var5                                 0\n",
2380 |        "source                               0\n",
2381 |        "Age                                  0\n",
2382 |        "EMI_Loan_Submitted_Missing           0\n",
2383 |        "Interest_Rate_Missing                0\n",
2384 |        "Loan_Amount_Submitted_Missing        0\n",
2385 |        "Loan_Tenure_Submitted_Missing        0\n",
2386 |        "Processing_Fee_Missing               0\n",
2387 |        "dtype: int64"
2388 |       ]
2389 |      },
2390 |      "execution_count": 37,
2391 |      "metadata": {},
2392 |      "output_type": "execute_result"
2393 |     }
2394 |    ],
2395 |    "source": [
2396 |     "data.apply(lambda x: sum(x.isnull()))"
2397 |    ]
2398 |   },
2399 |   {
2400 |    "cell_type": "code",
2401 |    "execution_count": 38,
2402 |    "metadata": {
2403 |     "collapsed": false
2404 |    },
2405 |    "outputs": [
2406 |     {
2407 |      "data": {
2408 |       "text/plain": [
2409 |        "Device_Type                       object\n",
2410 |        "Disbursed                        float64\n",
2411 |        "Existing_EMI                     float64\n",
2412 |        "Filled_Form                       object\n",
2413 |        "Gender                            object\n",
2414 |        "ID                                object\n",
2415 |        "Loan_Amount_Applied              float64\n",
2416 |        "Loan_Tenure_Applied              float64\n",
2417 |        "Mobile_Verified                   object\n",
2418 |        "Monthly_Income                     int64\n",
2419 |        "Source                            object\n",
2420 |        "Var1                              object\n",
2421 |        "Var2                              object\n",
2422 |        "Var4                               int64\n",
2423 |        "Var5                               int64\n",
2424 |        "source                            object\n",
2425 |        "Age                                int64\n",
2426 |        "EMI_Loan_Submitted_Missing         int64\n",
2427 |        "Interest_Rate_Missing              int64\n",
2428 |        "Loan_Amount_Submitted_Missing      int64\n",
2429 |        "Loan_Tenure_Submitted_Missing      int64\n",
2430 |        "Processing_Fee_Missing             int64\n",
2431 |        "dtype: object"
2432 |       ]
2433 |      },
2434 |      "execution_count": 38,
2435 |      "metadata": {},
2436 |      "output_type": "execute_result"
2437 |     }
2438 |    ],
2439 |    "source": [
2440 |     "data.dtypes"
2441 |    ]
2442 |   },
2443 |   {
2444 |    "cell_type": "markdown",
2445 |    "metadata": {},
2446 |    "source": [
2447 |     "### 数值编码"
2448 |    ]
2449 |   },
2450 |   {
2451 |    "cell_type": "code",
2452 |    "execution_count": 39,
2453 |    "metadata": {
2454 |     "collapsed": true
2455 |    },
2456 |    "outputs": [],
2457 |    "source": [
2458 |     "from sklearn.preprocessing import LabelEncoder\n",
2459 |     "le = LabelEncoder()\n",
2460 |     "var_to_encode = ['Device_Type','Filled_Form','Gender','Var1','Var2','Mobile_Verified','Source']\n",
2461 |     "for col in var_to_encode:\n",
2462 |     "    data[col] = le.fit_transform(data[col])"
2463 |    ]
2464 |   },
2465 |   {
2466 |    "cell_type": "code",
2467 |    "execution_count": 40,
2468 |    "metadata": {
2469 |     "collapsed": false
2470 |    },
2471 |    "outputs": [
2472 |     {
2473 |      "data": {
2474 |       "text/html": [
2475 |        "<div>\n",
2476 |        "<table border=\"1\" class=\"dataframe\">\n",
2477 |        "  <thead>\n",
2478 |        "    <tr style=\"text-align: right;\">\n",
2479 |        "      <th></th>\n",
2480 |        "      <th>Device_Type</th>\n",
2481 |        "      <th>Disbursed</th>\n",
2482 |        "      <th>Existing_EMI</th>\n",
2483 |        "      <th>Filled_Form</th>\n",
2484 |        "      <th>Gender</th>\n",
2485 |        "      <th>ID</th>\n",
2486 |        "      <th>Loan_Amount_Applied</th>\n",
2487 |        "      <th>Loan_Tenure_Applied</th>\n",
2488 |        "      <th>Mobile_Verified</th>\n",
2489 |        "      <th>Monthly_Income</th>\n",
2490 |        "      <th>...</th>\n",
2491 |        "      <th>Var2</th>\n",
2492 |        "      <th>Var4</th>\n",
2493 |        "      <th>Var5</th>\n",
2494 |        "      <th>source</th>\n",
2495 |        "      <th>Age</th>\n",
2496 |        "      <th>EMI_Loan_Submitted_Missing</th>\n",
2497 |        "      <th>Interest_Rate_Missing</th>\n",
2498 |        "      <th>Loan_Amount_Submitted_Missing</th>\n",
2499 |        "      <th>Loan_Tenure_Submitted_Missing</th>\n",
2500 |        "      <th>Processing_Fee_Missing</th>\n",
2501 |        "    </tr>\n",
2502 |        "  </thead>\n",
2503 |        "  <tbody>\n",
2504 |        "    <tr>\n",
2505 |        "      <th>0</th>\n",
2506 |        "      <td>1</td>\n",
2507 |        "      <td>0.0</td>\n",
2508 |        "      <td>0.0</td>\n",
2509 |        "      <td>0</td>\n",
2510 |        "      <td>0</td>\n",
2511 |        "      <td>ID000002C20</td>\n",
2512 |        "      <td>300000.0</td>\n",
2513 |        "      <td>5.0</td>\n",
2514 |        "      <td>0</td>\n",
2515 |        "      <td>20000</td>\n",
2516 |        "      <td>...</td>\n",
2517 |        "      <td>6</td>\n",
2518 |        "      <td>1</td>\n",
2519 |        "      <td>0</td>\n",
2520 |        "      <td>train</td>\n",
2521 |        "      <td>37</td>\n",
2522 |        "      <td>1</td>\n",
2523 |        "      <td>1</td>\n",
2524 |        "      <td>1</td>\n",
2525 |        "      <td>1</td>\n",
2526 |        "      <td>1</td>\n",
2527 |        "    </tr>\n",
2528 |        "    <tr>\n",
2529 |        "      <th>1</th>\n",
2530 |        "      <td>1</td>\n",
2531 |        "      <td>0.0</td>\n",
2532 |        "      <td>0.0</td>\n",
2533 |        "      <td>0</td>\n",
2534 |        "      <td>1</td>\n",
2535 |        "      <td>ID000004E40</td>\n",
2536 |        "      <td>200000.0</td>\n",
2537 |        "      <td>2.0</td>\n",
2538 |        "      <td>1</td>\n",
2539 |        "      <td>35000</td>\n",
2540 |        "      <td>...</td>\n",
2541 |        "      <td>6</td>\n",
2542 |        "      <td>3</td>\n",
2543 |        "      <td>13</td>\n",
2544 |        "      <td>train</td>\n",
2545 |        "      <td>30</td>\n",
2546 |        "      <td>0</td>\n",
2547 |        "      <td>0</td>\n",
2548 |        "      <td>0</td>\n",
2549 |        "      <td>0</td>\n",
2550 |        "      <td>1</td>\n",
2551 |        "    </tr>\n",
2552 |        "    <tr>\n",
2553 |        "      <th>2</th>\n",
2554 |        "      <td>1</td>\n",
2555 |        "      <td>0.0</td>\n",
2556 |        "      <td>0.0</td>\n",
2557 |        "      <td>0</td>\n",
2558 |        "      <td>1</td>\n",
2559 |        "      <td>ID000007H20</td>\n",
2560 |        "      <td>600000.0</td>\n",
2561 |        "      <td>4.0</td>\n",
2562 |        "      <td>1</td>\n",
2563 |        "      <td>22500</td>\n",
2564 |        "      <td>...</td>\n",
2565 |        "      <td>1</td>\n",
2566 |        "      <td>1</td>\n",
2567 |        "      <td>0</td>\n",
2568 |        "      <td>train</td>\n",
2569 |        "      <td>34</td>\n",
2570 |        "      <td>1</td>\n",
2571 |        "      <td>1</td>\n",
2572 |        "      <td>0</td>\n",
2573 |        "      <td>0</td>\n",
2574 |        "      <td>1</td>\n",
2575 |        "    </tr>\n",
2576 |        "    <tr>\n",
2577 |        "      <th>3</th>\n",
2578 |        "      <td>1</td>\n",
2579 |        "      <td>0.0</td>\n",
2580 |        "      <td>0.0</td>\n",
2581 |        "      <td>0</td>\n",
2582 |        "      <td>1</td>\n",
2583 |        "      <td>ID000008I30</td>\n",
2584 |        "      <td>1000000.0</td>\n",
2585 |        "      <td>5.0</td>\n",
2586 |        "      <td>1</td>\n",
2587 |        "      <td>35000</td>\n",
2588 |        "      <td>...</td>\n",
2589 |        "      <td>1</td>\n",
2590 |        "      <td>3</td>\n",
2591 |        "      <td>10</td>\n",
2592 |        "      <td>train</td>\n",
2593 |        "      <td>28</td>\n",
2594 |        "      <td>1</td>\n",
2595 |        "      <td>1</td>\n",
2596 |        "      <td>0</td>\n",
2597 |        "      <td>0</td>\n",
2598 |        "      <td>1</td>\n",
2599 |        "    </tr>\n",
2600 |        "    <tr>\n",
2601 |        "      <th>4</th>\n",
2602 |        "      <td>1</td>\n",
2603 |        "      <td>0.0</td>\n",
2604 |        "      <td>25000.0</td>\n",
2605 |        "      <td>0</td>\n",
2606 |        "      <td>1</td>\n",
2607 |        "      <td>ID000009J40</td>\n",
2608 |        "      <td>500000.0</td>\n",
2609 |        "      <td>2.0</td>\n",
2610 |        "      <td>1</td>\n",
2611 |        "      <td>100000</td>\n",
2612 |        "      <td>...</td>\n",
2613 |        "      <td>1</td>\n",
2614 |        "      <td>3</td>\n",
2615 |        "      <td>17</td>\n",
2616 |        "      <td>train</td>\n",
2617 |        "      <td>31</td>\n",
2618 |        "      <td>1</td>\n",
2619 |        "      <td>1</td>\n",
2620 |        "      <td>0</td>\n",
2621 |        "      <td>0</td>\n",
2622 |        "      <td>1</td>\n",
2623 |        "    </tr>\n",
2624 |        "  </tbody>\n",
2625 |        "</table>\n",
2626 |        "<p>5 rows × 22 columns</p>\n",
2627 |        "</div>"
2628 |       ],
2629 |       "text/plain": [
2630 |        "   Device_Type  Disbursed  Existing_EMI  Filled_Form  Gender           ID  \\\n",
2631 |        "0            1        0.0           0.0            0       0  ID000002C20   \n",
2632 |        "1            1        0.0           0.0            0       1  ID000004E40   \n",
2633 |        "2            1        0.0           0.0            0       1  ID000007H20   \n",
2634 |        "3            1        0.0           0.0            0       1  ID000008I30   \n",
2635 |        "4            1        0.0       25000.0            0       1  ID000009J40   \n",
2636 |        "\n",
2637 |        "   Loan_Amount_Applied  Loan_Tenure_Applied  Mobile_Verified  Monthly_Income  \\\n",
2638 |        "0             300000.0                  5.0                0           20000   \n",
2639 |        "1             200000.0                  2.0                1           35000   \n",
2640 |        "2             600000.0                  4.0                1           22500   \n",
2641 |        "3            1000000.0                  5.0                1           35000   \n",
2642 |        "4             500000.0                  2.0                1          100000   \n",
2643 |        "\n",
2644 |        "            ...            Var2  Var4  Var5  source  Age  \\\n",
2645 |        "0           ...               6     1     0   train   37   \n",
2646 |        "1           ...               6     3    13   train   30   \n",
2647 |        "2           ...               1     1     0   train   34   \n",
2648 |        "3           ...               1     3    10   train   28   \n",
2649 |        "4           ...               1     3    17   train   31   \n",
2650 |        "\n",
2651 |        "  EMI_Loan_Submitted_Missing  Interest_Rate_Missing  \\\n",
2652 |        "0                          1                      1   \n",
2653 |        "1                          0                      0   \n",
2654 |        "2                          1                      1   \n",
2655 |        "3                          1                      1   \n",
2656 |        "4                          1                      1   \n",
2657 |        "\n",
2658 |        "   Loan_Amount_Submitted_Missing  Loan_Tenure_Submitted_Missing  \\\n",
2659 |        "0                              1                              1   \n",
2660 |        "1                              0                              0   \n",
2661 |        "2                              0                              0   \n",
2662 |        "3                              0                              0   \n",
2663 |        "4                              0                              0   \n",
2664 |        "\n",
2665 |        "   Processing_Fee_Missing  \n",
2666 |        "0                       1  \n",
2667 |        "1                       1  \n",
2668 |        "2                       1  \n",
2669 |        "3                       1  \n",
2670 |        "4                       1  \n",
2671 |        "\n",
2672 |        "[5 rows x 22 columns]"
2673 |       ]
2674 |      },
2675 |      "execution_count": 40,
2676 |      "metadata": {},
2677 |      "output_type": "execute_result"
2678 |     }
2679 |    ],
2680 |    "source": [
2681 |     "data.head()"
2682 |    ]
2683 |   },
2684 |   {
2685 |    "cell_type": "code",
2686 |    "execution_count": 41,
2687 |    "metadata": {
2688 |     "collapsed": false
2689 |    },
2690 |    "outputs": [
2691 |     {
2692 |      "data": {
2693 |       "text/plain": [
2694 |        "Device_Type                        int64\n",
2695 |        "Disbursed                        float64\n",
2696 |        "Existing_EMI                     float64\n",
2697 |        "Filled_Form                        int64\n",
2698 |        "Gender                             int64\n",
2699 |        "ID                                object\n",
2700 |        "Loan_Amount_Applied              float64\n",
2701 |        "Loan_Tenure_Applied              float64\n",
2702 |        "Mobile_Verified                    int64\n",
2703 |        "Monthly_Income                     int64\n",
2704 |        "Source                             int64\n",
2705 |        "Var1                               int64\n",
2706 |        "Var2                               int64\n",
2707 |        "Var4                               int64\n",
2708 |        "Var5                               int64\n",
2709 |        "source                            object\n",
2710 |        "Age                                int64\n",
2711 |        "EMI_Loan_Submitted_Missing         int64\n",
2712 |        "Interest_Rate_Missing              int64\n",
2713 |        "Loan_Amount_Submitted_Missing      int64\n",
2714 |        "Loan_Tenure_Submitted_Missing      int64\n",
2715 |        "Processing_Fee_Missing             int64\n",
2716 |        "dtype: object"
2717 |       ]
2718 |      },
2719 |      "execution_count": 41,
2720 |      "metadata": {},
2721 |      "output_type": "execute_result"
2722 |     }
2723 |    ],
2724 |    "source": [
2725 |     "data.dtypes"
2726 |    ]
2727 |   },
2728 |   {
2729 |    "cell_type": "markdown",
2730 |    "metadata": {},
2731 |    "source": [
2732 |     "### 类别型的One-Hot 编码"
2733 |    ]
2734 |   },
2735 |   {
2736 |    "cell_type": "code",
2737 |    "execution_count": 42,
2738 |    "metadata": {
2739 |     "collapsed": false
2740 |    },
2741 |    "outputs": [
2742 |     {
2743 |      "data": {
2744 |       "text/plain": [
2745 |        "Index([u'Disbursed', u'Existing_EMI', u'ID', u'Loan_Amount_Applied',\n",
2746 |        "       u'Loan_Tenure_Applied', u'Monthly_Income', u'Var4', u'Var5', u'source',\n",
2747 |        "       u'Age', u'EMI_Loan_Submitted_Missing', u'Interest_Rate_Missing',\n",
2748 |        "       u'Loan_Amount_Submitted_Missing', u'Loan_Tenure_Submitted_Missing',\n",
2749 |        "       u'Processing_Fee_Missing', u'Device_Type_0', u'Device_Type_1',\n",
2750 |        "       u'Filled_Form_0', u'Filled_Form_1', u'Gender_0', u'Gender_1', u'Var1_0',\n",
2751 |        "       u'Var1_1', u'Var1_2', u'Var1_3', u'Var1_4', u'Var1_5', u'Var1_6',\n",
2752 |        "       u'Var1_7', u'Var1_8', u'Var1_9', u'Var1_10', u'Var1_11', u'Var1_12',\n",
2753 |        "       u'Var1_13', u'Var1_14', u'Var1_15', u'Var1_16', u'Var1_17', u'Var1_18',\n",
2754 |        "       u'Var2_0', u'Var2_1', u'Var2_2', u'Var2_3', u'Var2_4', u'Var2_5',\n",
2755 |        "       u'Var2_6', u'Mobile_Verified_0', u'Mobile_Verified_1', u'Source_0',\n",
2756 |        "       u'Source_1', u'Source_2'],\n",
2757 |        "      dtype='object')"
2758 |       ]
2759 |      },
2760 |      "execution_count": 42,
2761 |      "metadata": {},
2762 |      "output_type": "execute_result"
2763 |     }
2764 |    ],
2765 |    "source": [
2766 |     "data = pd.get_dummies(data, columns=var_to_encode)\n",
2767 |     "data.columns"
2768 |    ]
2769 |   },
2770 |   {
2771 |    "cell_type": "markdown",
2772 |    "metadata": {},
2773 |    "source": [
2774 |     "### 区分训练和测试数据"
2775 |    ]
2776 |   },
2777 |   {
2778 |    "cell_type": "code",
2779 |    "execution_count": 43,
2780 |    "metadata": {
2781 |     "collapsed": true
2782 |    },
2783 |    "outputs": [],
2784 |    "source": [
2785 |     "train = data.loc[data['source']=='train']\n",
2786 |     "test = data.loc[data['source']=='test']"
2787 |    ]
2788 |   },
2789 |   {
2790 |    "cell_type": "code",
2791 |    "execution_count": 44,
2792 |    "metadata": {
2793 |     "collapsed": false
2794 |    },
2795 |    "outputs": [
2796 |     {
2797 |      "name": "stderr",
2798 |      "output_type": "stream",
2799 |      "text": [
2800 |       "/Library/Python/2.7/site-packages/ipykernel/__main__.py:1: SettingWithCopyWarning: \n",
2801 |       "A value is trying to be set on a copy of a slice from a DataFrame\n",
2802 |       "\n",
2803 |       "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
2804 |       "  if __name__ == '__main__':\n",
2805 |       "/Library/Python/2.7/site-packages/ipykernel/__main__.py:2: SettingWithCopyWarning: \n",
2806 |       "A value is trying to be set on a copy of a slice from a DataFrame\n",
2807 |       "\n",
2808 |       "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
2809 |       "  from ipykernel import kernelapp as app\n"
2810 |      ]
2811 |     }
2812 |    ],
2813 |    "source": [
2814 |     "train.drop('source',axis=1,inplace=True)\n",
2815 |     "test.drop(['source','Disbursed'],axis=1,inplace=True)"
2816 |    ]
2817 |   },
2818 |   {
2819 |    "cell_type": "code",
2820 |    "execution_count": 45,
2821 |    "metadata": {
2822 |     "collapsed": true
2823 |    },
2824 |    "outputs": [],
2825 |    "source": [
2826 |     "train.to_csv('train_modified.csv',index=False)\n",
2827 |     "test.to_csv('test_modified.csv',index=False)"
2828 |    ]
2829 |   },
2830 |   {
2831 |    "cell_type": "code",
2832 |    "execution_count": null,
2833 |    "metadata": {
2834 |     "collapsed": true
2835 |    },
2836 |    "outputs": [],
2837 |    "source": []
2838 |   }
2839 |  ],
2840 |  "metadata": {
2841 |   "kernelspec": {
2842 |    "display_name": "Python 2",
2843 |    "language": "python",
2844 |    "name": "python2"
2845 |   },
2846 |   "language_info": {
2847 |    "codemirror_mode": {
2848 |     "name": "ipython",
2849 |     "version": 2
2850 |    },
2851 |    "file_extension": ".py",
2852 |    "mimetype": "text/x-python",
2853 |    "name": "python",
2854 |    "nbconvert_exporter": "python",
2855 |    "pygments_lexer": "ipython2",
2856 |    "version": "2.7.10"
2857 |   }
2858 |  },
2859 |  "nbformat": 4,
2860 |  "nbformat_minor": 0
2861 | }
2862 | 


--------------------------------------------------------------------------------