├── .gitattributes
├── README.md
├── analysis&machine-learning
    └── jobs_ml_reg.ipynb
├── analysis&visualizations
    └── jobs_analysis.ipynb
├── city_data
    ├── city.csv
    ├── city.ipynb
    └── city.py
├── pics
    ├── 1.png
    ├── 10.png
    ├── 11.png
    ├── 12.png
    ├── 13.png
    ├── 14.png
    ├── 2.png
    ├── 3.png
    ├── 4.png
    ├── 5.png
    ├── 6.png
    ├── 7.png
    ├── 8.png
    ├── 9.png
    ├── ML部分
    │   ├── 为标签编码.png
    │   ├── 决策树训练集误差.png
    │   ├── 划分训练集与测试集.png
    │   ├── 变量重要性.png
    │   ├── 属性合并.png
    │   ├── 筛选岗位数量前150.png
    │   ├── 编码分类结果.png
    │   ├── 过滤后的分布.png
    │   ├── 过滤薪资.png
    │   ├── 重要性分析.png
    │   └── 随机森林训练集误差.png
    └── url.png
└── spiders
    ├── citylist_spider.py
    └── jobs_spider.py


/.gitattributes:
--------------------------------------------------------------------------------
1 | *.ipynb linguist-language=python
2 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | # BOSS直聘的数据分析
  2 | ---
  3 | ## 库依赖
  4 | - 基于Python3.6
  5 | - Jupyter Notebook
  6 | - pyecharts
  7 | - pymongo
  8 | - pandas
  9 | - numpy
 10 | ---
 11 | ## 爬虫实现过程
 12 | ### 分析URL
 13 | ![url](pics/url.png)
 14 | - c后的编号对应不同城市
 15 | - page后的数字则对应页码
 16 | 
 17 | ---
 18 | 
 19 | ### 爬取所有省、市对应的code
 20 | city.py实现地区与对应code的爬取
 21 | <img src="pics/1.png" width=200 height=280 />
 22 | ---
 23 | ### 根据codelist爬取所有地区的职位
 24 | - 爬取内容包含：signal、省、市、职位名称、薪资、公司名称、工作经验、学历要求、公司规模。
 25 | <img src="pics/3.png" width=300 height=40 />
 26 | - signal字段作用在于重复爬取时跳过已爬取的页面。
 27 | - 存入MongoDB中
 28 | <img src="pics/2.png" width=300 height=180 />
 29 | 
 30 | ---
 31 | ## 进行数据分析
 32 | ### 读取数据
 33 | 利用pandas读取数据库中数据
 34 | <img src="pics/4.png" width=360 height=190 />
 35 | 
 36 | ---
 37 | ### 添加新列：salary
 38 | 利用正则提取出[职位薪资]
 39 | <img src="pics/5.png" width=360 height=190 />
 40 | 
 41 | ---
 42 | ### 数据清洗 
 43 | - 移除重复数据
 44 | 
 45 | <img src="pics/6.png" width=360 height=300 />
 46 | 
 47 | ---
 48 | 
 49 | - 数据筛选过滤，去除过高和过低的薪资
 50 | <img src="pics/7.png" width=360 height=300 />
 51 | <img src="pics/9.png" width=360 height=300 />
 52 | 
 53 | - 去除与“数据分析”无关的岗位信息
 54 | 
 55 | ---
 56 | ## 数据可视化
 57 | <img src="pics/10.png" width=360 height=300 />
 58 | 
 59 | ---
 60 | 
 61 | <img src="pics/11.png" width=360 height=300 />
 62 | 
 63 | ---
 64 | 
 65 | <img src="pics/12.png" width=360 height=300 />
 66 | 
 67 | ---
 68 | 
 69 | <img src="pics/14.png" width=360 height=300 />
 70 | 
 71 | ---
 72 | 
 73 | <img src="pics/13.png" width=360 height=300 />
 74 | 
 75 | ---
 76 | 
 77 | ## 机器学习部分分析
 78 | <img src="pics/ML部分/过滤薪资.png" width=360 height=300 />
 79 | 
 80 | ---
 81 | 
 82 | <img src="pics/ML部分/过滤后的分布.png" width=360 height=300 />
 83 | 
 84 | ---
 85 | 
 86 | <img src="pics/ML部分/筛选岗位数量前150.png" width=300 height=200 />
 87 | 
 88 | ---
 89 | 
 90 | <img src="pics/ML部分/属性合并.png" width=300 height=500 />
 91 | 
 92 | ---
 93 | 
 94 | <img src="pics/ML部分/划分训练集与测试集.png" width=600 height=300 />
 95 | 
 96 | ---
 97 | 
 98 | <img src="pics/ML部分/为标签编码.png" width=350 height=400 />
 99 | 
100 | ---
101 | ### **变量重要性**
102 | 
103 | <img src="pics/ML部分/变量重要性.png" width=350 height=400 />
104 | 
105 | ---
106 | 
107 | <img src="pics/ML部分/重要性分析.png" width=500 height=150 />
108 | 


--------------------------------------------------------------------------------
/analysis&machine-learning/jobs_ml_reg.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {},
   6 |    "source": [
   7 |     "# **机器学习**"
   8 |    ]
   9 |   },
  10 |   {
  11 |    "cell_type": "code",
  12 |    "execution_count": 262,
  13 |    "metadata": {
  14 |     "collapsed": true
  15 |    },
  16 |    "outputs": [],
  17 |    "source": [
  18 |     "import pymongo\n",
  19 |     "import pandas as pd\n",
  20 |     "from pandas import Series\n",
  21 |     "client = pymongo.MongoClient('localhost',27017)\n",
  22 |     "db  = client['Graduation_project']\n",
  23 |     "table = db['jobs_info']\n",
  24 |     "data = pd.DataFrame(list(table.find()))\n",
  25 |     "del data['_id']\n",
  26 |     "del data['signal']"
  27 |    ]
  28 |   },
  29 |   {
  30 |    "cell_type": "code",
  31 |    "execution_count": 263,
  32 |    "metadata": {},
  33 |    "outputs": [
  34 |     {
  35 |      "data": {
  36 |       "text/plain": [
  37 |        "(65059, 8)"
  38 |       ]
  39 |      },
  40 |      "execution_count": 263,
  41 |      "metadata": {},
  42 |      "output_type": "execute_result"
  43 |     }
  44 |    ],
  45 |    "source": [
  46 |     "data.shape # 65059 rows × 8 columns"
  47 |    ]
  48 |   },
  49 |   {
  50 |    "cell_type": "code",
  51 |    "execution_count": 264,
  52 |    "metadata": {
  53 |     "collapsed": true
  54 |    },
  55 |    "outputs": [],
  56 |    "source": [
  57 |     "import re\n",
  58 |     "# 均值函数\n",
  59 |     "def average(job_salary):\n",
  60 |     "    # 取薪资均值----------------\n",
  61 |     "    pattern = re.compile('\\d+')\n",
  62 |     "    salary = job_salary\n",
  63 |     "    try:\n",
  64 |     "        res = re.findall(pattern, salary)\n",
  65 |     "        avg_salary = 0\n",
  66 |     "        sum = 0\n",
  67 |     "        for i in res:\n",
  68 |     "            a = int(i)\n",
  69 |     "            sum = sum + a\n",
  70 |     "            avg_salary = sum / 2\n",
  71 |     "    except Exception:\n",
  72 |     "        avg_salary = 0\n",
  73 |     "    # 函数返回值\n",
  74 |     "    return avg_salary"
  75 |    ]
  76 |   },
  77 |   {
  78 |    "cell_type": "code",
  79 |    "execution_count": 265,
  80 |    "metadata": {
  81 |     "collapsed": true
  82 |    },
  83 |    "outputs": [],
  84 |    "source": [
  85 |     "salary_list = []\n",
  86 |     "for i in range(0,65059):\n",
  87 |     "    avg_sal = average(data['职位薪资'][i])\n",
  88 |     "    salary_list.append(avg_sal)"
  89 |    ]
  90 |   },
  91 |   {
  92 |    "cell_type": "code",
  93 |    "execution_count": 266,
  94 |    "metadata": {},
  95 |    "outputs": [],
  96 |    "source": [
  97 |     "sal = Series(salary_list)"
  98 |    ]
  99 |   },
 100 |   {
 101 |    "cell_type": "code",
 102 |    "execution_count": 267,
 103 |    "metadata": {},
 104 |    "outputs": [],
 105 |    "source": [
 106 |     "data.insert(8,'salary',sal)"
 107 |    ]
 108 |   },
 109 |   {
 110 |    "cell_type": "code",
 111 |    "execution_count": 268,
 112 |    "metadata": {},
 113 |    "outputs": [],
 114 |    "source": [
 115 |     "#data"
 116 |    ]
 117 |   },
 118 |   {
 119 |    "cell_type": "code",
 120 |    "execution_count": 269,
 121 |    "metadata": {
 122 |     "scrolled": false
 123 |    },
 124 |    "outputs": [
 125 |     {
 126 |      "data": {
 127 |       "text/html": [
 128 |        "<div>\n",
 129 |        "<style scoped>\n",
 130 |        "    .dataframe tbody tr th:only-of-type {\n",
 131 |        "        vertical-align: middle;\n",
 132 |        "    }\n",
 133 |        "\n",
 134 |        "    .dataframe tbody tr th {\n",
 135 |        "        vertical-align: top;\n",
 136 |        "    }\n",
 137 |        "\n",
 138 |        "    .dataframe thead th {\n",
 139 |        "        text-align: right;\n",
 140 |        "    }\n",
 141 |        "</style>\n",
 142 |        "<table border=\"1\" class=\"dataframe\">\n",
 143 |        "  <thead>\n",
 144 |        "    <tr style=\"text-align: right;\">\n",
 145 |        "      <th></th>\n",
 146 |        "      <th>城市</th>\n",
 147 |        "      <th>职位名称</th>\n",
 148 |        "      <th>工作经验</th>\n",
 149 |        "      <th>公司规模</th>\n",
 150 |        "      <th>学历要求</th>\n",
 151 |        "      <th>salary</th>\n",
 152 |        "    </tr>\n",
 153 |        "  </thead>\n",
 154 |        "  <tbody>\n",
 155 |        "    <tr>\n",
 156 |        "      <th>0</th>\n",
 157 |        "      <td>北京</td>\n",
 158 |        "      <td>数据分析</td>\n",
 159 |        "      <td>经验不限</td>\n",
 160 |        "      <td>10000人以上</td>\n",
 161 |        "      <td>本科</td>\n",
 162 |        "      <td>8.0</td>\n",
 163 |        "    </tr>\n",
 164 |        "    <tr>\n",
 165 |        "      <th>1</th>\n",
 166 |        "      <td>北京</td>\n",
 167 |        "      <td>数据分析师</td>\n",
 168 |        "      <td>1-3年</td>\n",
 169 |        "      <td>100-499人</td>\n",
 170 |        "      <td>本科</td>\n",
 171 |        "      <td>27.5</td>\n",
 172 |        "    </tr>\n",
 173 |        "    <tr>\n",
 174 |        "      <th>2</th>\n",
 175 |        "      <td>北京</td>\n",
 176 |        "      <td>数据分析师</td>\n",
 177 |        "      <td>3-5年</td>\n",
 178 |        "      <td>1000-9999人</td>\n",
 179 |        "      <td>本科</td>\n",
 180 |        "      <td>20.0</td>\n",
 181 |        "    </tr>\n",
 182 |        "    <tr>\n",
 183 |        "      <th>3</th>\n",
 184 |        "      <td>北京</td>\n",
 185 |        "      <td>数据分析</td>\n",
 186 |        "      <td>经验不限</td>\n",
 187 |        "      <td>10000人以上</td>\n",
 188 |        "      <td>本科</td>\n",
 189 |        "      <td>9.0</td>\n",
 190 |        "    </tr>\n",
 191 |        "    <tr>\n",
 192 |        "      <th>4</th>\n",
 193 |        "      <td>北京</td>\n",
 194 |        "      <td>数据分析师</td>\n",
 195 |        "      <td>3-5年</td>\n",
 196 |        "      <td>10000人以上</td>\n",
 197 |        "      <td>本科</td>\n",
 198 |        "      <td>12.5</td>\n",
 199 |        "    </tr>\n",
 200 |        "  </tbody>\n",
 201 |        "</table>\n",
 202 |        "</div>"
 203 |       ],
 204 |       "text/plain": [
 205 |        "   城市   职位名称  工作经验        公司规模 学历要求  salary\n",
 206 |        "0  北京   数据分析  经验不限    10000人以上   本科     8.0\n",
 207 |        "1  北京  数据分析师  1-3年    100-499人   本科    27.5\n",
 208 |        "2  北京  数据分析师  3-5年  1000-9999人   本科    20.0\n",
 209 |        "3  北京   数据分析  经验不限    10000人以上   本科     9.0\n",
 210 |        "4  北京  数据分析师  3-5年    10000人以上   本科    12.5"
 211 |       ]
 212 |      },
 213 |      "execution_count": 269,
 214 |      "metadata": {},
 215 |      "output_type": "execute_result"
 216 |     }
 217 |    ],
 218 |    "source": [
 219 |     "job_data = data[['城市','职位名称','工作经验','公司规模','学历要求','salary']]\n",
 220 |     "job_data.head()"
 221 |    ]
 222 |   },
 223 |   {
 224 |    "cell_type": "code",
 225 |    "execution_count": 270,
 226 |    "metadata": {},
 227 |    "outputs": [
 228 |     {
 229 |      "data": {
 230 |       "text/plain": [
 231 |        "(65059, 6)"
 232 |       ]
 233 |      },
 234 |      "execution_count": 270,
 235 |      "metadata": {},
 236 |      "output_type": "execute_result"
 237 |     }
 238 |    ],
 239 |    "source": [
 240 |     "job_data.shape # 65059条数据"
 241 |    ]
 242 |   },
 243 |   {
 244 |    "cell_type": "markdown",
 245 |    "metadata": {},
 246 |    "source": [
 247 |     "### 薪资分布"
 248 |    ]
 249 |   },
 250 |   {
 251 |    "cell_type": "code",
 252 |    "execution_count": 271,
 253 |    "metadata": {
 254 |     "scrolled": false
 255 |    },
 256 |    "outputs": [
 257 |     {
 258 |      "data": {
 259 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfYAAAE/CAYAAAC0DOHAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
 260 |       "text/plain": [
 261 |        "<matplotlib.figure.Figure at 0x20c6a80c438>"
 262 |       ]
 263 |      },
 264 |      "metadata": {},
 265 |      "output_type": "display_data"
 266 |     }
 267 |    ],
 268 |    "source": [
 269 |     "%matplotlib inline\n",
 270 |     "import matplotlib.pyplot as plt\n",
 271 |     "job_data.hist(bins=50, figsize=(8,5))\n",
 272 |     "# save_fig(\"attribute_histogram_plots\")\n",
 273 |     "plt.show()"
 274 |    ]
 275 |   },
 276 |   {
 277 |    "cell_type": "markdown",
 278 |    "metadata": {},
 279 |    "source": [
 280 |     "# 数据清洗"
 281 |    ]
 282 |   },
 283 |   {
 284 |    "cell_type": "markdown",
 285 |    "metadata": {},
 286 |    "source": [
 287 |     "## 去重"
 288 |    ]
 289 |   },
 290 |   {
 291 |    "cell_type": "code",
 292 |    "execution_count": 272,
 293 |    "metadata": {},
 294 |    "outputs": [
 295 |     {
 296 |      "data": {
 297 |       "text/html": [
 298 |        "<div>\n",
 299 |        "<style scoped>\n",
 300 |        "    .dataframe tbody tr th:only-of-type {\n",
 301 |        "        vertical-align: middle;\n",
 302 |        "    }\n",
 303 |        "\n",
 304 |        "    .dataframe tbody tr th {\n",
 305 |        "        vertical-align: top;\n",
 306 |        "    }\n",
 307 |        "\n",
 308 |        "    .dataframe thead th {\n",
 309 |        "        text-align: right;\n",
 310 |        "    }\n",
 311 |        "</style>\n",
 312 |        "<table border=\"1\" class=\"dataframe\">\n",
 313 |        "  <thead>\n",
 314 |        "    <tr style=\"text-align: right;\">\n",
 315 |        "      <th></th>\n",
 316 |        "      <th>城市</th>\n",
 317 |        "      <th>职位名称</th>\n",
 318 |        "      <th>工作经验</th>\n",
 319 |        "      <th>公司规模</th>\n",
 320 |        "      <th>学历要求</th>\n",
 321 |        "      <th>salary</th>\n",
 322 |        "    </tr>\n",
 323 |        "  </thead>\n",
 324 |        "  <tbody>\n",
 325 |        "    <tr>\n",
 326 |        "      <th>0</th>\n",
 327 |        "      <td>北京</td>\n",
 328 |        "      <td>数据分析</td>\n",
 329 |        "      <td>经验不限</td>\n",
 330 |        "      <td>10000人以上</td>\n",
 331 |        "      <td>本科</td>\n",
 332 |        "      <td>8.0</td>\n",
 333 |        "    </tr>\n",
 334 |        "    <tr>\n",
 335 |        "      <th>1</th>\n",
 336 |        "      <td>北京</td>\n",
 337 |        "      <td>数据分析师</td>\n",
 338 |        "      <td>1-3年</td>\n",
 339 |        "      <td>100-499人</td>\n",
 340 |        "      <td>本科</td>\n",
 341 |        "      <td>27.5</td>\n",
 342 |        "    </tr>\n",
 343 |        "    <tr>\n",
 344 |        "      <th>2</th>\n",
 345 |        "      <td>北京</td>\n",
 346 |        "      <td>数据分析师</td>\n",
 347 |        "      <td>3-5年</td>\n",
 348 |        "      <td>1000-9999人</td>\n",
 349 |        "      <td>本科</td>\n",
 350 |        "      <td>20.0</td>\n",
 351 |        "    </tr>\n",
 352 |        "    <tr>\n",
 353 |        "      <th>3</th>\n",
 354 |        "      <td>北京</td>\n",
 355 |        "      <td>数据分析</td>\n",
 356 |        "      <td>经验不限</td>\n",
 357 |        "      <td>10000人以上</td>\n",
 358 |        "      <td>本科</td>\n",
 359 |        "      <td>9.0</td>\n",
 360 |        "    </tr>\n",
 361 |        "    <tr>\n",
 362 |        "      <th>4</th>\n",
 363 |        "      <td>北京</td>\n",
 364 |        "      <td>数据分析师</td>\n",
 365 |        "      <td>3-5年</td>\n",
 366 |        "      <td>10000人以上</td>\n",
 367 |        "      <td>本科</td>\n",
 368 |        "      <td>12.5</td>\n",
 369 |        "    </tr>\n",
 370 |        "  </tbody>\n",
 371 |        "</table>\n",
 372 |        "</div>"
 373 |       ],
 374 |       "text/plain": [
 375 |        "   城市   职位名称  工作经验        公司规模 学历要求  salary\n",
 376 |        "0  北京   数据分析  经验不限    10000人以上   本科     8.0\n",
 377 |        "1  北京  数据分析师  1-3年    100-499人   本科    27.5\n",
 378 |        "2  北京  数据分析师  3-5年  1000-9999人   本科    20.0\n",
 379 |        "3  北京   数据分析  经验不限    10000人以上   本科     9.0\n",
 380 |        "4  北京  数据分析师  3-5年    10000人以上   本科    12.5"
 381 |       ]
 382 |      },
 383 |      "execution_count": 272,
 384 |      "metadata": {},
 385 |      "output_type": "execute_result"
 386 |     }
 387 |    ],
 388 |    "source": [
 389 |     "jobs = job_data.drop_duplicates()\n",
 390 |     "jobs.head()"
 391 |    ]
 392 |   },
 393 |   {
 394 |    "cell_type": "code",
 395 |    "execution_count": 273,
 396 |    "metadata": {},
 397 |    "outputs": [
 398 |     {
 399 |      "data": {
 400 |       "text/plain": [
 401 |        "(59142, 6)"
 402 |       ]
 403 |      },
 404 |      "execution_count": 273,
 405 |      "metadata": {},
 406 |      "output_type": "execute_result"
 407 |     }
 408 |    ],
 409 |    "source": [
 410 |     "jobs.shape # 61759条数据"
 411 |    ]
 412 |   },
 413 |   {
 414 |    "cell_type": "markdown",
 415 |    "metadata": {},
 416 |    "source": [
 417 |     "## 过滤"
 418 |    ]
 419 |   },
 420 |   {
 421 |    "cell_type": "code",
 422 |    "execution_count": 274,
 423 |    "metadata": {},
 424 |    "outputs": [],
 425 |    "source": [
 426 |     "jobs_copy = jobs.copy()"
 427 |    ]
 428 |   },
 429 |   {
 430 |    "cell_type": "markdown",
 431 |    "metadata": {},
 432 |    "source": [
 433 |     "### 过滤出薪资在6-27k之间"
 434 |    ]
 435 |   },
 436 |   {
 437 |    "cell_type": "code",
 438 |    "execution_count": 275,
 439 |    "metadata": {},
 440 |    "outputs": [],
 441 |    "source": [
 442 |     "jobs_copy = jobs_copy[jobs_copy['salary']<27]"
 443 |    ]
 444 |   },
 445 |   {
 446 |    "cell_type": "code",
 447 |    "execution_count": 276,
 448 |    "metadata": {},
 449 |    "outputs": [
 450 |     {
 451 |      "data": {
 452 |       "text/html": [
 453 |        "<div>\n",
 454 |        "<style scoped>\n",
 455 |        "    .dataframe tbody tr th:only-of-type {\n",
 456 |        "        vertical-align: middle;\n",
 457 |        "    }\n",
 458 |        "\n",
 459 |        "    .dataframe tbody tr th {\n",
 460 |        "        vertical-align: top;\n",
 461 |        "    }\n",
 462 |        "\n",
 463 |        "    .dataframe thead th {\n",
 464 |        "        text-align: right;\n",
 465 |        "    }\n",
 466 |        "</style>\n",
 467 |        "<table border=\"1\" class=\"dataframe\">\n",
 468 |        "  <thead>\n",
 469 |        "    <tr style=\"text-align: right;\">\n",
 470 |        "      <th></th>\n",
 471 |        "      <th>城市</th>\n",
 472 |        "      <th>职位名称</th>\n",
 473 |        "      <th>工作经验</th>\n",
 474 |        "      <th>公司规模</th>\n",
 475 |        "      <th>学历要求</th>\n",
 476 |        "      <th>salary</th>\n",
 477 |        "    </tr>\n",
 478 |        "  </thead>\n",
 479 |        "  <tbody>\n",
 480 |        "    <tr>\n",
 481 |        "      <th>0</th>\n",
 482 |        "      <td>北京</td>\n",
 483 |        "      <td>数据分析</td>\n",
 484 |        "      <td>经验不限</td>\n",
 485 |        "      <td>10000人以上</td>\n",
 486 |        "      <td>本科</td>\n",
 487 |        "      <td>8.0</td>\n",
 488 |        "    </tr>\n",
 489 |        "    <tr>\n",
 490 |        "      <th>2</th>\n",
 491 |        "      <td>北京</td>\n",
 492 |        "      <td>数据分析师</td>\n",
 493 |        "      <td>3-5年</td>\n",
 494 |        "      <td>1000-9999人</td>\n",
 495 |        "      <td>本科</td>\n",
 496 |        "      <td>20.0</td>\n",
 497 |        "    </tr>\n",
 498 |        "    <tr>\n",
 499 |        "      <th>3</th>\n",
 500 |        "      <td>北京</td>\n",
 501 |        "      <td>数据分析</td>\n",
 502 |        "      <td>经验不限</td>\n",
 503 |        "      <td>10000人以上</td>\n",
 504 |        "      <td>本科</td>\n",
 505 |        "      <td>9.0</td>\n",
 506 |        "    </tr>\n",
 507 |        "    <tr>\n",
 508 |        "      <th>4</th>\n",
 509 |        "      <td>北京</td>\n",
 510 |        "      <td>数据分析师</td>\n",
 511 |        "      <td>3-5年</td>\n",
 512 |        "      <td>10000人以上</td>\n",
 513 |        "      <td>本科</td>\n",
 514 |        "      <td>12.5</td>\n",
 515 |        "    </tr>\n",
 516 |        "    <tr>\n",
 517 |        "      <th>8</th>\n",
 518 |        "      <td>北京</td>\n",
 519 |        "      <td>数据分析师</td>\n",
 520 |        "      <td>1-3年</td>\n",
 521 |        "      <td>10000人以上</td>\n",
 522 |        "      <td>本科</td>\n",
 523 |        "      <td>25.5</td>\n",
 524 |        "    </tr>\n",
 525 |        "  </tbody>\n",
 526 |        "</table>\n",
 527 |        "</div>"
 528 |       ],
 529 |       "text/plain": [
 530 |        "   城市   职位名称  工作经验        公司规模 学历要求  salary\n",
 531 |        "0  北京   数据分析  经验不限    10000人以上   本科     8.0\n",
 532 |        "2  北京  数据分析师  3-5年  1000-9999人   本科    20.0\n",
 533 |        "3  北京   数据分析  经验不限    10000人以上   本科     9.0\n",
 534 |        "4  北京  数据分析师  3-5年    10000人以上   本科    12.5\n",
 535 |        "8  北京  数据分析师  1-3年    10000人以上   本科    25.5"
 536 |       ]
 537 |      },
 538 |      "execution_count": 276,
 539 |      "metadata": {},
 540 |      "output_type": "execute_result"
 541 |     }
 542 |    ],
 543 |    "source": [
 544 |     "jobs_copy = jobs_copy[jobs_copy['salary']>6]\n",
 545 |     "jobs_copy.head()"
 546 |    ]
 547 |   },
 548 |   {
 549 |    "cell_type": "code",
 550 |    "execution_count": 277,
 551 |    "metadata": {},
 552 |    "outputs": [
 553 |     {
 554 |      "data": {
 555 |       "text/plain": [
 556 |        "(27293, 6)"
 557 |       ]
 558 |      },
 559 |      "execution_count": 277,
 560 |      "metadata": {},
 561 |      "output_type": "execute_result"
 562 |     }
 563 |    ],
 564 |    "source": [
 565 |     "jobs_copy.shape"
 566 |    ]
 567 |   },
 568 |   {
 569 |    "cell_type": "code",
 570 |    "execution_count": 278,
 571 |    "metadata": {
 572 |     "scrolled": true
 573 |    },
 574 |    "outputs": [
 575 |     {
 576 |      "data": {
 577 |       "text/html": [
 578 |        "<div>\n",
 579 |        "<style scoped>\n",
 580 |        "    .dataframe tbody tr th:only-of-type {\n",
 581 |        "        vertical-align: middle;\n",
 582 |        "    }\n",
 583 |        "\n",
 584 |        "    .dataframe tbody tr th {\n",
 585 |        "        vertical-align: top;\n",
 586 |        "    }\n",
 587 |        "\n",
 588 |        "    .dataframe thead th {\n",
 589 |        "        text-align: right;\n",
 590 |        "    }\n",
 591 |        "</style>\n",
 592 |        "<table border=\"1\" class=\"dataframe\">\n",
 593 |        "  <thead>\n",
 594 |        "    <tr style=\"text-align: right;\">\n",
 595 |        "      <th></th>\n",
 596 |        "      <th>salary</th>\n",
 597 |        "    </tr>\n",
 598 |        "  </thead>\n",
 599 |        "  <tbody>\n",
 600 |        "    <tr>\n",
 601 |        "      <th>count</th>\n",
 602 |        "      <td>27293.000000</td>\n",
 603 |        "    </tr>\n",
 604 |        "    <tr>\n",
 605 |        "      <th>mean</th>\n",
 606 |        "      <td>10.915528</td>\n",
 607 |        "    </tr>\n",
 608 |        "    <tr>\n",
 609 |        "      <th>std</th>\n",
 610 |        "      <td>4.546630</td>\n",
 611 |        "    </tr>\n",
 612 |        "    <tr>\n",
 613 |        "      <th>min</th>\n",
 614 |        "      <td>6.500000</td>\n",
 615 |        "    </tr>\n",
 616 |        "    <tr>\n",
 617 |        "      <th>25%</th>\n",
 618 |        "      <td>7.500000</td>\n",
 619 |        "    </tr>\n",
 620 |        "    <tr>\n",
 621 |        "      <th>50%</th>\n",
 622 |        "      <td>9.000000</td>\n",
 623 |        "    </tr>\n",
 624 |        "    <tr>\n",
 625 |        "      <th>75%</th>\n",
 626 |        "      <td>12.500000</td>\n",
 627 |        "    </tr>\n",
 628 |        "    <tr>\n",
 629 |        "      <th>max</th>\n",
 630 |        "      <td>26.500000</td>\n",
 631 |        "    </tr>\n",
 632 |        "  </tbody>\n",
 633 |        "</table>\n",
 634 |        "</div>"
 635 |       ],
 636 |       "text/plain": [
 637 |        "             salary\n",
 638 |        "count  27293.000000\n",
 639 |        "mean      10.915528\n",
 640 |        "std        4.546630\n",
 641 |        "min        6.500000\n",
 642 |        "25%        7.500000\n",
 643 |        "50%        9.000000\n",
 644 |        "75%       12.500000\n",
 645 |        "max       26.500000"
 646 |       ]
 647 |      },
 648 |      "execution_count": 278,
 649 |      "metadata": {},
 650 |      "output_type": "execute_result"
 651 |     }
 652 |    ],
 653 |    "source": [
 654 |     "jobs_copy.describe()"
 655 |    ]
 656 |   },
 657 |   {
 658 |    "cell_type": "markdown",
 659 |    "metadata": {},
 660 |    "source": [
 661 |     "### 过滤后的薪资分布图"
 662 |    ]
 663 |   },
 664 |   {
 665 |    "cell_type": "code",
 666 |    "execution_count": 279,
 667 |    "metadata": {},
 668 |    "outputs": [
 669 |     {
 670 |      "data": {
 671 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAE/CAYAAACqxdFzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
 672 |       "text/plain": [
 673 |        "<matplotlib.figure.Figure at 0x20c658d55f8>"
 674 |       ]
 675 |      },
 676 |      "metadata": {},
 677 |      "output_type": "display_data"
 678 |     }
 679 |    ],
 680 |    "source": [
 681 |     "%matplotlib inline\n",
 682 |     "import matplotlib.pyplot as plt\n",
 683 |     "jobs_copy.hist(bins=50, figsize=(8,5))\n",
 684 |     "plt.show()"
 685 |    ]
 686 |   },
 687 |   {
 688 |    "cell_type": "markdown",
 689 |    "metadata": {},
 690 |    "source": [
 691 |     "## 不相关职位筛选"
 692 |    ]
 693 |   },
 694 |   {
 695 |    "cell_type": "markdown",
 696 |    "metadata": {},
 697 |    "source": [
 698 |     "### 筛选出职位名称包含：商业|数据分析|挖掘|分析|BI|BA|数据"
 699 |    ]
 700 |   },
 701 |   {
 702 |    "cell_type": "code",
 703 |    "execution_count": 280,
 704 |    "metadata": {},
 705 |    "outputs": [
 706 |     {
 707 |      "data": {
 708 |       "text/plain": [
 709 |        "数据分析师    5114\n",
 710 |        "电商运营     1144\n",
 711 |        "数据分析     1049\n",
 712 |        "产品经理      881\n",
 713 |        "网络推广      609\n",
 714 |        "Name: 职位名称, dtype: int64"
 715 |       ]
 716 |      },
 717 |      "execution_count": 280,
 718 |      "metadata": {},
 719 |      "output_type": "execute_result"
 720 |     }
 721 |    ],
 722 |    "source": [
 723 |     "jobs_copy['职位名称'].value_counts().head()"
 724 |    ]
 725 |   },
 726 |   {
 727 |    "cell_type": "code",
 728 |    "execution_count": 281,
 729 |    "metadata": {
 730 |     "collapsed": true
 731 |    },
 732 |    "outputs": [],
 733 |    "source": [
 734 |     "da_test = jobs_copy.astype(str)\n",
 735 |     "\n",
 736 |     "temp_0 = da_test[\n",
 737 |     "    da_test['职位名称'].str.contains('商业|数据分析|挖掘|分析|BI|BA|数据')\n",
 738 |     "            ]\n",
 739 |     "\n",
 740 |     "#data_analysis = jobs_result['职位名称']\n",
 741 |     "temp = list(temp_0['职位名称']) # 将含有关键字的列表表示为test1\n",
 742 |     "jobs_data = da_test[da_test['职位名称'].isin(temp)]"
 743 |    ]
 744 |   },
 745 |   {
 746 |    "cell_type": "code",
 747 |    "execution_count": 282,
 748 |    "metadata": {},
 749 |    "outputs": [],
 750 |    "source": [
 751 |     "#jobs_data['职位名称'].value_counts()"
 752 |    ]
 753 |   },
 754 |   {
 755 |    "cell_type": "code",
 756 |    "execution_count": 283,
 757 |    "metadata": {},
 758 |    "outputs": [
 759 |     {
 760 |      "data": {
 761 |       "text/plain": [
 762 |        "(13716, 6)"
 763 |       ]
 764 |      },
 765 |      "execution_count": 283,
 766 |      "metadata": {},
 767 |      "output_type": "execute_result"
 768 |     }
 769 |    ],
 770 |    "source": [
 771 |     "jobs_data.shape"
 772 |    ]
 773 |   },
 774 |   {
 775 |    "cell_type": "code",
 776 |    "execution_count": 284,
 777 |    "metadata": {},
 778 |    "outputs": [
 779 |     {
 780 |      "data": {
 781 |       "text/html": [
 782 |        "<div>\n",
 783 |        "<style scoped>\n",
 784 |        "    .dataframe tbody tr th:only-of-type {\n",
 785 |        "        vertical-align: middle;\n",
 786 |        "    }\n",
 787 |        "\n",
 788 |        "    .dataframe tbody tr th {\n",
 789 |        "        vertical-align: top;\n",
 790 |        "    }\n",
 791 |        "\n",
 792 |        "    .dataframe thead th {\n",
 793 |        "        text-align: right;\n",
 794 |        "    }\n",
 795 |        "</style>\n",
 796 |        "<table border=\"1\" class=\"dataframe\">\n",
 797 |        "  <thead>\n",
 798 |        "    <tr style=\"text-align: right;\">\n",
 799 |        "      <th></th>\n",
 800 |        "      <th>城市</th>\n",
 801 |        "      <th>职位名称</th>\n",
 802 |        "      <th>工作经验</th>\n",
 803 |        "      <th>公司规模</th>\n",
 804 |        "      <th>学历要求</th>\n",
 805 |        "      <th>salary</th>\n",
 806 |        "    </tr>\n",
 807 |        "  </thead>\n",
 808 |        "  <tbody>\n",
 809 |        "    <tr>\n",
 810 |        "      <th>0</th>\n",
 811 |        "      <td>北京</td>\n",
 812 |        "      <td>数据分析</td>\n",
 813 |        "      <td>经验不限</td>\n",
 814 |        "      <td>10000人以上</td>\n",
 815 |        "      <td>本科</td>\n",
 816 |        "      <td>8.0</td>\n",
 817 |        "    </tr>\n",
 818 |        "    <tr>\n",
 819 |        "      <th>2</th>\n",
 820 |        "      <td>北京</td>\n",
 821 |        "      <td>数据分析师</td>\n",
 822 |        "      <td>3-5年</td>\n",
 823 |        "      <td>1000-9999人</td>\n",
 824 |        "      <td>本科</td>\n",
 825 |        "      <td>20.0</td>\n",
 826 |        "    </tr>\n",
 827 |        "    <tr>\n",
 828 |        "      <th>3</th>\n",
 829 |        "      <td>北京</td>\n",
 830 |        "      <td>数据分析</td>\n",
 831 |        "      <td>经验不限</td>\n",
 832 |        "      <td>10000人以上</td>\n",
 833 |        "      <td>本科</td>\n",
 834 |        "      <td>9.0</td>\n",
 835 |        "    </tr>\n",
 836 |        "    <tr>\n",
 837 |        "      <th>4</th>\n",
 838 |        "      <td>北京</td>\n",
 839 |        "      <td>数据分析师</td>\n",
 840 |        "      <td>3-5年</td>\n",
 841 |        "      <td>10000人以上</td>\n",
 842 |        "      <td>本科</td>\n",
 843 |        "      <td>12.5</td>\n",
 844 |        "    </tr>\n",
 845 |        "    <tr>\n",
 846 |        "      <th>8</th>\n",
 847 |        "      <td>北京</td>\n",
 848 |        "      <td>数据分析师</td>\n",
 849 |        "      <td>1-3年</td>\n",
 850 |        "      <td>10000人以上</td>\n",
 851 |        "      <td>本科</td>\n",
 852 |        "      <td>25.5</td>\n",
 853 |        "    </tr>\n",
 854 |        "  </tbody>\n",
 855 |        "</table>\n",
 856 |        "</div>"
 857 |       ],
 858 |       "text/plain": [
 859 |        "   城市   职位名称  工作经验        公司规模 学历要求 salary\n",
 860 |        "0  北京   数据分析  经验不限    10000人以上   本科    8.0\n",
 861 |        "2  北京  数据分析师  3-5年  1000-9999人   本科   20.0\n",
 862 |        "3  北京   数据分析  经验不限    10000人以上   本科    9.0\n",
 863 |        "4  北京  数据分析师  3-5年    10000人以上   本科   12.5\n",
 864 |        "8  北京  数据分析师  1-3年    10000人以上   本科   25.5"
 865 |       ]
 866 |      },
 867 |      "execution_count": 284,
 868 |      "metadata": {},
 869 |      "output_type": "execute_result"
 870 |     }
 871 |    ],
 872 |    "source": [
 873 |     "dt_test = jobs_data.astype(str)\n",
 874 |     "y = dt_test[\n",
 875 |     "    dt_test['职位名称'].str.contains('转行')\n",
 876 |     "           ]\n",
 877 |     "\n",
 878 |     "test1 = list(y['职位名称']) # 将含有关键字的列表表示为test1\n",
 879 |     "test2 = list(dt_test['职位名称']) # 将全部表示为test2\n",
 880 |     "ret = list(set(test2) ^ set(test1)) # 列表求差集的方法将含有关键字的行除去\n",
 881 |     "\n",
 882 |     "jobs_data = dt_test[dt_test['职位名称'].isin(ret)]\n",
 883 |     "jobs_data.head()"
 884 |    ]
 885 |   },
 886 |   {
 887 |    "cell_type": "code",
 888 |    "execution_count": 285,
 889 |    "metadata": {},
 890 |    "outputs": [
 891 |     {
 892 |      "name": "stdout",
 893 |      "output_type": "stream",
 894 |      "text": [
 895 |       "<class 'pandas.core.frame.DataFrame'>\n",
 896 |       "Int64Index: 13665 entries, 0 to 65053\n",
 897 |       "Data columns (total 6 columns):\n",
 898 |       "城市        13665 non-null object\n",
 899 |       "职位名称      13665 non-null object\n",
 900 |       "工作经验      13665 non-null object\n",
 901 |       "公司规模      13665 non-null object\n",
 902 |       "学历要求      13665 non-null object\n",
 903 |       "salary    13665 non-null float64\n",
 904 |       "dtypes: float64(1), object(5)\n",
 905 |       "memory usage: 747.3+ KB\n"
 906 |      ]
 907 |     }
 908 |    ],
 909 |    "source": [
 910 |     "jobs_data_copy = jobs_data.copy()\n",
 911 |     "jobs_data_copy['salary'] = pd.to_numeric(jobs_data_copy['salary'])\n",
 912 |     "jobs_data_copy.info()"
 913 |    ]
 914 |   },
 915 |   {
 916 |    "cell_type": "code",
 917 |    "execution_count": 286,
 918 |    "metadata": {},
 919 |    "outputs": [
 920 |     {
 921 |      "data": {
 922 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAE/CAYAAACqxdFzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAHGNJREFUeJzt3X+Q3Hd93/HnOxZ2DDKWjMvVkTSR\nE1QnTlQSczUmpPSEEv9ksNuJW7sOlsAdDcFQCGKwCDOYIUMrN3UokJRUYA1y6lgm/Kg1WBQ0hivD\nDHZBBCwbQyQ7in2ykADZMsImVPDuH/u5sD7t3u7d7u3tffx8zNzc7uf72e/3897vfu913+9+97uR\nmUiSpLr83HwPQJIk9Z8BL0lShQx4SZIqZMBLklQhA16SpAoZ8JIkVciAlwRARGREvGi+xyGpPwx4\nSZIqZMBL6klELJrvMUg6kQEvVSgiboiIAxHxg4j4dkSsjYjzI+LLEfFERByMiD+LiJPbPP6yiPib\niHgyIh6NiHc3TVtZDudfFxGPAJ+PiLsi4k1T5nFfRFwxt5VKaseAlyoTEecAbwT+RWaeBlwE7Ad+\nAvwhcCbwMmAt8IY2s/khcC2wBLgM+IMWYf2vgF8t898G/H7TGF4MLAN29qUoSTNmwEv1+QlwCnBu\nRDwnM/dn5kOZuTsz78nM45m5H/gfNEL6BJk5npl7MvOnmXkfcHuLvu/OzB9m5tPAncCqiFhVpr0G\nuCMzfzwXBUrqzICXKpOZ+4C3AO8GDkfE9oj4hYj4ZxHx6Yj4TkQ8CfwnGnvzJ4iIl0bEFyLiuxFx\nFHh9i76PNi3zH4CPAb8fET8HXA38Zd+Lk9Q1A16qUGb+VWb+NvCLQAI3AR8CvgWsysznA38ERJtZ\n/BWwA1iRmacDf9Gi79SvotwGXEPj0P9TmfnlftQiaXYMeKkyEXFORLwyIk4BfgQ8TeOw/WnAk8Cx\niPgV4A+mmc1pwJHM/FFEnA/8+07LLYH+U+Bm3HuX5p0BL9XnFGAz8D3gO8ALaeytv41GUP8A+DBw\nxzTzeAPwnoj4AfAuGoffu3ErsBr4n7MauaS+icypR9kkaXYi4lpgQ3l7QNI8cg9eUl9ExHNp7Plv\nme+xSDLgJfVBRFwEfBc4ROMEPUnzzEP0kiRVyD14SZIqZMBLklShof4WqDPPPDNXrlw538OY1g9/\n+EOe97znzfcw+qKWWmqpA6xlWNVSSy11QF217N69+3uZ+U96nc9QB/zKlSv56le/Ot/DmNb4+Dhj\nY2PzPYy+qKWWWuoAaxlWtdRSSx1QVy0R8ff9mI+H6CVJqlDHgI+IrRFxOCLubzHtbeV7oc8s9yMi\nPhAR+8p3QZ/X1HddROwtP+v6W4YkSWrWzR78R4GLpzZGxArgd4FHmpovAVaVnw00vtyCiDgDuBF4\nKXA+cGNELO1l4JIkqb2OAZ+ZXwSOtJj0PuDtPPMbpS4Hbs2Ge4AlEXEWcBGwKzOPZObjwC5a/NMg\nSZL6Y1bvwUfEq4EDmfmNKZOW0fQd0cBEaWvXLkmS5sCMz6Iv15t+J3Bhq8kt2nKa9lbz30Dj8D4j\nIyOMj4/PdIgDdezYsaEfY7dqqaWWOsBahlUttdRSB9RVS7/M5mNyvwycDXwjIgCWA18r3xk9Aaxo\n6rsceKy0j01pH28188zcQvmyitHR0Rz2jz3U9NGMWmqppQ6wlmFVSy211AF11dIvMz5En5l7MvOF\nmbkyM1fSCO/zMvM7wA7g2nI2/QXA0cw8CHwWuDAilpaT6y4sbZIkaQ508zG524EvA+dExEREXDdN\n953Aw8A+4MM0vjqSzDwC/DHwlfLzntImSZLmQMdD9Jl5dYfpK5tuJ3B9m35bga0zHJ8kSZoFr2Qn\nSVKFhvpa9NKz1cpNdwGwcfVx1pfbzfZvvmzQQ5K0wLgHL0lShQx4SZIqZMBLklQhA16SpAoZ8JIk\nVciAlySpQga8JEkVMuAlSaqQAS9JUoUMeEmSKmTAS5JUIQNekqQKGfCSJFXIgJckqUIGvCRJFTLg\nJUmqkAEvSVKFDHhJkipkwEuSVCEDXpKkChnwkiRVyICXJKlCBrwkSRUy4CVJqpABL0lShQx4SZIq\nZMBLklShjgEfEVsj4nBE3N/U9icR8a2IuC8iPhURS5qmvSMi9kXEtyPioqb2i0vbvojY1P9SJEnS\npG724D8KXDylbRfw65n5z4G/Bd4BEBHnAlcBv1Ye898j4qSIOAn4c+AS4Fzg6tJXkiTNgY4Bn5lf\nBI5MaftcZh4vd+8BlpfblwPbM/MfMvPvgH3A+eVnX2Y+nJk/BraXvpIkaQ704z341wGfKbeXAY82\nTZsobe3aJUnSHFjUy4Mj4p3AceC2yaYW3ZLW/0hkm3luADYAjIyMMD4+3ssQ59yxY8eGfozdqqWW\nGurYuLpxgGzk1J/dbrYQ66thvUyqpZZa6oC6aumXWQd8RKwDXgWszczJsJ4AVjR1Ww48Vm63a3+G\nzNwCbAEYHR3NsbGx2Q5xIMbHxxn2MXarllpqqGP9pruARrjfvOfEzXT/NWMDHlHvalgvk2qppZY6\noK5a+mVWh+gj4mLgBuDVmflU06QdwFURcUpEnA2sAv4v8BVgVUScHREn0zgRb0dvQ5ckSe103IOP\niNuBMeDMiJgAbqRx1vwpwK6IALgnM1+fmQ9ExMeAb9I4dH99Zv6kzOeNwGeBk4CtmfnAHNQjSZLo\nIuAz8+oWzbdM0/+9wHtbtO8Eds5odJIkaVa8kp0kSRUy4CVJqpABL0lShQx4SZIqZMBLklQhA16S\npAoZ8JIkVciAlySpQga8JEkVMuAlSaqQAS9JUoUMeEmSKmTAS5JUIQNekqQKGfCSJFXIgJckqUIG\nvCRJFTLgJUmqkAEvSVKFDHhJkipkwEuSVCEDXpKkChnwkiRVyICXJKlCBrwkSRUy4CVJqpABL0lS\nhQx4SZIqZMBLklShjgEfEVsj4nBE3N/UdkZE7IqIveX30tIeEfGBiNgXEfdFxHlNj1lX+u+NiHVz\nU44kSQJY1EWfjwJ/Btza1LYJuDszN0fEpnL/BuASYFX5eSnwIeClEXEGcCMwCiSwOyJ2ZObj/Spk\nUFZuuusZ9zeuPs76prb9my8b9JAkSTpBxz34zPwicGRK8+XAtnJ7G3BFU/ut2XAPsCQizgIuAnZl\n5pES6ruAi/tRgCRJOtFs34MfycyDAOX3C0v7MuDRpn4Tpa1duyRJmgPdHKKfiWjRltO0nziDiA3A\nBoCRkRHGx8f7Nrh+2Lj6+DPuj5z6zLZhG+9MHDt2bEGPf1INdUy+pqa+viYtxPpqWC+Taqmlljqg\nrlr6ZbYBfygizsrMg+UQ/OHSPgGsaOq3HHistI9NaR9vNePM3AJsARgdHc2xsbFW3ebN+hbvwd+8\n52dP4/5rxgY8ov4ZHx9n2J7v2aihjsnX2dTX16SF+DqrYb1MqqWWWuqAumrpl9keot8BTJ4Jvw64\ns6n92nI2/QXA0XII/7PAhRGxtJxxf2FpkyRJc6DjHnxE3E5j7/vMiJigcTb8ZuBjEXEd8AhwZem+\nE7gU2Ac8BbwWIDOPRMQfA18p/d6TmVNP3JMkSX3SMeAz8+o2k9a26JvA9W3msxXYOqPRSZKkWfFK\ndpIkVciAlySpQga8JEkVMuAlSaqQAS9JUoUMeEmSKmTAS5JUoX5fi37oTf2616n8uldJUg3cg5ck\nqUIGvCRJFTLgJUmqkAEvSVKFDHhJkipkwEuSVKFn3cfkOun0MTpJqsWeA0dZ70eHq+UevCRJFXIP\nvs+8kI4kaRi4By9JUoXcg19gujlHwKMEkiT34CVJqpABL0lShQx4SZIqZMBLklQhA16SpAoZ8JIk\nVciAlySpQga8JEkVMuAlSaqQV7LTwHm9fkmaez3twUfEH0bEAxFxf0TcHhE/HxFnR8S9EbE3Iu6I\niJNL31PK/X1l+sp+FCBJkk4064CPiGXAfwRGM/PXgZOAq4CbgPdl5irgceC68pDrgMcz80XA+0o/\nSZI0B3p9D34RcGpELAKeCxwEXgl8vEzfBlxRbl9e7lOmr42I6HH5kiSphVkHfGYeAP4r8AiNYD8K\n7AaeyMzjpdsEsKzcXgY8Wh57vPR/wWyXL0mS2ovMnN0DI5YCnwD+HfAE8Nfl/o3lMDwRsQLYmZmr\nI+IB4KLMnCjTHgLOz8zvT5nvBmADwMjIyEu2b98+q/G1s+fA0b7Ob+RUOPR09/1XLzu9p+V1M/7Z\nLuPYsWMsXrx4Vo+diU419PocDaqOuTT5HLV7ffX6HM2HGtbLpFpqOXzkaMe/XwvltVbLOgFYs2bN\n7swc7XU+vZxF/zvA32XmdwEi4pPAbwFLImJR2UtfDjxW+k8AK4CJckj/dODI1Jlm5hZgC8Do6GiO\njY31MMQTre/i+9RnYuPq49y8p/uncf81Yz0tr5vxz3YZ4+Pj9Pv5bqVTDb0+R4OqYy5NPkftXl+9\nPkfzoYb1MqmWWj54250d/34tlNdaLeukn3p5D/4R4IKIeG55L30t8E3gC8DvlT7rgDvL7R3lPmX6\n53O2hw8kSdK0enkP/l4aJ8t9DdhT5rUFuAF4a0Tso/Ee+y3lIbcALyjtbwU29TBuSZI0jZ4udJOZ\nNwI3Tml+GDi/Rd8fAVf2sjxJktQdL1UrSVKFDHhJkipkwEuSVCG/bEbV2XPgaOeP4vmFNpIq5x68\nJEkVMuAlSaqQAS9JUoUMeEmSKmTAS5JUIQNekqQKGfCSJFXIgJckqUIGvCRJFTLgJUmqkAEvSVKF\nDHhJkipkwEuSVCEDXpKkChnwkiRVyICXJKlCBrwkSRVaNN8DkKZauemuaafv33zZgEYiSQuXe/CS\nJFXIgJckqUIGvCRJFTLgJUmqkAEvSVKFDHhJkipkwEuSVKGeAj4ilkTExyPiWxHxYES8LCLOiIhd\nEbG3/F5a+kZEfCAi9kXEfRFxXn9KkCRJU/W6B/9+4H9n5q8ALwYeBDYBd2fmKuDuch/gEmBV+dkA\nfKjHZUuSpDZmHfAR8XzgFcAtAJn548x8Argc2Fa6bQOuKLcvB27NhnuAJRFx1qxHLkmS2orMnN0D\nI34D2AJ8k8be+27gzcCBzFzS1O/xzFwaEZ8GNmfml0r73cANmfnVKfPdQGMPn5GRkZds3759VuNr\nZ8+Bo32d38ipcOjp7vuvXnZ6T8vrZvydltFuHpO19DrGTnpdB53Gd/jI0Y7rZK5r7NXkc9Tu9TXs\n42/l2LFjLF68eL6H0Re11FLDtjKplnUCsGbNmt2ZOdrrfHq5Fv0i4DzgTZl5b0S8n58djm8lWrSd\n8N9FZm6h8Y8Do6OjOTY21sMQT7S+w3XOZ2rj6uPcvKf7p3H/NWM9La+b8XdaRrt5TNbS6xg76XUd\ndBrfB2+7s+M6mesaezX5HLV7fQ37+FsZHx+n39vzfKmllhq2lUm1rJN+6uU9+AlgIjPvLfc/TiPw\nD00eei+/Dzf1X9H0+OXAYz0sX5IktTHrgM/M7wCPRsQ5pWktjcP1O4B1pW0dcGe5vQO4tpxNfwFw\nNDMPznb5kiSpvV6/LvZNwG0RcTLwMPBaGv80fCwirgMeAa4sfXcClwL7gKdKX0mSNAd6CvjM/DrQ\n6kSAtS36JnB9L8uTJEnd8Up2kiRVyICXJKlCBrwkSRUy4CVJqlCvZ9FLUrVWTnNRqPWb7mL/5ssG\nPCKpe+7BS5JUIQNekqQKGfCSJFXIgJckqUIGvCRJFTLgJUmqkAEvSVKFDHhJkipkwEuSVCEDXpKk\nChnwkiRVyICXJKlCBrwkSRUy4CVJqpABL0lShQx4SZIqZMBLklQhA16SpAoZ8JIkVciAlySpQga8\nJEkVMuAlSaqQAS9JUoV6DviIOCki/iYiPl3unx0R90bE3oi4IyJOLu2nlPv7yvSVvS5bkiS11o89\n+DcDDzbdvwl4X2auAh4Hrivt1wGPZ+aLgPeVfpIkaQ70FPARsRy4DPhIuR/AK4GPly7bgCvK7cvL\nfcr0taW/JEnqs1734P8b8Hbgp+X+C4AnMvN4uT8BLCu3lwGPApTpR0t/SZLUZ5GZs3tgxKuASzPz\nDRExBrwNeC3w5XIYnohYAezMzNUR8QBwUWZOlGkPAedn5venzHcDsAFgZGTkJdu3b59dZW3sOXC0\nr/MbORUOPd19/9XLTu9ped2Mv9My2s1jspZex9hJr+ug0/gOHznacZ3MdY29mnyO2r2+hn38rRw7\ndozFixfP9zBmZL63lblWw7YyaSG+vtpZs2bN7swc7XU+i3p47MuBV0fEpcDPA8+nsUe/JCIWlb30\n5cBjpf8EsAKYiIhFwOnAkakzzcwtwBaA0dHRHBsb62GIJ1q/6a6+zm/j6uPcvKf7p3H/NWM9La+b\n8XdaRrt5TNbS6fErO4xh/+bLZrX8bnUa3wdvu7PjOul1Pcy1yeeo3etr2Mffyvj4OP3enudar9vK\nsKthW5m0EF9fc23Wh+gz8x2ZuTwzVwJXAZ/PzGuALwC/V7qtA+4st3eU+5Tpn8/ZHj6QJEnTmovP\nwd8AvDUi9tF4j/2W0n4L8ILS/lZg0xwsW5Ik0dsh+n+UmePAeLn9MHB+iz4/Aq7sx/IkSdL0vJKd\nJEkVMuAlSaqQAS9JUoUMeEmSKtSXk+zUP50+Yy5JUjcMeEnzotcLJkmanofoJUmqkAEvSVKFDHhJ\nkipkwEuSVCEDXpKkChnwkiRVyICXJKlCBrwkSRUy4CVJqpBXspM0J6ZeqW7j6uOs91LM0sC4By9J\nUoUMeEmSKmTAS5JUIQNekqQKGfCSJFXIgJckqUJ+TE6q0NSPqLWyf/NlAxiJpPliwFeomz/ukqS6\neYhekqQKGfCSJFXIQ/Tqu2fDWwSdavT9bUnzzT14SZIqZMBLklShWQd8RKyIiC9ExIMR8UBEvLm0\nnxERuyJib/m9tLRHRHwgIvZFxH0RcV6/ipAkSc/Uyx78cWBjZv4qcAFwfUScC2wC7s7MVcDd5T7A\nJcCq8rMB+FAPy5YkSdOY9Ul2mXkQOFhu/yAiHgSWAZcDY6XbNmAcuKG035qZCdwTEUsi4qwyH0mS\nZmzyhNeNq4+zvsXJr8/mE1778h58RKwEfhO4FxiZDO3y+4Wl2zLg0aaHTZQ2SZLUZ9HYoe5hBhGL\ngf8DvDczPxkRT2Tmkqbpj2fm0oi4C/jPmfml0n438PbM3D1lfhtoHMJnZGTkJdu3b+9pfFPtOXC0\nr/MbORUOPd19/9XLTp92er/HNxOTtQzzGKHz+A4fOdpxnXSaRyednoN+zb/d66sf66jXMXYydQz9\n3lYGod3z2O22MuwGsa3MtV63lWG0Zs2a3Zk52ut8evocfEQ8B/gEcFtmfrI0H5o89B4RZwGHS/sE\nsKLp4cuBx6bOMzO3AFsARkdHc2xsrJchnqDVIZxebFx9nJv3dP807r9mbNrp/R7fTEzWMsxjhM7P\n4Qdvu7PjOuk0j046PQf9mn+711c/1lGvY+xk6hj6va0MQrvnsdttZdgNYluZa71uKzXr5Sz6AG4B\nHszMP22atANYV26vA+5sar+2nE1/AXDU998lSZobvezBvxx4DbAnIr5e2v4I2Ax8LCKuAx4BrizT\ndgKXAvuAp4DX9rBsSZI0jV7Oov8SEG0mr23RP4HrZ7s8SZLUPa9kJ0lShQx4SZIqZMBLklQhA16S\npAoZ8JIkVciAlySpQj1dyU5aqFZ2uhLds/gLKiTVwYCXWuj0D4AkDTsP0UuSVCEDXpKkChnwkiRV\nyICXJKlCBrwkSRUy4CVJqpABL0lShQx4SZIqZMBLklQhA16SpAoZ8JIkVciAlySpQga8JEkV8tvk\npHngt9XNPZ/j4eBXM88fA15SS/5hlhY2D9FLklQh9+AlLUg1HIL3KInmknvwkiRVyD14SVJbHmVY\nuNyDlySpQu7BS5qVGt4Dl2o+QjHwgI+Ii4H3AycBH8nMzYMegyTp2eHZ/I/oQAM+Ik4C/hz4XWAC\n+EpE7MjMbw5yHJK0ENS8d6m5N+g9+POBfZn5MEBEbAcuBwx4aQaezXsl0iB1s60N6z9agw74ZcCj\nTfcngJcOeAySNBD+I6b5FJk5uIVFXAlclJn/odx/DXB+Zr6pqc8GYEO5ew7w7YENcHbOBL4334Po\nk1pqqaUOsJZhVUsttdQBddVyTmae1utMBr0HPwGsaLq/HHisuUNmbgG2DHJQvYiIr2bm6HyPox9q\nqaWWOsBahlUttdRSB9RXSz/mM+jPwX8FWBURZ0fEycBVwI4Bj0GSpOoNdA8+M49HxBuBz9L4mNzW\nzHxgkGOQJOnZYOCfg8/MncDOQS93Di2YtxO6UEsttdQB1jKsaqmlljrAWk4w0JPsJEnSYHgtekmS\nKmTAdyEizomIrzf9PBkRb5nSZywijjb1edd8jXeqiNgaEYcj4v6mtjMiYldE7C2/l7Z57LrSZ29E\nrBvcqFuOpVUdfxIR34qI+yLiUxGxpM1j90fEnrJu+nKGai/a1PLuiDjQ9Bq6tM1jL46Ib0fEvojY\nNLhRt9amljua6tgfEV9v89ihWS8RsSIivhARD0bEAxHx5tK+ELeVdrUsuO1lmloW1PYyTR1zt61k\npj8z+KFxcuB3gF+c0j4GfHq+x9dmzK8AzgPub2r7L8CmcnsTcFOLx50BPFx+Ly23lw5ZHRcCi8rt\nm1rVUabtB86c73XRoZZ3A2/r8LiTgIeAXwJOBr4BnDtstUyZfjPwrmFfL8BZwHnl9mnA3wLnLtBt\npV0tC257maaWBbW9tKtjSp++bivuwc/cWuChzPz7+R5ItzLzi8CRKc2XA9vK7W3AFS0eehGwKzOP\nZObjwC7g4jkbaAet6sjMz2Xm8XL3HhrXVhh6bdZJN/7xcs+Z+WNg8nLP82a6WiIigH8L3D7QQc1C\nZh7MzK+V2z8AHqRx9c2FuK20rGUhbi/TrJduDM320qmOudhWDPiZu4r2K+BlEfGNiPhMRPzaIAc1\nCyOZeRAaLzzghS36tLq0cLcb1nx4HfCZNtMS+FxE7I7G1RKH1RvL4dOtbQ4FL7R18i+BQ5m5t830\noVwvEbES+E3gXhb4tjKllmYLbntpUcuC3F7arJO+bysG/AxE4+I8rwb+usXkr9E4bP9i4IPA/xrk\n2OZItGgbyo9dRMQ7gePAbW26vDwzzwMuAa6PiFcMbHDd+xDwy8BvAAdpHK6basGsk+Jqpt8jGbr1\nEhGLgU8Ab8nMJ7t9WIu2eV8v7WpZiNtLi1oW5PYyzeur79uKAT8zlwBfy8xDUydk5pOZeazc3gk8\nJyLOHPQAZ+BQRJwFUH4fbtGn46WFh0E5oelVwDVZ3qyaKjMfK78PA5+icehuqGTmocz8SWb+FPgw\nrce4INYJQEQsAv4NcEe7PsO2XiLiOTT++N6WmZ8szQtyW2lTy4LcXlrVshC3l2nWyZxsKwb8zLT9\nDysi/ml5D4WIOJ/Gc/v9AY5tpnYAk2f6rgPubNHns8CFEbG0HP66sLQNjYi4GLgBeHVmPtWmz/Mi\n4rTJ2zTquL9V3/k0GSLFv6b1GBfS5Z5/B/hWZk60mjhs66Vsv7cAD2bmnzZNWnDbSrtaFuL2Mk0t\nC2p7meb1BXO1rczH2YQL8Qd4Lo3APr2p7fXA68vtNwIP0DhL8x7gt+Z7zE3jvJ3GIaz/R+M/2uuA\nFwB3A3vL7zNK31HgI02PfR2wr/y8dgjr2EfjPbavl5+/KH1/AdhZbv9SWS/fKOvonUO6Tv4S2APc\nR+OP0FlTayn3L6VxBu5Dw1pLaf/o5PbR1Hdo1wvw2zQO397X9Hq6dIFuK+1qWXDbyzS1LKjtpV0d\nZdqcbCteyU6SpAp5iF6SpAoZ8JIkVciAlySpQga8JEkVMuAlSaqQAS9JUoUMeEmSKmTAS5JUof8P\nLIwC2ZVVELMAAAAASUVORK5CYII=\n",
 923 |       "text/plain": [
 924 |        "<matplotlib.figure.Figure at 0x20c65cedc88>"
 925 |       ]
 926 |      },
 927 |      "metadata": {},
 928 |      "output_type": "display_data"
 929 |     }
 930 |    ],
 931 |    "source": [
 932 |     "%matplotlib inline\n",
 933 |     "import matplotlib.pyplot as plt\n",
 934 |     "jobs_data_copy.hist(bins=50, figsize=(8,5))\n",
 935 |     "# save_fig(\"attribute_histogram_plots\")\n",
 936 |     "plt.show()"
 937 |    ]
 938 |   },
 939 |   {
 940 |    "cell_type": "markdown",
 941 |    "metadata": {},
 942 |    "source": [
 943 |     "## 筛选岗位数量前150的城市"
 944 |    ]
 945 |   },
 946 |   {
 947 |    "cell_type": "code",
 948 |    "execution_count": 287,
 949 |    "metadata": {},
 950 |    "outputs": [],
 951 |    "source": [
 952 |     "cities = jobs_data_copy['城市'].value_counts()[:150].index"
 953 |    ]
 954 |   },
 955 |   {
 956 |    "cell_type": "code",
 957 |    "execution_count": 288,
 958 |    "metadata": {},
 959 |    "outputs": [],
 960 |    "source": [
 961 |     "temp_list = []\n",
 962 |     "for item in jobs_data_copy['城市'].values:\n",
 963 |     "    temp_list.append(item in cities)"
 964 |    ]
 965 |   },
 966 |   {
 967 |    "cell_type": "code",
 968 |    "execution_count": 289,
 969 |    "metadata": {},
 970 |    "outputs": [],
 971 |    "source": [
 972 |     "#temp_list"
 973 |    ]
 974 |   },
 975 |   {
 976 |    "cell_type": "code",
 977 |    "execution_count": 290,
 978 |    "metadata": {},
 979 |    "outputs": [],
 980 |    "source": [
 981 |     "#temp_list\n",
 982 |     "jobs_data_copy = jobs_data_copy[temp_list]"
 983 |    ]
 984 |   },
 985 |   {
 986 |    "cell_type": "code",
 987 |    "execution_count": 291,
 988 |    "metadata": {},
 989 |    "outputs": [
 990 |     {
 991 |      "data": {
 992 |       "text/html": [
 993 |        "<div>\n",
 994 |        "<style scoped>\n",
 995 |        "    .dataframe tbody tr th:only-of-type {\n",
 996 |        "        vertical-align: middle;\n",
 997 |        "    }\n",
 998 |        "\n",
 999 |        "    .dataframe tbody tr th {\n",
1000 |        "        vertical-align: top;\n",
1001 |        "    }\n",
1002 |        "\n",
1003 |        "    .dataframe thead th {\n",
1004 |        "        text-align: right;\n",
1005 |        "    }\n",
1006 |        "</style>\n",
1007 |        "<table border=\"1\" class=\"dataframe\">\n",
1008 |        "  <thead>\n",
1009 |        "    <tr style=\"text-align: right;\">\n",
1010 |        "      <th></th>\n",
1011 |        "      <th>城市</th>\n",
1012 |        "      <th>职位名称</th>\n",
1013 |        "      <th>工作经验</th>\n",
1014 |        "      <th>公司规模</th>\n",
1015 |        "      <th>学历要求</th>\n",
1016 |        "      <th>salary</th>\n",
1017 |        "    </tr>\n",
1018 |        "  </thead>\n",
1019 |        "  <tbody>\n",
1020 |        "    <tr>\n",
1021 |        "      <th>0</th>\n",
1022 |        "      <td>北京</td>\n",
1023 |        "      <td>数据分析</td>\n",
1024 |        "      <td>经验不限</td>\n",
1025 |        "      <td>10000人以上</td>\n",
1026 |        "      <td>本科</td>\n",
1027 |        "      <td>8.0</td>\n",
1028 |        "    </tr>\n",
1029 |        "    <tr>\n",
1030 |        "      <th>2</th>\n",
1031 |        "      <td>北京</td>\n",
1032 |        "      <td>数据分析师</td>\n",
1033 |        "      <td>3-5年</td>\n",
1034 |        "      <td>1000-9999人</td>\n",
1035 |        "      <td>本科</td>\n",
1036 |        "      <td>20.0</td>\n",
1037 |        "    </tr>\n",
1038 |        "    <tr>\n",
1039 |        "      <th>3</th>\n",
1040 |        "      <td>北京</td>\n",
1041 |        "      <td>数据分析</td>\n",
1042 |        "      <td>经验不限</td>\n",
1043 |        "      <td>10000人以上</td>\n",
1044 |        "      <td>本科</td>\n",
1045 |        "      <td>9.0</td>\n",
1046 |        "    </tr>\n",
1047 |        "    <tr>\n",
1048 |        "      <th>4</th>\n",
1049 |        "      <td>北京</td>\n",
1050 |        "      <td>数据分析师</td>\n",
1051 |        "      <td>3-5年</td>\n",
1052 |        "      <td>10000人以上</td>\n",
1053 |        "      <td>本科</td>\n",
1054 |        "      <td>12.5</td>\n",
1055 |        "    </tr>\n",
1056 |        "    <tr>\n",
1057 |        "      <th>8</th>\n",
1058 |        "      <td>北京</td>\n",
1059 |        "      <td>数据分析师</td>\n",
1060 |        "      <td>1-3年</td>\n",
1061 |        "      <td>10000人以上</td>\n",
1062 |        "      <td>本科</td>\n",
1063 |        "      <td>25.5</td>\n",
1064 |        "    </tr>\n",
1065 |        "  </tbody>\n",
1066 |        "</table>\n",
1067 |        "</div>"
1068 |       ],
1069 |       "text/plain": [
1070 |        "   城市   职位名称  工作经验        公司规模 学历要求  salary\n",
1071 |        "0  北京   数据分析  经验不限    10000人以上   本科     8.0\n",
1072 |        "2  北京  数据分析师  3-5年  1000-9999人   本科    20.0\n",
1073 |        "3  北京   数据分析  经验不限    10000人以上   本科     9.0\n",
1074 |        "4  北京  数据分析师  3-5年    10000人以上   本科    12.5\n",
1075 |        "8  北京  数据分析师  1-3年    10000人以上   本科    25.5"
1076 |       ]
1077 |      },
1078 |      "execution_count": 291,
1079 |      "metadata": {},
1080 |      "output_type": "execute_result"
1081 |     }
1082 |    ],
1083 |    "source": [
1084 |     "jobs_data_copy.head()"
1085 |    ]
1086 |   },
1087 |   {
1088 |    "cell_type": "code",
1089 |    "execution_count": 292,
1090 |    "metadata": {},
1091 |    "outputs": [],
1092 |    "source": [
1093 |     "#jobs_data_copy['城市'].value_counts()"
1094 |    ]
1095 |   },
1096 |   {
1097 |    "cell_type": "markdown",
1098 |    "metadata": {},
1099 |    "source": [
1100 |     "## 相同属性合并"
1101 |    ]
1102 |   },
1103 |   {
1104 |    "cell_type": "code",
1105 |    "execution_count": 293,
1106 |    "metadata": {},
1107 |    "outputs": [],
1108 |    "source": [
1109 |     "experience_shift = {\n",
1110 |     "    '1-3年': '1-3年',\n",
1111 |     "    '3-5年': '3-5年',\n",
1112 |     "    '5-10年': '5-10年',\n",
1113 |     "    '应届生':'1年以内',\n",
1114 |     "    '1年以下':'1年以内',\n",
1115 |     "    '1年以内': '1年以内',\n",
1116 |     "    '无经验': '1年以内',\n",
1117 |     "    '经验不限': '经验不限',\n",
1118 |     "    '不限': '经验不限'\n",
1119 |     "}\n",
1120 |     "\n",
1121 |     "scale_shift = {\n",
1122 |     "    '100-499人':'100-499人',\n",
1123 |     "    '0-20人':'0-20人',\n",
1124 |     "    '20人以下':'0-20人',\n",
1125 |     "    '20-99人':'20-99人',\n",
1126 |     "    '100-499人':'100-499人',\n",
1127 |     "    '500-999人':'500-999人',\n",
1128 |     "    '1000-9999人':'1000-9999人',\n",
1129 |     "    '10000人以上':'10000人以上',\n",
1130 |     "    '':'100-499人'\n",
1131 |     "}\n",
1132 |     "\n",
1133 |     "degree_shift = {\n",
1134 |     "    '中专/中技': '中专',\n",
1135 |     "    '中技': '中专',\n",
1136 |     "    '中专': '中专',\n",
1137 |     "    '高中': '高中',\n",
1138 |     "    '大专': '大专',\n",
1139 |     "    '本科': '本科',\n",
1140 |     "    '硕士': '硕士',\n",
1141 |     "    '博士': '博士',\n",
1142 |     "    '不限': '学历不限',\n",
1143 |     "    '学历不限': '学历不限'\n",
1144 |     "}"
1145 |    ]
1146 |   },
1147 |   {
1148 |    "cell_type": "code",
1149 |    "execution_count": 294,
1150 |    "metadata": {},
1151 |    "outputs": [],
1152 |    "source": [
1153 |     "jobs_data_copy['工作经验'] = jobs_data_copy['工作经验'].map(experience_shift)\n",
1154 |     "jobs_data_copy['公司规模'] = jobs_data_copy['公司规模'].map(scale_shift)\n",
1155 |     "jobs_data_copy['学历要求'] = jobs_data_copy['学历要求'].map(degree_shift)"
1156 |    ]
1157 |   },
1158 |   {
1159 |    "cell_type": "code",
1160 |    "execution_count": 295,
1161 |    "metadata": {},
1162 |    "outputs": [
1163 |     {
1164 |      "data": {
1165 |       "text/plain": [
1166 |        "1-3年     5219\n",
1167 |        "3-5年     4271\n",
1168 |        "经验不限     1309\n",
1169 |        "5-10年     825\n",
1170 |        "1年以内      825\n",
1171 |        "Name: 工作经验, dtype: int64"
1172 |       ]
1173 |      },
1174 |      "execution_count": 295,
1175 |      "metadata": {},
1176 |      "output_type": "execute_result"
1177 |     }
1178 |    ],
1179 |    "source": [
1180 |     "jobs_data_copy['工作经验'].value_counts()"
1181 |    ]
1182 |   },
1183 |   {
1184 |    "cell_type": "code",
1185 |    "execution_count": 296,
1186 |    "metadata": {},
1187 |    "outputs": [
1188 |     {
1189 |      "data": {
1190 |       "text/plain": [
1191 |        "100-499人      3843\n",
1192 |        "20-99人        2848\n",
1193 |        "1000-9999人    2503\n",
1194 |        "10000人以上      1648\n",
1195 |        "500-999人      1292\n",
1196 |        "0-20人          315\n",
1197 |        "Name: 公司规模, dtype: int64"
1198 |       ]
1199 |      },
1200 |      "execution_count": 296,
1201 |      "metadata": {},
1202 |      "output_type": "execute_result"
1203 |     }
1204 |    ],
1205 |    "source": [
1206 |     "jobs_data_copy['公司规模'].value_counts()"
1207 |    ]
1208 |   },
1209 |   {
1210 |    "cell_type": "code",
1211 |    "execution_count": 297,
1212 |    "metadata": {},
1213 |    "outputs": [
1214 |     {
1215 |      "data": {
1216 |       "text/plain": [
1217 |        "本科      8941\n",
1218 |        "大专      2479\n",
1219 |        "硕士       564\n",
1220 |        "学历不限     420\n",
1221 |        "中专        21\n",
1222 |        "博士        13\n",
1223 |        "高中        11\n",
1224 |        "Name: 学历要求, dtype: int64"
1225 |       ]
1226 |      },
1227 |      "execution_count": 297,
1228 |      "metadata": {},
1229 |      "output_type": "execute_result"
1230 |     }
1231 |    ],
1232 |    "source": [
1233 |     "jobs_data_copy['学历要求'].value_counts()"
1234 |    ]
1235 |   },
1236 |   {
1237 |    "cell_type": "markdown",
1238 |    "metadata": {},
1239 |    "source": [
1240 |     "# 划分训练集与测试集"
1241 |    ]
1242 |   },
1243 |   {
1244 |    "cell_type": "code",
1245 |    "execution_count": 298,
1246 |    "metadata": {
1247 |     "collapsed": true
1248 |    },
1249 |    "outputs": [],
1250 |    "source": [
1251 |     "from sklearn.model_selection import train_test_split\n",
1252 |     "\n",
1253 |     "train_set, test_set = train_test_split(jobs_data_copy, test_size=0.2, random_state=42)"
1254 |    ]
1255 |   },
1256 |   {
1257 |    "cell_type": "code",
1258 |    "execution_count": 299,
1259 |    "metadata": {
1260 |     "collapsed": true
1261 |    },
1262 |    "outputs": [],
1263 |    "source": [
1264 |     "datas_train = train_set.copy()\n",
1265 |     "datas_test = test_set.copy()"
1266 |    ]
1267 |   },
1268 |   {
1269 |    "cell_type": "code",
1270 |    "execution_count": 300,
1271 |    "metadata": {},
1272 |    "outputs": [
1273 |     {
1274 |      "data": {
1275 |       "text/html": [
1276 |        "<div>\n",
1277 |        "<style scoped>\n",
1278 |        "    .dataframe tbody tr th:only-of-type {\n",
1279 |        "        vertical-align: middle;\n",
1280 |        "    }\n",
1281 |        "\n",
1282 |        "    .dataframe tbody tr th {\n",
1283 |        "        vertical-align: top;\n",
1284 |        "    }\n",
1285 |        "\n",
1286 |        "    .dataframe thead th {\n",
1287 |        "        text-align: right;\n",
1288 |        "    }\n",
1289 |        "</style>\n",
1290 |        "<table border=\"1\" class=\"dataframe\">\n",
1291 |        "  <thead>\n",
1292 |        "    <tr style=\"text-align: right;\">\n",
1293 |        "      <th></th>\n",
1294 |        "      <th>城市</th>\n",
1295 |        "      <th>职位名称</th>\n",
1296 |        "      <th>工作经验</th>\n",
1297 |        "      <th>公司规模</th>\n",
1298 |        "      <th>学历要求</th>\n",
1299 |        "      <th>salary</th>\n",
1300 |        "    </tr>\n",
1301 |        "  </thead>\n",
1302 |        "  <tbody>\n",
1303 |        "    <tr>\n",
1304 |        "      <th>21110</th>\n",
1305 |        "      <td>南阳</td>\n",
1306 |        "      <td>数据分析</td>\n",
1307 |        "      <td>1-3年</td>\n",
1308 |        "      <td>20-99人</td>\n",
1309 |        "      <td>本科</td>\n",
1310 |        "      <td>6.5</td>\n",
1311 |        "    </tr>\n",
1312 |        "    <tr>\n",
1313 |        "      <th>56033</th>\n",
1314 |        "      <td>云浮</td>\n",
1315 |        "      <td>化妆品数据分析员</td>\n",
1316 |        "      <td>1-3年</td>\n",
1317 |        "      <td>100-499人</td>\n",
1318 |        "      <td>本科</td>\n",
1319 |        "      <td>8.5</td>\n",
1320 |        "    </tr>\n",
1321 |        "    <tr>\n",
1322 |        "      <th>37705</th>\n",
1323 |        "      <td>泉州</td>\n",
1324 |        "      <td>数据分析专员</td>\n",
1325 |        "      <td>1-3年</td>\n",
1326 |        "      <td>100-499人</td>\n",
1327 |        "      <td>本科</td>\n",
1328 |        "      <td>6.5</td>\n",
1329 |        "    </tr>\n",
1330 |        "    <tr>\n",
1331 |        "      <th>24622</th>\n",
1332 |        "      <td>淮安</td>\n",
1333 |        "      <td>数据分析</td>\n",
1334 |        "      <td>1-3年</td>\n",
1335 |        "      <td>1000-9999人</td>\n",
1336 |        "      <td>大专</td>\n",
1337 |        "      <td>6.5</td>\n",
1338 |        "    </tr>\n",
1339 |        "    <tr>\n",
1340 |        "      <th>37106</th>\n",
1341 |        "      <td>宁德</td>\n",
1342 |        "      <td>数据分析师（福州）</td>\n",
1343 |        "      <td>1-3年</td>\n",
1344 |        "      <td>100-499人</td>\n",
1345 |        "      <td>本科</td>\n",
1346 |        "      <td>11.5</td>\n",
1347 |        "    </tr>\n",
1348 |        "  </tbody>\n",
1349 |        "</table>\n",
1350 |        "</div>"
1351 |       ],
1352 |       "text/plain": [
1353 |        "       城市       职位名称  工作经验        公司规模 学历要求  salary\n",
1354 |        "21110  南阳       数据分析  1-3年      20-99人   本科     6.5\n",
1355 |        "56033  云浮   化妆品数据分析员  1-3年    100-499人   本科     8.5\n",
1356 |        "37705  泉州     数据分析专员  1-3年    100-499人   本科     6.5\n",
1357 |        "24622  淮安       数据分析  1-3年  1000-9999人   大专     6.5\n",
1358 |        "37106  宁德  数据分析师（福州）  1-3年    100-499人   本科    11.5"
1359 |       ]
1360 |      },
1361 |      "execution_count": 300,
1362 |      "metadata": {},
1363 |      "output_type": "execute_result"
1364 |     }
1365 |    ],
1366 |    "source": [
1367 |     "datas_train.head()"
1368 |    ]
1369 |   },
1370 |   {
1371 |    "cell_type": "markdown",
1372 |    "metadata": {},
1373 |    "source": [
1374 |     "## 为 标签/类别 属性编码"
1375 |    ]
1376 |   },
1377 |   {
1378 |    "cell_type": "markdown",
1379 |    "metadata": {},
1380 |    "source": [
1381 |     "### 标签/类别 训练集/测试集划分"
1382 |    ]
1383 |   },
1384 |   {
1385 |    "cell_type": "code",
1386 |    "execution_count": 301,
1387 |    "metadata": {},
1388 |    "outputs": [
1389 |     {
1390 |      "data": {
1391 |       "text/html": [
1392 |        "<div>\n",
1393 |        "<style scoped>\n",
1394 |        "    .dataframe tbody tr th:only-of-type {\n",
1395 |        "        vertical-align: middle;\n",
1396 |        "    }\n",
1397 |        "\n",
1398 |        "    .dataframe tbody tr th {\n",
1399 |        "        vertical-align: top;\n",
1400 |        "    }\n",
1401 |        "\n",
1402 |        "    .dataframe thead th {\n",
1403 |        "        text-align: right;\n",
1404 |        "    }\n",
1405 |        "</style>\n",
1406 |        "<table border=\"1\" class=\"dataframe\">\n",
1407 |        "  <thead>\n",
1408 |        "    <tr style=\"text-align: right;\">\n",
1409 |        "      <th></th>\n",
1410 |        "      <th>城市</th>\n",
1411 |        "      <th>公司规模</th>\n",
1412 |        "      <th>学历要求</th>\n",
1413 |        "      <th>工作经验</th>\n",
1414 |        "    </tr>\n",
1415 |        "  </thead>\n",
1416 |        "  <tbody>\n",
1417 |        "    <tr>\n",
1418 |        "      <th>21110</th>\n",
1419 |        "      <td>南阳</td>\n",
1420 |        "      <td>20-99人</td>\n",
1421 |        "      <td>本科</td>\n",
1422 |        "      <td>1-3年</td>\n",
1423 |        "    </tr>\n",
1424 |        "    <tr>\n",
1425 |        "      <th>56033</th>\n",
1426 |        "      <td>云浮</td>\n",
1427 |        "      <td>100-499人</td>\n",
1428 |        "      <td>本科</td>\n",
1429 |        "      <td>1-3年</td>\n",
1430 |        "    </tr>\n",
1431 |        "    <tr>\n",
1432 |        "      <th>37705</th>\n",
1433 |        "      <td>泉州</td>\n",
1434 |        "      <td>100-499人</td>\n",
1435 |        "      <td>本科</td>\n",
1436 |        "      <td>1-3年</td>\n",
1437 |        "    </tr>\n",
1438 |        "    <tr>\n",
1439 |        "      <th>24622</th>\n",
1440 |        "      <td>淮安</td>\n",
1441 |        "      <td>1000-9999人</td>\n",
1442 |        "      <td>大专</td>\n",
1443 |        "      <td>1-3年</td>\n",
1444 |        "    </tr>\n",
1445 |        "    <tr>\n",
1446 |        "      <th>37106</th>\n",
1447 |        "      <td>宁德</td>\n",
1448 |        "      <td>100-499人</td>\n",
1449 |        "      <td>本科</td>\n",
1450 |        "      <td>1-3年</td>\n",
1451 |        "    </tr>\n",
1452 |        "  </tbody>\n",
1453 |        "</table>\n",
1454 |        "</div>"
1455 |       ],
1456 |       "text/plain": [
1457 |        "       城市        公司规模 学历要求  工作经验\n",
1458 |        "21110  南阳      20-99人   本科  1-3年\n",
1459 |        "56033  云浮    100-499人   本科  1-3年\n",
1460 |        "37705  泉州    100-499人   本科  1-3年\n",
1461 |        "24622  淮安  1000-9999人   大专  1-3年\n",
1462 |        "37106  宁德    100-499人   本科  1-3年"
1463 |       ]
1464 |      },
1465 |      "execution_count": 301,
1466 |      "metadata": {},
1467 |      "output_type": "execute_result"
1468 |     }
1469 |    ],
1470 |    "source": [
1471 |     "cata_train = datas_train[['城市','公司规模','学历要求','工作经验']] # 训练集\n",
1472 |     "cata_test = datas_test[['城市','公司规模','学历要求','工作经验']] # 测试集\n",
1473 |     "cata_train.head()"
1474 |    ]
1475 |   },
1476 |   {
1477 |    "cell_type": "code",
1478 |    "execution_count": 302,
1479 |    "metadata": {
1480 |     "collapsed": true
1481 |    },
1482 |    "outputs": [],
1483 |    "source": [
1484 |     "from sklearn.preprocessing import OneHotEncoder\n",
1485 |     "\n",
1486 |     "cat_encoder = OneHotEncoder(sparse=False)"
1487 |    ]
1488 |   },
1489 |   {
1490 |    "cell_type": "code",
1491 |    "execution_count": 303,
1492 |    "metadata": {
1493 |     "collapsed": true
1494 |    },
1495 |    "outputs": [],
1496 |    "source": [
1497 |     "job_cata_train = cat_encoder.fit_transform(cata_train)\n",
1498 |     "job_cata_test = cat_encoder.fit_transform(cata_test)"
1499 |    ]
1500 |   },
1501 |   {
1502 |    "cell_type": "code",
1503 |    "execution_count": 304,
1504 |    "metadata": {},
1505 |    "outputs": [
1506 |     {
1507 |      "data": {
1508 |       "text/plain": [
1509 |        "[array(['三明', '上海', '东沙群岛', '东莞', '东营', '中山', '临沂', '丽水', '乐山', '云浮', '亳州',\n",
1510 |        "        '仙桃', '佛山', '六安', '内江', '凉山', '北京', '十堰', '南京', '南充', '南平', '南通',\n",
1511 |        "        '南阳', '厦门', '台州', '合肥', '周口', '咸宁', '咸阳', '商丘', '商洛', '嘉兴', '大连',\n",
1512 |        "        '天津', '天门', '威海', '娄底', '孝感', '宁德', '安庆', '安康', '安阳', '宜宾', '宜昌',\n",
1513 |        "        '宣城', '宿州', '宿迁', '岳阳', '巴中', '常州', '常德', '平顶山', '广元', '广安', '广州',\n",
1514 |        "        '开封', '张家界', '徐州', '德阳', '怀化', '恩施', '惠州', '成都', '扬州', '揭阳', '攀枝花',\n",
1515 |        "        '新乡', '日照', '杭州', '枣庄', '株洲', '梅州', '武汉', '永州', '汕头', '汕尾', '江门',\n",
1516 |        "        '池州', '河源', '泉州', '泰安', '泰州', '泸州', '洛阳', '济南', '济宁', '淄博', '淮北',\n",
1517 |        "        '淮安', '深圳', '清远', '温州', '湖州', '湘潭', '湘西', '湛江', '滨州', '漯河', '漳州',\n",
1518 |        "        '潍坊', '潜江', '潮州', '烟台', '珠海', '甘孜', '益阳', '盐城', '眉山', '神农架', '福州',\n",
1519 |        "        '绍兴', '绵阳', '聊城', '肇庆', '自贡', '舟山', '芜湖', '苏州', '茂名', '荆州', '荆门',\n",
1520 |        "        '莆田', '莱芜', '菏泽', '衡阳', '衢州', '襄阳', '西安', '资阳', '达州', '连云港', '遂宁',\n",
1521 |        "        '邵阳', '郑州', '郴州', '鄂州', '金华', '铜川', '镇江', '长沙', '阜阳', '阳江', '阿坝',\n",
1522 |        "        '随州', '雅安', '韶关', '黄冈', '黄山', '黄石', '龙岩'], dtype=object),\n",
1523 |        " array(['0-20人', '100-499人', '1000-9999人', '10000人以上', '20-99人',\n",
1524 |        "        '500-999人'], dtype=object),\n",
1525 |        " array(['中专', '博士', '大专', '学历不限', '本科', '硕士', '高中'], dtype=object),\n",
1526 |        " array(['1-3年', '1年以内', '3-5年', '5-10年', '经验不限'], dtype=object)]"
1527 |       ]
1528 |      },
1529 |      "execution_count": 304,
1530 |      "metadata": {},
1531 |      "output_type": "execute_result"
1532 |     }
1533 |    ],
1534 |    "source": [
1535 |     "cat_encoder.categories_"
1536 |    ]
1537 |   },
1538 |   {
1539 |    "cell_type": "markdown",
1540 |    "metadata": {},
1541 |    "source": [
1542 |     "### 参数构造函数"
1543 |    ]
1544 |   },
1545 |   {
1546 |    "cell_type": "code",
1547 |    "execution_count": 378,
1548 |    "metadata": {},
1549 |    "outputs": [
1550 |     {
1551 |      "data": {
1552 |       "text/plain": [
1553 |        "array(['三明', '上海', '东沙群岛', '东莞', '东营', '中山', '临沂', '丽水', '乐山', '云浮', '亳州',\n",
1554 |        "       '仙桃', '佛山', '六安', '内江', '凉山', '北京', '十堰', '南京', '南充', '南平', '南通',\n",
1555 |        "       '南阳', '厦门', '台州', '合肥', '周口', '咸宁', '咸阳', '商丘', '商洛', '嘉兴', '大连',\n",
1556 |        "       '天津', '天门', '威海', '娄底', '孝感', '宁德', '安庆', '安康', '安阳', '宜宾', '宜昌',\n",
1557 |        "       '宣城', '宿州', '宿迁', '岳阳', '巴中', '常州', '常德', '平顶山', '广元', '广安', '广州',\n",
1558 |        "       '开封', '张家界', '徐州', '德阳', '怀化', '恩施', '惠州', '成都', '扬州', '揭阳', '攀枝花',\n",
1559 |        "       '新乡', '日照', '杭州', '枣庄', '株洲', '梅州', '武汉', '永州', '汕头', '汕尾', '江门',\n",
1560 |        "       '池州', '河源', '泉州', '泰安', '泰州', '泸州', '洛阳', '济南', '济宁', '淄博', '淮北',\n",
1561 |        "       '淮安', '深圳', '清远', '温州', '湖州', '湘潭', '湘西', '湛江', '滨州', '漯河', '漳州',\n",
1562 |        "       '潍坊', '潜江', '潮州', '烟台', '珠海', '甘孜', '益阳', '盐城', '眉山', '神农架', '福州',\n",
1563 |        "       '绍兴', '绵阳', '聊城', '肇庆', '自贡', '舟山', '芜湖', '苏州', '茂名', '荆州', '荆门',\n",
1564 |        "       '莆田', '莱芜', '菏泽', '衡阳', '衢州', '襄阳', '西安', '资阳', '达州', '连云港', '遂宁',\n",
1565 |        "       '邵阳', '郑州', '郴州', '鄂州', '金华', '铜川', '镇江', '长沙', '阜阳', '阳江', '阿坝',\n",
1566 |        "       '随州', '雅安', '韶关', '黄冈', '黄山', '黄石', '龙岩'], dtype=object)"
1567 |       ]
1568 |      },
1569 |      "execution_count": 378,
1570 |      "metadata": {},
1571 |      "output_type": "execute_result"
1572 |     }
1573 |    ],
1574 |    "source": [
1575 |     "cat_encoder.categories_[0] "
1576 |    ]
1577 |   },
1578 |   {
1579 |    "cell_type": "code",
1580 |    "execution_count": 444,
1581 |    "metadata": {},
1582 |    "outputs": [],
1583 |    "source": [
1584 |     "import numpy as np\n",
1585 |     "\n",
1586 |     "def func_params(templist):\n",
1587 |     "    temp = []\n",
1588 |     "    city,scale,degree,exp = templist\n",
1589 |     "    citypara = cat_encoder.categories_[0] == '{city}'.format(city=city)\n",
1590 |     "    scalepara = cat_encoder.categories_[1] == '{scale}'.format(scale=scale)\n",
1591 |     "    degreepara = cat_encoder.categories_[2] == '{degree}'.format(degree=degree)\n",
1592 |     "    exppara = cat_encoder.categories_[3] == '{exp}'.format(exp=exp)\n",
1593 |     "    # citypara\n",
1594 |     "    for item in citypara:\n",
1595 |     "        #print(item)\n",
1596 |     "        if item == False:\n",
1597 |     "            item = float(0)\n",
1598 |     "            temp.append(item)\n",
1599 |     "        else:\n",
1600 |     "            item = float(1)\n",
1601 |     "            temp.append(item)\n",
1602 |     "    for item in scalepara:\n",
1603 |     "        if item == False:\n",
1604 |     "            item = float(0)\n",
1605 |     "            temp.append(item)\n",
1606 |     "        else:\n",
1607 |     "            item = float(1)\n",
1608 |     "            temp.append(item)\n",
1609 |     "    for item in scalepara:\n",
1610 |     "        if item == False:\n",
1611 |     "            item = float(0)\n",
1612 |     "            temp.append(item)\n",
1613 |     "        else:\n",
1614 |     "            item = float(1)\n",
1615 |     "            temp.append(item)\n",
1616 |     "    for item in scalepara:\n",
1617 |     "        if item == False:\n",
1618 |     "            item = float(0)\n",
1619 |     "            temp.append(item)\n",
1620 |     "        else:\n",
1621 |     "            item = float(1)\n",
1622 |     "            temp.append(item)\n",
1623 |     "    temp = np.array(temp, dtype = float).reshape(1, -1)\n",
1624 |     "    return temp"
1625 |    ]
1626 |   },
1627 |   {
1628 |    "cell_type": "code",
1629 |    "execution_count": 445,
1630 |    "metadata": {},
1631 |    "outputs": [],
1632 |    "source": [
1633 |     "paramlist = func_params(['上海','1000-9999人','硕士','1-3年'])\n",
1634 |     "#paramlist"
1635 |    ]
1636 |   },
1637 |   {
1638 |    "cell_type": "code",
1639 |    "execution_count": 372,
1640 |    "metadata": {},
1641 |    "outputs": [
1642 |     {
1643 |      "data": {
1644 |       "text/plain": [
1645 |        "array(['0-20人', '100-499人', '1000-9999人', '10000人以上', '20-99人',\n",
1646 |        "       '500-999人'], dtype=object)"
1647 |       ]
1648 |      },
1649 |      "execution_count": 372,
1650 |      "metadata": {},
1651 |      "output_type": "execute_result"
1652 |     }
1653 |    ],
1654 |    "source": [
1655 |     "cat_encoder.categories_[1]"
1656 |    ]
1657 |   },
1658 |   {
1659 |    "cell_type": "code",
1660 |    "execution_count": 432,
1661 |    "metadata": {},
1662 |    "outputs": [],
1663 |    "source": [
1664 |     "#cat_encoder.categories_[3]"
1665 |    ]
1666 |   },
1667 |   {
1668 |    "cell_type": "markdown",
1669 |    "metadata": {},
1670 |    "source": [
1671 |     "# 用于机器学习的数据"
1672 |    ]
1673 |   },
1674 |   {
1675 |    "cell_type": "markdown",
1676 |    "metadata": {},
1677 |    "source": [
1678 |     "## x_train,y_train"
1679 |    ]
1680 |   },
1681 |   {
1682 |    "cell_type": "code",
1683 |    "execution_count": 305,
1684 |    "metadata": {},
1685 |    "outputs": [],
1686 |    "source": [
1687 |     "# 训练集\n",
1688 |     "x_train = job_cata_train\n",
1689 |     "y_train = datas_train['salary'].values.reshape(-1, 1)\n",
1690 |     "#y_train"
1691 |    ]
1692 |   },
1693 |   {
1694 |    "cell_type": "code",
1695 |    "execution_count": 439,
1696 |    "metadata": {},
1697 |    "outputs": [],
1698 |    "source": [
1699 |     "#x_train[0]"
1700 |    ]
1701 |   },
1702 |   {
1703 |    "cell_type": "code",
1704 |    "execution_count": 306,
1705 |    "metadata": {},
1706 |    "outputs": [
1707 |     {
1708 |      "data": {
1709 |       "text/plain": [
1710 |        "(9959, 168)"
1711 |       ]
1712 |      },
1713 |      "execution_count": 306,
1714 |      "metadata": {},
1715 |      "output_type": "execute_result"
1716 |     }
1717 |    ],
1718 |    "source": [
1719 |     "x_train.shape"
1720 |    ]
1721 |   },
1722 |   {
1723 |    "cell_type": "code",
1724 |    "execution_count": 307,
1725 |    "metadata": {},
1726 |    "outputs": [
1727 |     {
1728 |      "data": {
1729 |       "text/plain": [
1730 |        "(9959, 1)"
1731 |       ]
1732 |      },
1733 |      "execution_count": 307,
1734 |      "metadata": {},
1735 |      "output_type": "execute_result"
1736 |     }
1737 |    ],
1738 |    "source": [
1739 |     "y_train.shape"
1740 |    ]
1741 |   },
1742 |   {
1743 |    "cell_type": "markdown",
1744 |    "metadata": {},
1745 |    "source": [
1746 |     "## x_test,y_test"
1747 |    ]
1748 |   },
1749 |   {
1750 |    "cell_type": "code",
1751 |    "execution_count": 308,
1752 |    "metadata": {
1753 |     "collapsed": true
1754 |    },
1755 |    "outputs": [],
1756 |    "source": [
1757 |     "# 测试集\n",
1758 |     "x_test = job_cata_test\n",
1759 |     "y_test = datas_test['salary'].values.reshape(-1, 1)"
1760 |    ]
1761 |   },
1762 |   {
1763 |    "cell_type": "code",
1764 |    "execution_count": 309,
1765 |    "metadata": {},
1766 |    "outputs": [
1767 |     {
1768 |      "data": {
1769 |       "text/plain": [
1770 |        "(2490, 168)"
1771 |       ]
1772 |      },
1773 |      "execution_count": 309,
1774 |      "metadata": {},
1775 |      "output_type": "execute_result"
1776 |     }
1777 |    ],
1778 |    "source": [
1779 |     "x_test.shape"
1780 |    ]
1781 |   },
1782 |   {
1783 |    "cell_type": "code",
1784 |    "execution_count": 310,
1785 |    "metadata": {},
1786 |    "outputs": [
1787 |     {
1788 |      "data": {
1789 |       "text/plain": [
1790 |        "(2490, 1)"
1791 |       ]
1792 |      },
1793 |      "execution_count": 310,
1794 |      "metadata": {},
1795 |      "output_type": "execute_result"
1796 |     }
1797 |    ],
1798 |    "source": [
1799 |     "y_test.shape"
1800 |    ]
1801 |   },
1802 |   {
1803 |    "cell_type": "markdown",
1804 |    "metadata": {},
1805 |    "source": [
1806 |     "# 机器学习建模"
1807 |    ]
1808 |   },
1809 |   {
1810 |    "cell_type": "markdown",
1811 |    "metadata": {},
1812 |    "source": [
1813 |     "## 决策树"
1814 |    ]
1815 |   },
1816 |   {
1817 |    "cell_type": "code",
1818 |    "execution_count": 311,
1819 |    "metadata": {
1820 |     "scrolled": true
1821 |    },
1822 |    "outputs": [
1823 |     {
1824 |      "data": {
1825 |       "text/plain": [
1826 |        "DecisionTreeRegressor(criterion='mse', max_depth=None, max_features=None,\n",
1827 |        "           max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
1828 |        "           min_impurity_split=None, min_samples_leaf=1,\n",
1829 |        "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
1830 |        "           presort=False, random_state=42, splitter='best')"
1831 |       ]
1832 |      },
1833 |      "execution_count": 311,
1834 |      "metadata": {},
1835 |      "output_type": "execute_result"
1836 |     }
1837 |    ],
1838 |    "source": [
1839 |     "from sklearn.tree import DecisionTreeRegressor\n",
1840 |     "\n",
1841 |     "tree_reg = DecisionTreeRegressor(random_state=42)\n",
1842 |     "tree_reg.fit(x_train, y_train)"
1843 |    ]
1844 |   },
1845 |   {
1846 |    "cell_type": "markdown",
1847 |    "metadata": {},
1848 |    "source": [
1849 |     "### 检验"
1850 |    ]
1851 |   },
1852 |   {
1853 |    "cell_type": "markdown",
1854 |    "metadata": {},
1855 |    "source": [
1856 |     "#### 训练集检验"
1857 |    ]
1858 |   },
1859 |   {
1860 |    "cell_type": "code",
1861 |    "execution_count": 312,
1862 |    "metadata": {
1863 |     "scrolled": true
1864 |    },
1865 |    "outputs": [
1866 |     {
1867 |      "data": {
1868 |       "text/plain": [
1869 |        "array([[ 7.66666667],\n",
1870 |        "       [11.33333333],\n",
1871 |        "       [10.        ],\n",
1872 |        "       [ 6.5       ],\n",
1873 |        "       [ 9.875     ],\n",
1874 |        "       [20.        ],\n",
1875 |        "       [15.8125    ],\n",
1876 |        "       [ 8.38461538],\n",
1877 |        "       [ 7.8       ],\n",
1878 |        "       [10.25      ]])"
1879 |       ]
1880 |      },
1881 |      "execution_count": 312,
1882 |      "metadata": {},
1883 |      "output_type": "execute_result"
1884 |     }
1885 |    ],
1886 |    "source": [
1887 |     "y_pred_tree = tree_reg.predict(x_train)\n",
1888 |     "y_pred_tree[:10].reshape(10, 1)"
1889 |    ]
1890 |   },
1891 |   {
1892 |    "cell_type": "code",
1893 |    "execution_count": 354,
1894 |    "metadata": {},
1895 |    "outputs": [],
1896 |    "source": [
1897 |     "#datas_train.head(10)"
1898 |    ]
1899 |   },
1900 |   {
1901 |    "cell_type": "code",
1902 |    "execution_count": 314,
1903 |    "metadata": {
1904 |     "scrolled": true
1905 |    },
1906 |    "outputs": [
1907 |     {
1908 |      "data": {
1909 |       "text/plain": [
1910 |        "array([[ 6.5],\n",
1911 |        "       [ 8.5],\n",
1912 |        "       [ 6.5],\n",
1913 |        "       [ 6.5],\n",
1914 |        "       [11.5],\n",
1915 |        "       [20. ],\n",
1916 |        "       [16.5],\n",
1917 |        "       [ 7. ],\n",
1918 |        "       [ 6.5],\n",
1919 |        "       [ 8. ]])"
1920 |       ]
1921 |      },
1922 |      "execution_count": 314,
1923 |      "metadata": {},
1924 |      "output_type": "execute_result"
1925 |     }
1926 |    ],
1927 |    "source": [
1928 |     "y_train[:10]"
1929 |    ]
1930 |   },
1931 |   {
1932 |    "cell_type": "code",
1933 |    "execution_count": 315,
1934 |    "metadata": {},
1935 |    "outputs": [
1936 |     {
1937 |      "data": {
1938 |       "text/plain": [
1939 |        "3.0382473722545784"
1940 |       ]
1941 |      },
1942 |      "execution_count": 315,
1943 |      "metadata": {},
1944 |      "output_type": "execute_result"
1945 |     }
1946 |    ],
1947 |    "source": [
1948 |     "from sklearn.metrics import mean_squared_error\n",
1949 |     "import numpy as np\n",
1950 |     "\n",
1951 |     "tree_mse = mean_squared_error(y_train, y_pred_tree)\n",
1952 |     "tree_rmse = np.sqrt(tree_mse)\n",
1953 |     "tree_rmse"
1954 |    ]
1955 |   },
1956 |   {
1957 |    "cell_type": "markdown",
1958 |    "metadata": {},
1959 |    "source": [
1960 |     "训练集预测误差：3.0382473722545784"
1961 |    ]
1962 |   },
1963 |   {
1964 |    "cell_type": "markdown",
1965 |    "metadata": {},
1966 |    "source": [
1967 |     "#### 测试集检验"
1968 |    ]
1969 |   },
1970 |   {
1971 |    "cell_type": "code",
1972 |    "execution_count": 316,
1973 |    "metadata": {},
1974 |    "outputs": [
1975 |     {
1976 |      "data": {
1977 |       "text/plain": [
1978 |        "4.782476424480568"
1979 |       ]
1980 |      },
1981 |      "execution_count": 316,
1982 |      "metadata": {},
1983 |      "output_type": "execute_result"
1984 |     }
1985 |    ],
1986 |    "source": [
1987 |     "from sklearn.metrics import mean_squared_error\n",
1988 |     "import numpy as np\n",
1989 |     "\n",
1990 |     "y_pred_tree_test = tree_reg.predict(x_test)\n",
1991 |     "\n",
1992 |     "tree_mse = mean_squared_error(y_test, y_pred_tree_test)\n",
1993 |     "tree_rmse = np.sqrt(tree_mse)\n",
1994 |     "tree_rmse"
1995 |    ]
1996 |   },
1997 |   {
1998 |    "cell_type": "markdown",
1999 |    "metadata": {},
2000 |    "source": [
2001 |     "测试集误差：4.782476424480568"
2002 |    ]
2003 |   },
2004 |   {
2005 |    "cell_type": "markdown",
2006 |    "metadata": {},
2007 |    "source": [
2008 |     "### 交叉验证"
2009 |    ]
2010 |   },
2011 |   {
2012 |    "cell_type": "code",
2013 |    "execution_count": 317,
2014 |    "metadata": {
2015 |     "collapsed": true
2016 |    },
2017 |    "outputs": [],
2018 |    "source": [
2019 |     "from sklearn.model_selection import cross_val_score\n",
2020 |     "\n",
2021 |     "scores = cross_val_score(tree_reg, x_train, y_train,\n",
2022 |     "                         scoring=\"neg_mean_squared_error\", cv=10)\n",
2023 |     "tree_rmse_scores = np.sqrt(-scores)"
2024 |    ]
2025 |   },
2026 |   {
2027 |    "cell_type": "code",
2028 |    "execution_count": 318,
2029 |    "metadata": {},
2030 |    "outputs": [
2031 |     {
2032 |      "name": "stdout",
2033 |      "output_type": "stream",
2034 |      "text": [
2035 |       "Scores: [4.62433783 4.74066502 4.8478161  4.60423202 4.78729759 4.55902835\n",
2036 |       " 4.47123285 4.46471566 4.62152351 4.81591793]\n",
2037 |       "Mean: 4.653676687092153\n",
2038 |       "Standard deviation: 0.1310433320106498\n"
2039 |      ]
2040 |     }
2041 |    ],
2042 |    "source": [
2043 |     "def display_scores(scores):\n",
2044 |     "    print(\"Scores:\", scores)\n",
2045 |     "    print(\"Mean:\", scores.mean())\n",
2046 |     "    print(\"Standard deviation:\", scores.std())\n",
2047 |     "\n",
2048 |     "display_scores(tree_rmse_scores)"
2049 |    ]
2050 |   },
2051 |   {
2052 |    "cell_type": "markdown",
2053 |    "metadata": {},
2054 |    "source": [
2055 |     "## 随机森林"
2056 |    ]
2057 |   },
2058 |   {
2059 |    "cell_type": "code",
2060 |    "execution_count": 319,
2061 |    "metadata": {},
2062 |    "outputs": [
2063 |     {
2064 |      "name": "stderr",
2065 |      "output_type": "stream",
2066 |      "text": [
2067 |       "C:\\Users\\13626\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
2068 |       "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
2069 |       "C:\\Users\\13626\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
2070 |       "  after removing the cwd from sys.path.\n"
2071 |      ]
2072 |     },
2073 |     {
2074 |      "data": {
2075 |       "text/plain": [
2076 |        "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
2077 |        "           max_features='auto', max_leaf_nodes=None,\n",
2078 |        "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
2079 |        "           min_samples_leaf=1, min_samples_split=2,\n",
2080 |        "           min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=None,\n",
2081 |        "           oob_score=False, random_state=42, verbose=0, warm_start=False)"
2082 |       ]
2083 |      },
2084 |      "execution_count": 319,
2085 |      "metadata": {},
2086 |      "output_type": "execute_result"
2087 |     }
2088 |    ],
2089 |    "source": [
2090 |     "from sklearn.ensemble import RandomForestRegressor\n",
2091 |     "\n",
2092 |     "forest_reg = RandomForestRegressor(random_state=42)\n",
2093 |     "forest_reg.fit(x_train, y_train)"
2094 |    ]
2095 |   },
2096 |   {
2097 |    "cell_type": "markdown",
2098 |    "metadata": {},
2099 |    "source": [
2100 |     "### 训练集检验"
2101 |    ]
2102 |   },
2103 |   {
2104 |    "cell_type": "code",
2105 |    "execution_count": 320,
2106 |    "metadata": {},
2107 |    "outputs": [
2108 |     {
2109 |      "data": {
2110 |       "text/plain": [
2111 |        "3.1980674188547304"
2112 |       ]
2113 |      },
2114 |      "execution_count": 320,
2115 |      "metadata": {},
2116 |      "output_type": "execute_result"
2117 |     }
2118 |    ],
2119 |    "source": [
2120 |     "y_pred_rf = forest_reg.predict(x_train)\n",
2121 |     "forest_mse = mean_squared_error(y_train, y_pred_rf)\n",
2122 |     "forest_rmse = np.sqrt(forest_mse)\n",
2123 |     "forest_rmse"
2124 |    ]
2125 |   },
2126 |   {
2127 |    "cell_type": "markdown",
2128 |    "metadata": {},
2129 |    "source": [
2130 |     "误差为3.1980674188547304"
2131 |    ]
2132 |   },
2133 |   {
2134 |    "cell_type": "markdown",
2135 |    "metadata": {},
2136 |    "source": [
2137 |     "### 测试集检验"
2138 |    ]
2139 |   },
2140 |   {
2141 |    "cell_type": "code",
2142 |    "execution_count": 322,
2143 |    "metadata": {},
2144 |    "outputs": [
2145 |     {
2146 |      "data": {
2147 |       "text/plain": [
2148 |        "4.5536604518702575"
2149 |       ]
2150 |      },
2151 |      "execution_count": 322,
2152 |      "metadata": {},
2153 |      "output_type": "execute_result"
2154 |     }
2155 |    ],
2156 |    "source": [
2157 |     "y_pred_rf_test = forest_reg.predict(x_test)\n",
2158 |     "forest_mse = mean_squared_error(y_test, y_pred_rf_test)\n",
2159 |     "forest_rmse = np.sqrt(forest_mse)\n",
2160 |     "forest_rmse"
2161 |    ]
2162 |   },
2163 |   {
2164 |    "cell_type": "markdown",
2165 |    "metadata": {},
2166 |    "source": [
2167 |     "测试集标准差4.5536604518702575"
2168 |    ]
2169 |   },
2170 |   {
2171 |    "cell_type": "code",
2172 |    "execution_count": 355,
2173 |    "metadata": {},
2174 |    "outputs": [],
2175 |    "source": [
2176 |     "#datas_test.head(10)"
2177 |    ]
2178 |   },
2179 |   {
2180 |    "cell_type": "code",
2181 |    "execution_count": 324,
2182 |    "metadata": {},
2183 |    "outputs": [
2184 |     {
2185 |      "data": {
2186 |       "text/plain": [
2187 |        "array([[ 7.73      ],\n",
2188 |        "       [11.4861727 ],\n",
2189 |        "       [ 9.33285714],\n",
2190 |        "       [ 7.4       ],\n",
2191 |        "       [ 9.275     ],\n",
2192 |        "       [16.1       ],\n",
2193 |        "       [15.60247655],\n",
2194 |        "       [ 8.45447109],\n",
2195 |        "       [ 8.05528666],\n",
2196 |        "       [11.12662698]])"
2197 |       ]
2198 |      },
2199 |      "execution_count": 324,
2200 |      "metadata": {},
2201 |      "output_type": "execute_result"
2202 |     }
2203 |    ],
2204 |    "source": [
2205 |     "y_pred_rf[:10].reshape(10, 1)"
2206 |    ]
2207 |   },
2208 |   {
2209 |    "cell_type": "markdown",
2210 |    "metadata": {},
2211 |    "source": [
2212 |     "# 变量重要性"
2213 |    ]
2214 |   },
2215 |   {
2216 |    "cell_type": "code",
2217 |    "execution_count": 353,
2218 |    "metadata": {},
2219 |    "outputs": [
2220 |     {
2221 |      "data": {
2222 |       "text/plain": [
2223 |        "[(0.058026406821841056, '大专'),\n",
2224 |        " (0.031263812251122784, '10000人以上'),\n",
2225 |        " (0.020290832343490904, '1000-9999人'),\n",
2226 |        " (0.018095521323951716, '20-99人'),\n",
2227 |        " (0.015872193036716557, '100-499人'),\n",
2228 |        " (0.015170706225377684, '硕士'),\n",
2229 |        " (0.014925513158681906, '北京'),\n",
2230 |        " (0.014193573528348177, '500-999人'),\n",
2231 |        " (0.013911459035214633, '0-20人'),\n",
2232 |        " (0.011434958956342767, '深圳'),\n",
2233 |        " (0.011310119828051196, '本科'),\n",
2234 |        " (0.01002918014973044, '丽水'),\n",
2235 |        " (0.0096919398776096, '杭州'),\n",
2236 |        " (0.009393127333069419, '温州'),\n",
2237 |        " (0.00912590555299951, '学历不限'),\n",
2238 |        " (0.009077835450701817, '台州'),\n",
2239 |        " (0.008986219951521799, '舟山'),\n",
2240 |        " (0.008834491150945197, '绍兴'),\n",
2241 |        " (0.008656335871500801, '金华'),\n",
2242 |        " (0.008413046832938394, '湖州')]"
2243 |       ]
2244 |      },
2245 |      "execution_count": 353,
2246 |      "metadata": {},
2247 |      "output_type": "execute_result"
2248 |     }
2249 |    ],
2250 |    "source": [
2251 |     "#forest_reg.feature_importances_\n",
2252 |     "labellist = []\n",
2253 |     "\n",
2254 |     "for item in cat_encoder.categories_[0]:\n",
2255 |     "    labellist.append(item)\n",
2256 |     "    \n",
2257 |     "for item in cat_encoder.categories_[1]:\n",
2258 |     "    labellist.append(item)\n",
2259 |     "    \n",
2260 |     "for item in cat_encoder.categories_[2]:\n",
2261 |     "    labellist.append(item)\n",
2262 |     "#labellist\n",
2263 |     "sorted(zip(forest_reg.feature_importances_,labellist), reverse=True)[:20]"
2264 |    ]
2265 |   },
2266 |   {
2267 |    "cell_type": "code",
2268 |    "execution_count": 369,
2269 |    "metadata": {},
2270 |    "outputs": [
2271 |     {
2272 |      "data": {
2273 |       "text/html": [
2274 |        "<div>\n",
2275 |        "<style scoped>\n",
2276 |        "    .dataframe tbody tr th:only-of-type {\n",
2277 |        "        vertical-align: middle;\n",
2278 |        "    }\n",
2279 |        "\n",
2280 |        "    .dataframe tbody tr th {\n",
2281 |        "        vertical-align: top;\n",
2282 |        "    }\n",
2283 |        "\n",
2284 |        "    .dataframe thead th {\n",
2285 |        "        text-align: right;\n",
2286 |        "    }\n",
2287 |        "</style>\n",
2288 |        "<table border=\"1\" class=\"dataframe\">\n",
2289 |        "  <thead>\n",
2290 |        "    <tr style=\"text-align: right;\">\n",
2291 |        "      <th></th>\n",
2292 |        "      <th>重要性</th>\n",
2293 |        "    </tr>\n",
2294 |        "  </thead>\n",
2295 |        "  <tbody>\n",
2296 |        "    <tr>\n",
2297 |        "      <th>大专</th>\n",
2298 |        "      <td>0.058026</td>\n",
2299 |        "    </tr>\n",
2300 |        "    <tr>\n",
2301 |        "      <th>10000人以上</th>\n",
2302 |        "      <td>0.031264</td>\n",
2303 |        "    </tr>\n",
2304 |        "    <tr>\n",
2305 |        "      <th>1000-9999人</th>\n",
2306 |        "      <td>0.020291</td>\n",
2307 |        "    </tr>\n",
2308 |        "    <tr>\n",
2309 |        "      <th>20-99人</th>\n",
2310 |        "      <td>0.018096</td>\n",
2311 |        "    </tr>\n",
2312 |        "    <tr>\n",
2313 |        "      <th>100-499人</th>\n",
2314 |        "      <td>0.015872</td>\n",
2315 |        "    </tr>\n",
2316 |        "    <tr>\n",
2317 |        "      <th>硕士</th>\n",
2318 |        "      <td>0.015171</td>\n",
2319 |        "    </tr>\n",
2320 |        "    <tr>\n",
2321 |        "      <th>北京</th>\n",
2322 |        "      <td>0.014926</td>\n",
2323 |        "    </tr>\n",
2324 |        "    <tr>\n",
2325 |        "      <th>500-999人</th>\n",
2326 |        "      <td>0.014194</td>\n",
2327 |        "    </tr>\n",
2328 |        "    <tr>\n",
2329 |        "      <th>0-20人</th>\n",
2330 |        "      <td>0.013911</td>\n",
2331 |        "    </tr>\n",
2332 |        "    <tr>\n",
2333 |        "      <th>深圳</th>\n",
2334 |        "      <td>0.011435</td>\n",
2335 |        "    </tr>\n",
2336 |        "    <tr>\n",
2337 |        "      <th>本科</th>\n",
2338 |        "      <td>0.011310</td>\n",
2339 |        "    </tr>\n",
2340 |        "    <tr>\n",
2341 |        "      <th>丽水</th>\n",
2342 |        "      <td>0.010029</td>\n",
2343 |        "    </tr>\n",
2344 |        "    <tr>\n",
2345 |        "      <th>杭州</th>\n",
2346 |        "      <td>0.009692</td>\n",
2347 |        "    </tr>\n",
2348 |        "    <tr>\n",
2349 |        "      <th>温州</th>\n",
2350 |        "      <td>0.009393</td>\n",
2351 |        "    </tr>\n",
2352 |        "    <tr>\n",
2353 |        "      <th>学历不限</th>\n",
2354 |        "      <td>0.009126</td>\n",
2355 |        "    </tr>\n",
2356 |        "    <tr>\n",
2357 |        "      <th>台州</th>\n",
2358 |        "      <td>0.009078</td>\n",
2359 |        "    </tr>\n",
2360 |        "    <tr>\n",
2361 |        "      <th>舟山</th>\n",
2362 |        "      <td>0.008986</td>\n",
2363 |        "    </tr>\n",
2364 |        "    <tr>\n",
2365 |        "      <th>绍兴</th>\n",
2366 |        "      <td>0.008834</td>\n",
2367 |        "    </tr>\n",
2368 |        "    <tr>\n",
2369 |        "      <th>金华</th>\n",
2370 |        "      <td>0.008656</td>\n",
2371 |        "    </tr>\n",
2372 |        "    <tr>\n",
2373 |        "      <th>湖州</th>\n",
2374 |        "      <td>0.008413</td>\n",
2375 |        "    </tr>\n",
2376 |        "  </tbody>\n",
2377 |        "</table>\n",
2378 |        "</div>"
2379 |       ],
2380 |       "text/plain": [
2381 |        "                 重要性\n",
2382 |        "大专          0.058026\n",
2383 |        "10000人以上    0.031264\n",
2384 |        "1000-9999人  0.020291\n",
2385 |        "20-99人      0.018096\n",
2386 |        "100-499人    0.015872\n",
2387 |        "硕士          0.015171\n",
2388 |        "北京          0.014926\n",
2389 |        "500-999人    0.014194\n",
2390 |        "0-20人       0.013911\n",
2391 |        "深圳          0.011435\n",
2392 |        "本科          0.011310\n",
2393 |        "丽水          0.010029\n",
2394 |        "杭州          0.009692\n",
2395 |        "温州          0.009393\n",
2396 |        "学历不限        0.009126\n",
2397 |        "台州          0.009078\n",
2398 |        "舟山          0.008986\n",
2399 |        "绍兴          0.008834\n",
2400 |        "金华          0.008656\n",
2401 |        "湖州          0.008413"
2402 |       ]
2403 |      },
2404 |      "execution_count": 369,
2405 |      "metadata": {},
2406 |      "output_type": "execute_result"
2407 |     }
2408 |    ],
2409 |    "source": [
2410 |     "from pandas import DataFrame\n",
2411 |     "# 转换成 DataFrame格式\n",
2412 |     "sorted(zip(forest_reg.feature_importances_,labellist), reverse=True)[:20][0]\n",
2413 |     "list1 = []\n",
2414 |     "list2 = []\n",
2415 |     "for item in sorted(zip(forest_reg.feature_importances_,labellist), reverse=True)[:20]:\n",
2416 |     "    list1.append(item[1])\n",
2417 |     "    list2.append(item[0])\n",
2418 |     "df_importance = DataFrame(list2,index=list1,columns=['重要性'])\n",
2419 |     "df_importance"
2420 |    ]
2421 |   },
2422 |   {
2423 |    "cell_type": "markdown",
2424 |    "metadata": {},
2425 |    "source": [
2426 |     "# 变量重要性分析"
2427 |    ]
2428 |   },
2429 |   {
2430 |    "cell_type": "markdown",
2431 |    "metadata": {},
2432 |    "source": [
2433 |     "## 选取前20个重要变量"
2434 |    ]
2435 |   },
2436 |   {
2437 |    "cell_type": "markdown",
2438 |    "metadata": {},
2439 |    "source": [
2440 |     "1.很多公司招聘时往往将学历要求设置为'大专'（即不设门槛，只挑选有能力者）"
2441 |    ]
2442 |   },
2443 |   {
2444 |    "cell_type": "markdown",
2445 |    "metadata": {},
2446 |    "source": [
2447 |     "2.公司规模是影响数据分析师薪资方面的重要变量，其中'10000人以上'的大公司在薪资高低方面的影响力最大。"
2448 |    ]
2449 |   },
2450 |   {
2451 |    "cell_type": "markdown",
2452 |    "metadata": {},
2453 |    "source": [
2454 |     "3.也有许多公司将'硕士学历'设为数据分析师的门槛，因此'硕士学历'在数据分析师薪资方面影响较大。"
2455 |    ]
2456 |   },
2457 |   {
2458 |    "cell_type": "markdown",
2459 |    "metadata": {},
2460 |    "source": [
2461 |     "4.城市/地域方面：可以看到北京、深圳的职位对薪资的影响最大；其次，丽水、杭州、温州、台州、舟山、绍兴、金华等均为浙江省的市，可以明显看出浙江省数据分析师的薪资待遇较高。"
2462 |    ]
2463 |   },
2464 |   {
2465 |    "cell_type": "code",
2466 |    "execution_count": 370,
2467 |    "metadata": {},
2468 |    "outputs": [],
2469 |    "source": [
2470 |     "#datas_train[datas_train['学历要求'] == '大专'].head()"
2471 |    ]
2472 |   },
2473 |   {
2474 |    "cell_type": "markdown",
2475 |    "metadata": {},
2476 |    "source": [
2477 |     "# 自定义变量预测"
2478 |    ]
2479 |   },
2480 |   {
2481 |    "cell_type": "code",
2482 |    "execution_count": 451,
2483 |    "metadata": {},
2484 |    "outputs": [
2485 |     {
2486 |      "data": {
2487 |       "text/plain": [
2488 |        "[array(['三明', '上海', '东沙群岛', '东莞', '东营', '中山', '临沂', '丽水', '乐山', '云浮', '亳州',\n",
2489 |        "        '仙桃', '佛山', '六安', '内江', '凉山', '北京', '十堰', '南京', '南充', '南平', '南通',\n",
2490 |        "        '南阳', '厦门', '台州', '合肥', '周口', '咸宁', '咸阳', '商丘', '商洛', '嘉兴', '大连',\n",
2491 |        "        '天津', '天门', '威海', '娄底', '孝感', '宁德', '安庆', '安康', '安阳', '宜宾', '宜昌',\n",
2492 |        "        '宣城', '宿州', '宿迁', '岳阳', '巴中', '常州', '常德', '平顶山', '广元', '广安', '广州',\n",
2493 |        "        '开封', '张家界', '徐州', '德阳', '怀化', '恩施', '惠州', '成都', '扬州', '揭阳', '攀枝花',\n",
2494 |        "        '新乡', '日照', '杭州', '枣庄', '株洲', '梅州', '武汉', '永州', '汕头', '汕尾', '江门',\n",
2495 |        "        '池州', '河源', '泉州', '泰安', '泰州', '泸州', '洛阳', '济南', '济宁', '淄博', '淮北',\n",
2496 |        "        '淮安', '深圳', '清远', '温州', '湖州', '湘潭', '湘西', '湛江', '滨州', '漯河', '漳州',\n",
2497 |        "        '潍坊', '潜江', '潮州', '烟台', '珠海', '甘孜', '益阳', '盐城', '眉山', '神农架', '福州',\n",
2498 |        "        '绍兴', '绵阳', '聊城', '肇庆', '自贡', '舟山', '芜湖', '苏州', '茂名', '荆州', '荆门',\n",
2499 |        "        '莆田', '莱芜', '菏泽', '衡阳', '衢州', '襄阳', '西安', '资阳', '达州', '连云港', '遂宁',\n",
2500 |        "        '邵阳', '郑州', '郴州', '鄂州', '金华', '铜川', '镇江', '长沙', '阜阳', '阳江', '阿坝',\n",
2501 |        "        '随州', '雅安', '韶关', '黄冈', '黄山', '黄石', '龙岩'], dtype=object),\n",
2502 |        " array(['0-20人', '100-499人', '1000-9999人', '10000人以上', '20-99人',\n",
2503 |        "        '500-999人'], dtype=object),\n",
2504 |        " array(['中专', '博士', '大专', '学历不限', '本科', '硕士', '高中'], dtype=object),\n",
2505 |        " array(['1-3年', '1年以内', '3-5年', '5-10年', '经验不限'], dtype=object)]"
2506 |       ]
2507 |      },
2508 |      "execution_count": 451,
2509 |      "metadata": {},
2510 |      "output_type": "execute_result"
2511 |     }
2512 |    ],
2513 |    "source": [
2514 |     "cat_encoder.categories_ #可选预测变量"
2515 |    ]
2516 |   },
2517 |   {
2518 |    "cell_type": "markdown",
2519 |    "metadata": {},
2520 |    "source": [
2521 |     "## 决策树预测 "
2522 |    ]
2523 |   },
2524 |   {
2525 |    "cell_type": "code",
2526 |    "execution_count": 452,
2527 |    "metadata": {},
2528 |    "outputs": [
2529 |     {
2530 |      "data": {
2531 |       "text/plain": [
2532 |        "array([15.])"
2533 |       ]
2534 |      },
2535 |      "execution_count": 452,
2536 |      "metadata": {},
2537 |      "output_type": "execute_result"
2538 |     }
2539 |    ],
2540 |    "source": [
2541 |     "#paramlist = func_params(['上海','1000-9999人','硕士','1-3年'])\n",
2542 |     "Y_pred_dt = tree_reg.predict(paramlist)\n",
2543 |     "Y_pred_dt"
2544 |    ]
2545 |   },
2546 |   {
2547 |    "cell_type": "code",
2548 |    "execution_count": 458,
2549 |    "metadata": {},
2550 |    "outputs": [
2551 |     {
2552 |      "data": {
2553 |       "text/plain": [
2554 |        "array([13.5])"
2555 |       ]
2556 |      },
2557 |      "execution_count": 458,
2558 |      "metadata": {},
2559 |      "output_type": "execute_result"
2560 |     }
2561 |    ],
2562 |    "source": [
2563 |     "paramlist2 = func_params(['北京','500-999人','本科','1-3年'])\n",
2564 |     "tree_reg.predict(paramlist2)"
2565 |    ]
2566 |   },
2567 |   {
2568 |    "cell_type": "markdown",
2569 |    "metadata": {},
2570 |    "source": [
2571 |     "## 随机森林预测"
2572 |    ]
2573 |   },
2574 |   {
2575 |    "cell_type": "code",
2576 |    "execution_count": 469,
2577 |    "metadata": {},
2578 |    "outputs": [
2579 |     {
2580 |      "data": {
2581 |       "text/plain": [
2582 |        "array([12.63888889])"
2583 |       ]
2584 |      },
2585 |      "execution_count": 469,
2586 |      "metadata": {},
2587 |      "output_type": "execute_result"
2588 |     }
2589 |    ],
2590 |    "source": [
2591 |     "paramlist3 = func_params(['深圳','100-499人','硕士','1年以内'])\n",
2592 |     "tree_reg.predict(paramlist3)"
2593 |    ]
2594 |   },
2595 |   {
2596 |    "cell_type": "code",
2597 |    "execution_count": 476,
2598 |    "metadata": {},
2599 |    "outputs": [
2600 |     {
2601 |      "data": {
2602 |       "text/plain": [
2603 |        "array([15.])"
2604 |       ]
2605 |      },
2606 |      "execution_count": 476,
2607 |      "metadata": {},
2608 |      "output_type": "execute_result"
2609 |     }
2610 |    ],
2611 |    "source": [
2612 |     "paramlist4 = func_params(['上海','1000-9999人','硕士','1年以内'])\n",
2613 |     "tree_reg.predict(paramlist4)"
2614 |    ]
2615 |   },
2616 |   {
2617 |    "cell_type": "code",
2618 |    "execution_count": null,
2619 |    "metadata": {
2620 |     "collapsed": true
2621 |    },
2622 |    "outputs": [],
2623 |    "source": []
2624 |   }
2625 |  ],
2626 |  "metadata": {
2627 |   "kernelspec": {
2628 |    "display_name": "Python 3",
2629 |    "language": "python",
2630 |    "name": "python3"
2631 |   },
2632 |   "language_info": {
2633 |    "codemirror_mode": {
2634 |     "name": "ipython",
2635 |     "version": 3
2636 |    },
2637 |    "file_extension": ".py",
2638 |    "mimetype": "text/x-python",
2639 |    "name": "python",
2640 |    "nbconvert_exporter": "python",
2641 |    "pygments_lexer": "ipython3",
2642 |    "version": "3.6.3"
2643 |   }
2644 |  },
2645 |  "nbformat": 4,
2646 |  "nbformat_minor": 2
2647 | }
2648 | 


--------------------------------------------------------------------------------
/city_data/city.csv:
--------------------------------------------------------------------------------
  1 | ,code,市,省
  2 | 0,101010100,北京,北京
  3 | 1,101020100,上海,上海
  4 | 2,101030100,天津,天津
  5 | 3,101040100,重庆,重庆
  6 | 4,101050100,哈尔滨,黑龙江
  7 | 5,101050200,齐齐哈尔,黑龙江
  8 | 6,101050300,牡丹江,黑龙江
  9 | 7,101050400,佳木斯,黑龙江
 10 | 8,101050500,绥化,黑龙江
 11 | 9,101050600,黑河,黑龙江
 12 | 10,101050700,伊春,黑龙江
 13 | 11,101050800,大庆,黑龙江
 14 | 12,101050900,七台河,黑龙江
 15 | 13,101051000,鸡西,黑龙江
 16 | 14,101051100,鹤岗,黑龙江
 17 | 15,101051200,双鸭山,黑龙江
 18 | 16,101051300,大兴安岭,黑龙江
 19 | 17,101060100,长春,吉林
 20 | 18,101060200,吉林,吉林
 21 | 19,101060300,四平,吉林
 22 | 20,101060400,通化,吉林
 23 | 21,101060500,白城,吉林
 24 | 22,101060600,辽源,吉林
 25 | 23,101060700,松原,吉林
 26 | 24,101060800,白山,吉林
 27 | 25,101060900,延边,吉林
 28 | 26,101070100,沈阳,辽宁
 29 | 27,101070200,大连,辽宁
 30 | 28,101070300,鞍山,辽宁
 31 | 29,101070400,抚顺,辽宁
 32 | 30,101070500,本溪,辽宁
 33 | 31,101070600,丹东,辽宁
 34 | 32,101070700,锦州,辽宁
 35 | 33,101070800,营口,辽宁
 36 | 34,101070900,阜新,辽宁
 37 | 35,101071000,辽阳,辽宁
 38 | 36,101071100,铁岭,辽宁
 39 | 37,101071200,朝阳,辽宁
 40 | 38,101071300,盘锦,辽宁
 41 | 39,101071400,葫芦岛,辽宁
 42 | 40,101080100,呼和浩特,内蒙古
 43 | 41,101080200,包头,内蒙古
 44 | 42,101080300,乌海,内蒙古
 45 | 43,101080400,通辽,内蒙古
 46 | 44,101080500,赤峰,内蒙古
 47 | 45,101080600,鄂尔多斯,内蒙古
 48 | 46,101080700,呼伦贝尔,内蒙古
 49 | 47,101080800,巴彦淖尔,内蒙古
 50 | 48,101080900,乌兰察布,内蒙古
 51 | 49,101081000,锡林郭勒,内蒙古
 52 | 50,101081100,兴安盟,内蒙古
 53 | 51,101081200,阿拉善,内蒙古
 54 | 52,101090100,石家庄,河北
 55 | 53,101090200,保定,河北
 56 | 54,101090300,张家口,河北
 57 | 55,101090400,承德,河北
 58 | 56,101090500,唐山,河北
 59 | 57,101090600,廊坊,河北
 60 | 58,101090700,沧州,河北
 61 | 59,101090800,衡水,河北
 62 | 60,101090900,邢台,河北
 63 | 61,101091000,邯郸,河北
 64 | 62,101091100,秦皇岛,河北
 65 | 63,101100100,太原,山西
 66 | 64,101100200,大同,山西
 67 | 65,101100300,阳泉,山西
 68 | 66,101100400,晋中,山西
 69 | 67,101100500,长治,山西
 70 | 68,101100600,晋城,山西
 71 | 69,101100700,临汾,山西
 72 | 70,101100800,运城,山西
 73 | 71,101100900,朔州,山西
 74 | 72,101101000,忻州,山西
 75 | 73,101101100,吕梁,山西
 76 | 74,101110100,西安,陕西
 77 | 75,101110200,咸阳,陕西
 78 | 76,101110300,延安,陕西
 79 | 77,101110400,榆林,陕西
 80 | 78,101110500,渭南,陕西
 81 | 79,101110600,商洛,陕西
 82 | 80,101110700,安康,陕西
 83 | 81,101110800,汉中,陕西
 84 | 82,101110900,宝鸡,陕西
 85 | 83,101111000,铜川,陕西
 86 | 84,101120100,济南,山东
 87 | 85,101120200,青岛,山东
 88 | 86,101120300,淄博,山东
 89 | 87,101120400,德州,山东
 90 | 88,101120500,烟台,山东
 91 | 89,101120600,潍坊,山东
 92 | 90,101120700,济宁,山东
 93 | 91,101120800,泰安,山东
 94 | 92,101120900,临沂,山东
 95 | 93,101121000,菏泽,山东
 96 | 94,101121100,滨州,山东
 97 | 95,101121200,东营,山东
 98 | 96,101121300,威海,山东
 99 | 97,101121400,枣庄,山东
100 | 98,101121500,日照,山东
101 | 99,101121600,莱芜,山东
102 | 100,101121700,聊城,山东
103 | 101,101130100,乌鲁木齐,新疆
104 | 102,101130200,克拉玛依,新疆
105 | 103,101130300,昌吉,新疆
106 | 104,101130400,巴音郭楞,新疆
107 | 105,101130500,博尔塔拉,新疆
108 | 106,101130600,伊犁,新疆
109 | 107,101130800,吐鲁番,新疆
110 | 108,101130900,哈密,新疆
111 | 109,101131000,阿克苏,新疆
112 | 110,101131100,克孜勒苏柯尔克孜,新疆
113 | 111,101131200,喀什,新疆
114 | 112,101131300,和田,新疆
115 | 113,101131400,塔城,新疆
116 | 114,101131500,阿勒泰,新疆
117 | 115,101131600,石河子,新疆
118 | 116,101131700,阿拉尔,新疆
119 | 117,101131800,图木舒克,新疆
120 | 118,101131900,五家渠,新疆
121 | 119,101132000,铁门关,新疆
122 | 120,101132100,北屯市,新疆
123 | 121,101132200,可克达拉市,新疆
124 | 122,101132300,昆玉市,新疆
125 | 123,101132400,双河市,新疆
126 | 124,101150100,西宁,青海
127 | 125,101150200,海东,青海
128 | 126,101150300,海北,青海
129 | 127,101150400,黄南,青海
130 | 128,101150500,海南,青海
131 | 129,101150600,果洛,青海
132 | 130,101150700,玉树,青海
133 | 131,101150800,海西,青海
134 | 132,101160100,兰州,甘肃
135 | 133,101160200,定西,甘肃
136 | 134,101160300,平凉,甘肃
137 | 135,101160400,庆阳,甘肃
138 | 136,101160500,武威,甘肃
139 | 137,101160600,金昌,甘肃
140 | 138,101160700,张掖,甘肃
141 | 139,101160800,酒泉,甘肃
142 | 140,101160900,天水,甘肃
143 | 141,101161000,白银,甘肃
144 | 142,101161100,陇南,甘肃
145 | 143,101161200,嘉峪关,甘肃
146 | 144,101161300,临夏,甘肃
147 | 145,101161400,甘南,甘肃
148 | 146,101170100,银川,宁夏
149 | 147,101170200,石嘴山,宁夏
150 | 148,101170300,吴忠,宁夏
151 | 149,101170400,固原,宁夏
152 | 150,101170500,中卫,宁夏
153 | 151,101180100,郑州,河南
154 | 152,101180200,安阳,河南
155 | 153,101180300,新乡,河南
156 | 154,101180400,许昌,河南
157 | 155,101180500,平顶山,河南
158 | 156,101180600,信阳,河南
159 | 157,101180700,南阳,河南
160 | 158,101180800,开封,河南
161 | 159,101180900,洛阳,河南
162 | 160,101181000,商丘,河南
163 | 161,101181100,焦作,河南
164 | 162,101181200,鹤壁,河南
165 | 163,101181300,濮阳,河南
166 | 164,101181400,周口,河南
167 | 165,101181500,漯河,河南
168 | 166,101181600,驻马店,河南
169 | 167,101181700,三门峡,河南
170 | 168,101181800,济源,河南
171 | 169,101190100,南京,江苏
172 | 170,101190200,无锡,江苏
173 | 171,101190300,镇江,江苏
174 | 172,101190400,苏州,江苏
175 | 173,101190500,南通,江苏
176 | 174,101190600,扬州,江苏
177 | 175,101190700,盐城,江苏
178 | 176,101190800,徐州,江苏
179 | 177,101190900,淮安,江苏
180 | 178,101191000,连云港,江苏
181 | 179,101191100,常州,江苏
182 | 180,101191200,泰州,江苏
183 | 181,101191300,宿迁,江苏
184 | 182,101200100,武汉,湖北
185 | 183,101200200,襄阳,湖北
186 | 184,101200300,鄂州,湖北
187 | 185,101200400,孝感,湖北
188 | 186,101200500,黄冈,湖北
189 | 187,101200600,黄石,湖北
190 | 188,101200700,咸宁,湖北
191 | 189,101200800,荆州,湖北
192 | 190,101200900,宜昌,湖北
193 | 191,101201000,十堰,湖北
194 | 192,101201100,随州,湖北
195 | 193,101201200,荆门,湖北
196 | 194,101201300,恩施,湖北
197 | 195,101201400,仙桃,湖北
198 | 196,101201500,潜江,湖北
199 | 197,101201600,天门,湖北
200 | 198,101201700,神农架,湖北
201 | 199,101210100,杭州,浙江
202 | 200,101210200,湖州,浙江
203 | 201,101210300,嘉兴,浙江
204 | 202,101210400,宁波,浙江
205 | 203,101210500,绍兴,浙江
206 | 204,101210600,台州,浙江
207 | 205,101210700,温州,浙江
208 | 206,101210800,丽水,浙江
209 | 207,101210900,金华,浙江
210 | 208,101211000,衢州,浙江
211 | 209,101211100,舟山,浙江
212 | 210,101220100,合肥,安徽
213 | 211,101220200,蚌埠,安徽
214 | 212,101220300,芜湖,安徽
215 | 213,101220400,淮南,安徽
216 | 214,101220500,马鞍山,安徽
217 | 215,101220600,安庆,安徽
218 | 216,101220700,宿州,安徽
219 | 217,101220800,阜阳,安徽
220 | 218,101220900,亳州,安徽
221 | 219,101221000,滁州,安徽
222 | 220,101221100,淮北,安徽
223 | 221,101221200,铜陵,安徽
224 | 222,101221300,宣城,安徽
225 | 223,101221400,六安,安徽
226 | 224,101221500,池州,安徽
227 | 225,101221600,黄山,安徽
228 | 226,101230100,福州,福建
229 | 227,101230200,厦门,福建
230 | 228,101230300,宁德,福建
231 | 229,101230400,莆田,福建
232 | 230,101230500,泉州,福建
233 | 231,101230600,漳州,福建
234 | 232,101230700,龙岩,福建
235 | 233,101230800,三明,福建
236 | 234,101230900,南平,福建
237 | 235,101240100,南昌,江西
238 | 236,101240200,九江,江西
239 | 237,101240300,上饶,江西
240 | 238,101240400,抚州,江西
241 | 239,101240500,宜春,江西
242 | 240,101240600,吉安,江西
243 | 241,101240700,赣州,江西
244 | 242,101240800,景德镇,江西
245 | 243,101240900,萍乡,江西
246 | 244,101241000,新余,江西
247 | 245,101241100,鹰潭,江西
248 | 246,101250100,长沙,湖南
249 | 247,101250200,湘潭,湖南
250 | 248,101250300,株洲,湖南
251 | 249,101250400,衡阳,湖南
252 | 250,101250500,郴州,湖南
253 | 251,101250600,常德,湖南
254 | 252,101250700,益阳,湖南
255 | 253,101250800,娄底,湖南
256 | 254,101250900,邵阳,湖南
257 | 255,101251000,岳阳,湖南
258 | 256,101251100,张家界,湖南
259 | 257,101251200,怀化,湖南
260 | 258,101251300,永州,湖南
261 | 259,101251400,湘西,湖南
262 | 260,101260100,贵阳,贵州
263 | 261,101260200,遵义,贵州
264 | 262,101260300,安顺,贵州
265 | 263,101260400,铜仁,贵州
266 | 264,101260500,毕节,贵州
267 | 265,101260600,六盘水,贵州
268 | 266,101260700,黔东南,贵州
269 | 267,101260800,黔南,贵州
270 | 268,101260900,黔西南,贵州
271 | 269,101270100,成都,四川
272 | 270,101270200,攀枝花,四川
273 | 271,101270300,自贡,四川
274 | 272,101270400,绵阳,四川
275 | 273,101270500,南充,四川
276 | 274,101270600,达州,四川
277 | 275,101270700,遂宁,四川
278 | 276,101270800,广安,四川
279 | 277,101270900,巴中,四川
280 | 278,101271000,泸州,四川
281 | 279,101271100,宜宾,四川
282 | 280,101271200,内江,四川
283 | 281,101271300,资阳,四川
284 | 282,101271400,乐山,四川
285 | 283,101271500,眉山,四川
286 | 284,101271600,雅安,四川
287 | 285,101271700,德阳,四川
288 | 286,101271800,广元,四川
289 | 287,101271900,阿坝,四川
290 | 288,101272000,凉山,四川
291 | 289,101272100,甘孜,四川
292 | 290,101280100,广州,广东
293 | 291,101280200,韶关,广东
294 | 292,101280300,惠州,广东
295 | 293,101280400,梅州,广东
296 | 294,101280500,汕头,广东
297 | 295,101280600,深圳,广东
298 | 296,101280700,珠海,广东
299 | 297,101280800,佛山,广东
300 | 298,101280900,肇庆,广东
301 | 299,101281000,湛江,广东
302 | 300,101281100,江门,广东
303 | 301,101281200,河源,广东
304 | 302,101281300,清远,广东
305 | 303,101281400,云浮,广东
306 | 304,101281500,潮州,广东
307 | 305,101281600,东莞,广东
308 | 306,101281700,中山,广东
309 | 307,101281800,阳江,广东
310 | 308,101281900,揭阳,广东
311 | 309,101282000,茂名,广东
312 | 310,101282100,汕尾,广东
313 | 311,101282200,东沙群岛,广东
314 | 312,101290100,昆明,云南
315 | 313,101290200,曲靖,云南
316 | 314,101290300,保山,云南
317 | 315,101290400,玉溪,云南
318 | 316,101290500,普洱,云南
319 | 317,101290700,昭通,云南
320 | 318,101290800,临沧,云南
321 | 319,101290900,丽江,云南
322 | 320,101291000,西双版纳,云南
323 | 321,101291100,文山,云南
324 | 322,101291200,红河,云南
325 | 323,101291300,德宏,云南
326 | 324,101291400,怒江,云南
327 | 325,101291500,迪庆,云南
328 | 326,101291600,大理,云南
329 | 327,101291700,楚雄,云南
330 | 328,101300100,南宁,广西
331 | 329,101300200,崇左,广西
332 | 330,101300300,柳州,广西
333 | 331,101300400,来宾,广西
334 | 332,101300500,桂林,广西
335 | 333,101300600,梧州,广西
336 | 334,101300700,贺州,广西
337 | 335,101300800,贵港,广西
338 | 336,101300900,玉林,广西
339 | 337,101301000,百色,广西
340 | 338,101301100,钦州,广西
341 | 339,101301200,河池,广西
342 | 340,101301300,北海,广西
343 | 341,101301400,防城港,广西
344 | 342,101310100,海口,海南
345 | 343,101310200,三亚,海南
346 | 344,101310300,三沙,海南
347 | 345,101310400,儋州,海南
348 | 346,101310500,五指山,海南
349 | 347,101310600,琼海,海南
350 | 348,101310700,文昌,海南
351 | 349,101310800,万宁,海南
352 | 350,101310900,东方,海南
353 | 351,101311000,定安,海南
354 | 352,101311100,屯昌,海南
355 | 353,101311200,澄迈,海南
356 | 354,101311300,临高,海南
357 | 355,101311400,白沙,海南
358 | 356,101311500,昌江,海南
359 | 357,101311600,乐东,海南
360 | 358,101311700,陵水,海南
361 | 359,101311800,保亭,海南
362 | 360,101311900,琼中,海南
363 | 361,101341100,台湾,台湾
364 | 362,101140100,拉萨,西藏
365 | 363,101140200,日喀则,西藏
366 | 364,101140300,昌都,西藏
367 | 365,101140400,林芝,西藏
368 | 366,101140500,山南,西藏
369 | 367,101140600,那曲,西藏
370 | 368,101140700,阿里,西藏
371 | 369,101320300,香港,香港
372 | 370,101330100,澳门,澳门
373 | 


--------------------------------------------------------------------------------
/city_data/city.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 9,
  6 |    "metadata": {
  7 |     "collapsed": true
  8 |    },
  9 |    "outputs": [],
 10 |    "source": [
 11 |     "import pymongo\n",
 12 |     "import pandas as pd\n",
 13 |     "client = pymongo.MongoClient('localhost',27017)\n",
 14 |     "db  = client['Graduation_project']\n",
 15 |     "table = db['city']\n",
 16 |     "city = pd.DataFrame(list(table.find()))\n",
 17 |     "del city['_id']"
 18 |    ]
 19 |   },
 20 |   {
 21 |    "cell_type": "code",
 22 |    "execution_count": 11,
 23 |    "metadata": {},
 24 |    "outputs": [
 25 |     {
 26 |      "data": {
 27 |       "text/html": [
 28 |        "<div>\n",
 29 |        "<style scoped>\n",
 30 |        "    .dataframe tbody tr th:only-of-type {\n",
 31 |        "        vertical-align: middle;\n",
 32 |        "    }\n",
 33 |        "\n",
 34 |        "    .dataframe tbody tr th {\n",
 35 |        "        vertical-align: top;\n",
 36 |        "    }\n",
 37 |        "\n",
 38 |        "    .dataframe thead th {\n",
 39 |        "        text-align: right;\n",
 40 |        "    }\n",
 41 |        "</style>\n",
 42 |        "<table border=\"1\" class=\"dataframe\">\n",
 43 |        "  <thead>\n",
 44 |        "    <tr style=\"text-align: right;\">\n",
 45 |        "      <th></th>\n",
 46 |        "      <th>code</th>\n",
 47 |        "      <th>市</th>\n",
 48 |        "      <th>省</th>\n",
 49 |        "    </tr>\n",
 50 |        "  </thead>\n",
 51 |        "  <tbody>\n",
 52 |        "    <tr>\n",
 53 |        "      <th>0</th>\n",
 54 |        "      <td>101010100</td>\n",
 55 |        "      <td>北京</td>\n",
 56 |        "      <td>北京</td>\n",
 57 |        "    </tr>\n",
 58 |        "    <tr>\n",
 59 |        "      <th>1</th>\n",
 60 |        "      <td>101020100</td>\n",
 61 |        "      <td>上海</td>\n",
 62 |        "      <td>上海</td>\n",
 63 |        "    </tr>\n",
 64 |        "    <tr>\n",
 65 |        "      <th>2</th>\n",
 66 |        "      <td>101030100</td>\n",
 67 |        "      <td>天津</td>\n",
 68 |        "      <td>天津</td>\n",
 69 |        "    </tr>\n",
 70 |        "    <tr>\n",
 71 |        "      <th>3</th>\n",
 72 |        "      <td>101040100</td>\n",
 73 |        "      <td>重庆</td>\n",
 74 |        "      <td>重庆</td>\n",
 75 |        "    </tr>\n",
 76 |        "    <tr>\n",
 77 |        "      <th>4</th>\n",
 78 |        "      <td>101050100</td>\n",
 79 |        "      <td>哈尔滨</td>\n",
 80 |        "      <td>黑龙江</td>\n",
 81 |        "    </tr>\n",
 82 |        "  </tbody>\n",
 83 |        "</table>\n",
 84 |        "</div>"
 85 |       ],
 86 |       "text/plain": [
 87 |        "        code    市    省\n",
 88 |        "0  101010100   北京   北京\n",
 89 |        "1  101020100   上海   上海\n",
 90 |        "2  101030100   天津   天津\n",
 91 |        "3  101040100   重庆   重庆\n",
 92 |        "4  101050100  哈尔滨  黑龙江"
 93 |       ]
 94 |      },
 95 |      "execution_count": 11,
 96 |      "metadata": {},
 97 |      "output_type": "execute_result"
 98 |     }
 99 |    ],
100 |    "source": [
101 |     "city.head()"
102 |    ]
103 |   },
104 |   {
105 |    "cell_type": "code",
106 |    "execution_count": 13,
107 |    "metadata": {
108 |     "collapsed": true
109 |    },
110 |    "outputs": [],
111 |    "source": [
112 |     "city.to_csv('city.csv')"
113 |    ]
114 |   },
115 |   {
116 |    "cell_type": "code",
117 |    "execution_count": 16,
118 |    "metadata": {},
119 |    "outputs": [
120 |     {
121 |      "data": {
122 |       "text/html": [
123 |        "<div>\n",
124 |        "<style scoped>\n",
125 |        "    .dataframe tbody tr th:only-of-type {\n",
126 |        "        vertical-align: middle;\n",
127 |        "    }\n",
128 |        "\n",
129 |        "    .dataframe tbody tr th {\n",
130 |        "        vertical-align: top;\n",
131 |        "    }\n",
132 |        "\n",
133 |        "    .dataframe thead th {\n",
134 |        "        text-align: right;\n",
135 |        "    }\n",
136 |        "</style>\n",
137 |        "<table border=\"1\" class=\"dataframe\">\n",
138 |        "  <thead>\n",
139 |        "    <tr style=\"text-align: right;\">\n",
140 |        "      <th></th>\n",
141 |        "      <th>code</th>\n",
142 |        "      <th>省</th>\n",
143 |        "      <th>市</th>\n",
144 |        "    </tr>\n",
145 |        "  </thead>\n",
146 |        "  <tbody>\n",
147 |        "    <tr>\n",
148 |        "      <th>0</th>\n",
149 |        "      <td>101010100</td>\n",
150 |        "      <td>北京</td>\n",
151 |        "      <td>北京</td>\n",
152 |        "    </tr>\n",
153 |        "    <tr>\n",
154 |        "      <th>1</th>\n",
155 |        "      <td>101020100</td>\n",
156 |        "      <td>上海</td>\n",
157 |        "      <td>上海</td>\n",
158 |        "    </tr>\n",
159 |        "    <tr>\n",
160 |        "      <th>2</th>\n",
161 |        "      <td>101030100</td>\n",
162 |        "      <td>天津</td>\n",
163 |        "      <td>天津</td>\n",
164 |        "    </tr>\n",
165 |        "    <tr>\n",
166 |        "      <th>3</th>\n",
167 |        "      <td>101040100</td>\n",
168 |        "      <td>重庆</td>\n",
169 |        "      <td>重庆</td>\n",
170 |        "    </tr>\n",
171 |        "    <tr>\n",
172 |        "      <th>4</th>\n",
173 |        "      <td>101050100</td>\n",
174 |        "      <td>黑龙江</td>\n",
175 |        "      <td>哈尔滨</td>\n",
176 |        "    </tr>\n",
177 |        "  </tbody>\n",
178 |        "</table>\n",
179 |        "</div>"
180 |       ],
181 |       "text/plain": [
182 |        "        code    省    市\n",
183 |        "0  101010100   北京   北京\n",
184 |        "1  101020100   上海   上海\n",
185 |        "2  101030100   天津   天津\n",
186 |        "3  101040100   重庆   重庆\n",
187 |        "4  101050100  黑龙江  哈尔滨"
188 |       ]
189 |      },
190 |      "execution_count": 16,
191 |      "metadata": {},
192 |      "output_type": "execute_result"
193 |     }
194 |    ],
195 |    "source": [
196 |     "import pandas as pd\n",
197 |     "\n",
198 |     "city = pd.read_csv('city.csv')\n",
199 |     "data = city[['code', '省', '市']]\n",
200 |     "data.head()"
201 |    ]
202 |   },
203 |   {
204 |    "cell_type": "code",
205 |    "execution_count": 20,
206 |    "metadata": {
207 |     "collapsed": true
208 |    },
209 |    "outputs": [],
210 |    "source": [
211 |     "city = data.values"
212 |    ]
213 |   },
214 |   {
215 |    "cell_type": "code",
216 |    "execution_count": 21,
217 |    "metadata": {},
218 |    "outputs": [
219 |     {
220 |      "data": {
221 |       "text/plain": [
222 |        "array([[101010100, '北京', '北京'],\n",
223 |        "       [101020100, '上海', '上海'],\n",
224 |        "       [101030100, '天津', '天津'],\n",
225 |        "       ...,\n",
226 |        "       [101140700, '西藏', '阿里'],\n",
227 |        "       [101320300, '香港', '香港'],\n",
228 |        "       [101330100, '澳门', '澳门']], dtype=object)"
229 |       ]
230 |      },
231 |      "execution_count": 21,
232 |      "metadata": {},
233 |      "output_type": "execute_result"
234 |     }
235 |    ],
236 |    "source": [
237 |     "city"
238 |    ]
239 |   },
240 |   {
241 |    "cell_type": "code",
242 |    "execution_count": 23,
243 |    "metadata": {},
244 |    "outputs": [
245 |     {
246 |      "data": {
247 |       "text/plain": [
248 |        "'上海'"
249 |       ]
250 |      },
251 |      "execution_count": 23,
252 |      "metadata": {},
253 |      "output_type": "execute_result"
254 |     }
255 |    ],
256 |    "source": [
257 |     "city[1][2]"
258 |    ]
259 |   },
260 |   {
261 |    "cell_type": "code",
262 |    "execution_count": 36,
263 |    "metadata": {
264 |     "collapsed": true
265 |    },
266 |    "outputs": [],
267 |    "source": [
268 |     "# 连接到MongoDB\n",
269 |     "MONGO_URL = 'localhost'\n",
270 |     "MONGO_DB = 'Graduation_project'\n",
271 |     "MONGO_COLLECTION = 'jobs_info'\n",
272 |     "client = pymongo.MongoClient(MONGO_URL, port=27017)\n",
273 |     "db = client[MONGO_DB]\n",
274 |     "\n",
275 |     "\n",
276 |     "# 检查是否已爬过\n",
277 |     "check = pd.DataFrame(list(db[MONGO_COLLECTION].find()))\n",
278 |     "check_list = check[['signal']]\n",
279 |     "grouped = check_list.groupby(check['signal'])"
280 |    ]
281 |   },
282 |   {
283 |    "cell_type": "code",
284 |    "execution_count": 44,
285 |    "metadata": {},
286 |    "outputs": [
287 |     {
288 |      "data": {
289 |       "text/plain": [
290 |        "True"
291 |       ]
292 |      },
293 |      "execution_count": 44,
294 |      "metadata": {},
295 |      "output_type": "execute_result"
296 |     }
297 |    ],
298 |    "source": [
299 |     "'1' in grouped.size().index"
300 |    ]
301 |   }
302 |  ],
303 |  "metadata": {
304 |   "kernelspec": {
305 |    "display_name": "Python 3",
306 |    "language": "python",
307 |    "name": "python3"
308 |   },
309 |   "language_info": {
310 |    "codemirror_mode": {
311 |     "name": "ipython",
312 |     "version": 3
313 |    },
314 |    "file_extension": ".py",
315 |    "mimetype": "text/x-python",
316 |    "name": "python",
317 |    "nbconvert_exporter": "python",
318 |    "pygments_lexer": "ipython3",
319 |    "version": "3.6.3"
320 |   }
321 |  },
322 |  "nbformat": 4,
323 |  "nbformat_minor": 2
324 | }
325 | 


--------------------------------------------------------------------------------
/city_data/city.py:
--------------------------------------------------------------------------------
 1 | # version:1.0
 2 | # author:brandon
 3 | # date:2018/10/20
 4 | 
 5 | # common imports
 6 | import pandas as pd
 7 | 
 8 | 
 9 | def city():
10 |     city_data = pd.read_csv('city.csv')
11 |     data = city_data[['code', '省', '市']]
12 |     # print(data.values)
13 |     city = data.values
14 |     return city


--------------------------------------------------------------------------------
/pics/1.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/1.png


--------------------------------------------------------------------------------
/pics/10.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/10.png


--------------------------------------------------------------------------------
/pics/11.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/11.png


--------------------------------------------------------------------------------
/pics/12.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/12.png


--------------------------------------------------------------------------------
/pics/13.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/13.png


--------------------------------------------------------------------------------
/pics/14.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/14.png


--------------------------------------------------------------------------------
/pics/2.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/2.png


--------------------------------------------------------------------------------
/pics/3.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/3.png


--------------------------------------------------------------------------------
/pics/4.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/4.png


--------------------------------------------------------------------------------
/pics/5.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/5.png


--------------------------------------------------------------------------------
/pics/6.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/6.png


--------------------------------------------------------------------------------
/pics/7.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/7.png


--------------------------------------------------------------------------------
/pics/8.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/8.png


--------------------------------------------------------------------------------
/pics/9.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/9.png


--------------------------------------------------------------------------------
/pics/ML部分/为标签编码.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/为标签编码.png


--------------------------------------------------------------------------------
/pics/ML部分/决策树训练集误差.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/决策树训练集误差.png


--------------------------------------------------------------------------------
/pics/ML部分/划分训练集与测试集.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/划分训练集与测试集.png


--------------------------------------------------------------------------------
/pics/ML部分/变量重要性.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/变量重要性.png


--------------------------------------------------------------------------------
/pics/ML部分/属性合并.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/属性合并.png


--------------------------------------------------------------------------------
/pics/ML部分/筛选岗位数量前150.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/筛选岗位数量前150.png


--------------------------------------------------------------------------------
/pics/ML部分/编码分类结果.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/编码分类结果.png


--------------------------------------------------------------------------------
/pics/ML部分/过滤后的分布.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/过滤后的分布.png


--------------------------------------------------------------------------------
/pics/ML部分/过滤薪资.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/过滤薪资.png


--------------------------------------------------------------------------------
/pics/ML部分/重要性分析.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/重要性分析.png


--------------------------------------------------------------------------------
/pics/ML部分/随机森林训练集误差.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/ML部分/随机森林训练集误差.png


--------------------------------------------------------------------------------
/pics/url.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/brandonchow1997/bosszhipin_spider/8e78ba18dee2f46d297ba3fcd11f5cc731989367/pics/url.png


--------------------------------------------------------------------------------
/spiders/citylist_spider.py:
--------------------------------------------------------------------------------
 1 | # version:1.0
 2 | # author:brandon
 3 | # date:2018/10/20
 4 | 
 5 | # common imports
 6 | import requests
 7 | import pymongo
 8 | # -------------
 9 | 
10 | 
11 | # 页面获取函数
12 | def get():
13 |     header = {
14 |         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) '
15 |                       'Chrome/69.0.3497.12 Safari/537.36 '
16 |     }
17 |     url = 'https://www.zhipin.com/common/data/city.json'
18 |     response = requests.get(url, headers=header)
19 |     return response.json()
20 | 
21 | 
22 | def parse(data):
23 |     cities = data['data']['cityList']
24 |     for provence in cities:
25 |         provence_name = provence['name']
26 |         citylist = provence['subLevelModelList']
27 |         for city in citylist:
28 |             city_code = city['code']
29 |             city_name = city['name']
30 |             city_info = {
31 |                 '省': provence_name,
32 |                 '市': city_name,
33 |                 'code': city_code
34 |             }
35 |             print(city_info)
36 |             save_to_mongo(city_info)
37 | 
38 | 
39 | # 连接到MongoDB
40 | MONGO_URL = 'localhost'
41 | MONGO_DB = 'Graduation_project'
42 | MONGO_COLLECTION = 'city'
43 | client = pymongo.MongoClient(MONGO_URL, port=27017)
44 | db = client[MONGO_DB]
45 | 
46 | 
47 | def save_to_mongo(data):
48 |     # 保存到MongoDB中
49 |     try:
50 |         if db[MONGO_COLLECTION].insert(data):
51 |             print('存储到 MongoDB 成功')
52 |     except Exception:
53 |         print('存储到 MongoDB 失败')
54 | 
55 | 
56 | if __name__ == '__main__':
57 |     city = get()
58 |     parse(city)
59 | 


--------------------------------------------------------------------------------
/spiders/jobs_spider.py:
--------------------------------------------------------------------------------
  1 | # version:1.0
  2 | # author:brandon
  3 | # date:2018/10/20
  4 | 
  5 | # common imports
  6 | import requests
  7 | import re
  8 | from lxml import etree
  9 | import time
 10 | import random
 11 | import pymongo
 12 | import pandas as pd
 13 | # -------------
 14 | # 导入模块
 15 | import city
 16 | 
 17 | 
 18 | # 页面获取函数
 19 | def get_page(page, city_code):
 20 |     header = {
 21 |         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) '
 22 |                       'Chrome/69.0.3497.12 Safari/537.36 '
 23 |     }
 24 |     print('正在爬取第', page, '页')
 25 |     url = 'https://www.zhipin.com/c{code}-p100511/?page={page}&ka=page-{page}'.format(code=city_code, page=page)
 26 |     response = requests.get(url, headers=header)
 27 |     return response.text
 28 | # --------------
 29 | 
 30 | 
 31 | # 页面解析函数
 32 | def parse(html, city, provence, page):
 33 |     # data = json.loads(data)
 34 |     # print(data)
 35 |     # 观察数据结构可得
 36 |     data = etree.HTML(html)
 37 |     # 取工资均值
 38 |     items = data.xpath('//*[@id="main"]/div/div[2]/ul/li')
 39 |     for item in items:
 40 |         job_title = item.xpath('./div/div[1]/h3/a/div[1]/text()')[0]
 41 |         job_salary = item.xpath('./div/div[1]/h3/a/span/text()')[0]
 42 |         job_company = item.xpath('./div/div[2]/div/h3/a/text()')[0]
 43 |         job_experience = item.xpath('./div/div[1]/p/text()[2]')[0]
 44 |         job_degree = item.xpath('./div/div[1]/p/text()[3]')[0]
 45 |         company_scale = item.xpath('./div/div[2]/div/p/text()[3]')[0]
 46 |         # 取薪资均值----------------
 47 |         avg_salary = average(job_salary)
 48 |         # -------------------------
 49 |         signal = city + str(page)
 50 |         print(provence, '|', city, '|', job_title, '|', job_salary, '|', job_company, '|', job_experience, '|', job_degree, '|', company_scale,
 51 |               '|', avg_salary)
 52 |         job = {
 53 |             'signal': signal,
 54 |             '省': provence,
 55 |             '城市': city,
 56 |             '职位名称': job_title,
 57 |             '职位薪资': job_salary,
 58 |             '公司名称': job_company,
 59 |             '工作经验': job_experience,
 60 |             '学历要求': job_degree,
 61 |             '公司规模': company_scale
 62 |         }
 63 |         save_to_mongo(job)
 64 | # ---------------------------------------
 65 | 
 66 | 
 67 | # 均值函数
 68 | def average(job_salary):
 69 |     # 取薪资均值----------------
 70 |     pattern = re.compile('\d+')
 71 |     salary = job_salary
 72 |     try:
 73 |         res = re.findall(pattern, salary)
 74 |         avg_salary = 0
 75 |         sum = 0
 76 |         for i in res:
 77 |             a = int(i)
 78 |             sum = sum + a
 79 |             avg_salary = sum / 2
 80 |     except Exception:
 81 |         avg_salary = 0
 82 |     # 函数返回值
 83 |     return avg_salary
 84 | 
 85 | 
 86 | # 连接到MongoDB
 87 | MONGO_URL = 'localhost'
 88 | MONGO_DB = 'Graduation_project'
 89 | MONGO_COLLECTION = 'jobs_info'
 90 | client = pymongo.MongoClient(MONGO_URL, port=27017)
 91 | db = client[MONGO_DB]
 92 | 
 93 | 
 94 | # 检查是否已爬过
 95 | check = pd.DataFrame(list(db[MONGO_COLLECTION].find()))
 96 | check_list = check[['signal']]
 97 | grouped = check_list.groupby(check['signal'])
 98 | # -----------------
 99 | 
100 | 
101 | def save_to_mongo(data):
102 |     # 保存到MongoDB中
103 |     try:
104 |         if db[MONGO_COLLECTION].insert(data):
105 |             print('存储到 MongoDB 成功')
106 |     except Exception:
107 |         print('存储到 MongoDB 失败')
108 | 
109 | 
110 | def jobspider(city_code, city, provence):
111 |     # 最大爬取页数
112 |     MAX_PAGE = 30
113 |     for i in range(1, MAX_PAGE + 1):
114 |         job_signal = city + str(i)
115 |         # print(job_signal)
116 |         if job_signal in grouped.size().index:
117 |             continue
118 |         else:
119 |             try:
120 |                 html = get_page(i, city_code)
121 |                 # ------------ 解析数据 ---------------
122 |                 parse(html, city, provence, i)
123 |                 print('-' * 100)
124 |                 time.sleep(random.randint(0, 3))
125 |             except Exception:
126 |                 break
127 | 
128 | 
129 | if __name__ == '__main__':
130 |     # 获取市ID
131 |     citylist = city.city()
132 |     for city in citylist:
133 |         city_code = city[0]
134 |         provence = city[1]
135 |         city = city[2]
136 |         # 职位爬虫
137 |         jobspider(city_code, city, provence)
138 |         # -----------------
139 | 


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