第3篇：缺失值.ipynb · 刘利民/轻松掌握Pandas

加入 Gitee

与超过 1200万开发者一起发现、参与优秀开源项目，私有仓库也完全免费：）

克隆/下载

第3篇：缺失值.ipynb 163.00 KB

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 第3篇：缺失值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "在我们拿到一个数据进行统计和分析时，很多时候我们都会遇到这样一个问题：数据缺失，比如学生信息中的某个人的学号和年龄为空，这也是大部分数据分析工作所会遇到的之一。缺失值是数据处理中经常会遇到的数据类型，对于这类数据的该如何处理将是本篇内容将要解决的问题。那么，我们首先会提出一个问题，pandas中有哪些数据是缺失值呢？"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 第1部分：何为缺失值\n",
    "缺失值从字面意思上可以理解为当前数据项的值为空，在Python中，提起空值可以想到None，可能也会想到空字符串`\"\"`,那么这两种是否就是pandas中的缺失值呢？这个疑问暂时保留，这里我们引入三种新的空值，np.nan和pd.NAT和pd.NA。下面我们将生成一个包含这几种空值的数据，看看pandas的缺失值到底是什么？"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "所需模块导入"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "生成一个含有空值的DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex       birth\n",
       "name                                  \n",
       "Tom    18.0    北京     None  2000-02-10\n",
       "Bob    30.0    上海     male  2001-12-12\n",
       "Mary    NaN  <NA>   female        None\n",
       "James  40.0    深圳     male  1978-08-08\n",
       "Andy    NaN            NaN         NaT\n",
       "Alice  30.0    晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index = pd.Series(data=[\"Tom\", \"Bob\", \"Mary\", \"James\", \"Andy\", \"Alice\"], name=\"name\")\n",
    "data = {\n",
    "    \"age\": [18, 30, None, 40, np.nan, 30],\n",
    "    \"city\": [\"北京\", \"上海\", pd.NA, \"深圳\", \"\", \"晋城\"],\n",
    "    \"sex\": [None, \"male\", \"female\", \"male\", np.nan, \"unknown\"],\n",
    "    \"birth\": [\"2000-02-10\", \"2001-12-12\", None, \"1978-08-08\", pd.NaT, \"1988-10-17\"]\n",
    "}\n",
    "user_info = pd.DataFrame(data=data, index=index)\n",
    "user_info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age      float64\n",
       "city      object\n",
       "sex       object\n",
       "birth     object\n",
       "dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们生成了一个包含None、\"\"、np.nan、pd.NAT、pd.NA的DataFrame，从生成的数据打印结果来看，每种数据类型都有自己的表示方式，下面我们查看一下他们中的哪些属于pandas中的缺失值。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 判断缺失值\n",
    "判断缺失值有两种方式\n",
    "- isna和notna，\n",
    "- isnull和notnull\n",
    "\n",
    "这两种方法的效果是一样的，Series和DataFrame均有此方法。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         age   city    sex  birth\n",
       "name                             \n",
       "Tom    False  False   True  False\n",
       "Bob    False  False  False  False\n",
       "Mary    True   True  False   True\n",
       "James  False  False  False  False\n",
       "Andy    True  False   True   True\n",
       "Alice  False  False  False  False"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.isna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         age   city    sex  birth\n",
       "name                             \n",
       "Tom    False  False   True  False\n",
       "Bob    False  False  False  False\n",
       "Mary    True   True  False   True\n",
       "James  False  False  False  False\n",
       "Andy    True  False   True   True\n",
       "Alice  False  False  False  False"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.isnull()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<function pandas.core.dtypes.missing.isna(obj)>,\n",
       " <function pandas.core.dtypes.missing.isna(obj)>)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.isna, pd.isnull"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "从上面打印结果来看，除了空字符串不属于缺失值以外，None、np.nan、pd.NA、pd.NAT均是缺失值。这几种缺失值类型有什么区别呢？"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 缺失值类型"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### None: python对象类型的缺失值\n",
    "pandas可以使用None作为缺失值标签，它是一个python对象，它不能作为Numpy/pandas数组类型的缺失值，只能用于'object'数组类型（即由python对象构成的数组）"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "首先来回顾一下None的运算"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "None == None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# None + None\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# TypeError                                 Traceback (most recent call last)\n",
    "# <ipython-input-31-87684764322b> in <module>\n",
    "# ----> 1 None + None\n",
    "\n",
    "# TypeError: unsupported operand type(s) for +: 'NoneType' and 'NoneType'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None在bool类型的Series中为False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([None], dtype='bool')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None在bool类型的Series中不会改变其类型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True, None], dtype='bool')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None和数字在一起时，None自动转换为np.nan，Series自动转换为float64类型。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([1, None])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "除非手动指定数据类型为O(object)类型，即表示任何Python数据类型。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1    None\n",
       "dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([1, None], dtype='O')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "总结：None表示Python数据类型中的缺失值，即Series类型为object时None才会为None，其他类型时会转换成对应的数据类型，比如bool类型时会转换成False,int类型时会转换成pd.nan。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### NaN：数值类型的缺失值\n",
    "NaN: Not a Number，是在任何系统中都兼容的特殊浮点数"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "np.nan不等于自己"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.nan == np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.nan == 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.nan == None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "np.nan在numpy中的类型为浮点，由此导致数据集读入时，即使原来是整数的列，只要有缺失值就会变为浮点型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "float"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    2\n",
       "2    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([1, 2, 3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    NaN\n",
       "2    3.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([1, np.nan, 3])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "此外，对于包含布尔类型和np.nan的Series，会自动转换为object类型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1      NaN\n",
       "2    False\n",
       "dtype: object"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,np.nan,False])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果是np.nan填充，那么它的值会自动变为True而不是False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1     True\n",
       "2    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,np.nan,False], dtype='bool')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "但当修改一个布尔列表时，会改变列表类型，而不是赋值为True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([True,False],dtype='bool')\n",
    "s[1]=np.nan\n",
    "s"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "总结：np.nan是numpy和pandas中的缺失值表示，其自身数据类型为float,但其与空值不同的是，转为布尔值为True,当Series同时包含bool和np.nan时，转为object类型，当Series同时包含int和np.nan时，转为float64类型，两个np.nan不相等。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### pd.NAT：时间类型的缺失值\n",
    "NaT是针对时间序列的缺失值，是Pandas的内置类型，可以完全看做时序版本的np.nan，与自己不等"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   2021-01-20\n",
       "dtype: datetime64[ns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_time = pd.Series([pd.Timestamp('20210120')])\n",
    "s_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   2021-01-20\n",
       "1          NaT\n",
       "dtype: datetime64[ns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_time[1] = None\n",
    "s_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   2021-01-20\n",
       "1          NaT\n",
       "2          NaT\n",
       "dtype: datetime64[ns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_time[2] = np.nan\n",
    "s_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   2021-01-20\n",
       "1          NaT\n",
       "2          NaT\n",
       "3          NaT\n",
       "dtype: datetime64[ns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_time[3] = pd.NaT\n",
    "s_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pandas._libs.tslibs.nattype.NaTType"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(pd.NaT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.NaT == pd.NaT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1      NaT\n",
       "2    False\n",
       "dtype: object"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,pd.NaT,False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1     True\n",
       "2    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([True,pd.NaT,False], dtype='bool')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Nullable类型与NA符号"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "这是Pandas在1.0新版本中引入的重大改变，其目的就是为了（在若干版本后）解决之前出现的混乱局面，统一缺失值处理方法\n",
    "> \"The goal of pd.NA is provide a “missing” indicator that can be used consistently across data types (instead of np.nan, None or pd.NaT depending on the data type).\"——User Guide for Pandas v-1.0\n",
    "\n",
    "官方鼓励用户使用新的数据类型和缺失类型pd.NA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Nullable整形**  \n",
    "对于该种类型而言，它与原来标记int上的符号区别在于首字母大写：'Int'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_original = pd.Series([1, 2], dtype=\"int64\")\n",
    "s_original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    2\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new = pd.Series([1, 2], dtype=\"Int64\")\n",
    "s_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "它的好处就在于，其中前面提到的三种缺失值都会被替换为统一的NA符号，且不改变数据类型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "s_original[1] = np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1    <NA>\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[1] = np.nan\n",
    "s_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1    <NA>\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[1] = None\n",
    "s_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1    <NA>\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[1] = pd.NaT\n",
    "s_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Nullable布尔**  \n",
    "对于该种类型而言，作用与上面的类似，记号为boolean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_original = pd.Series([1, 0], dtype=\"bool\")\n",
    "s_original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    False\n",
       "1     True\n",
       "dtype: boolean"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new = pd.Series([0, 1], dtype=\"boolean\")\n",
    "s_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    NaN\n",
       "1    0.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_original[0] = np.nan\n",
    "s_original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    False\n",
       "1    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_original = pd.Series([1, 0], dtype=\"bool\") #此处重新加一句是因为前面赋值改变了bool类型\n",
    "s_original[0] = None\n",
    "s_original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1    True\n",
       "dtype: boolean"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[0] = np.nan\n",
    "s_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1    True\n",
       "dtype: boolean"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[0] = None\n",
    "s_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1    True\n",
       "dtype: boolean"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_new[0] = pd.NaT\n",
    "s_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "pd.NA作为索引"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    cat\n",
       "dtype: object"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series(['dog','cat'])\n",
    "s[s_new]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**string类型**  \n",
    "该类型是1.0的一大创新，目的之一就是为了区分开原本含糊不清的object类型，这里将简要地提及string，因为它是第7章的主题内容\n",
    "它本质上也属于Nullable类型，因为并不会因为含有缺失而改变类型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    dog\n",
       "1    cat\n",
       "dtype: string"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series(['dog','cat'],dtype='string')\n",
    "s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1     cat\n",
       "dtype: string"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s[0] = np.nan\n",
    "s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1     cat\n",
       "dtype: string"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s[0] = None\n",
    "s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    <NA>\n",
       "1     cat\n",
       "dtype: string"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s[0] = pd.NaT\n",
    "s"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "此外，和object类型的一点重要区别就在于，在调用字符方法后，string类型返回的是Nullable类型，object则会根据缺失类型和数据类型而改变"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1    <NA>\n",
       "2       0\n",
       "dtype: Int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([\"a\", None, \"b\"], dtype=\"string\")\n",
    "s.str.count('a')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    NaN\n",
       "2    0.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s2 = pd.Series([\"a\", None, \"b\"], dtype=\"object\")\n",
    "s2.str.count(\"a\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    False\n",
       "1     <NA>\n",
       "2    False\n",
       "dtype: boolean"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.str.isdigit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    False\n",
       "1     None\n",
       "2    False\n",
       "dtype: object"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s2.str.isdigit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 缺失值的运算与分组"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1. 加号与乘号规则"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用加法时，缺失值为0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9.0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([2,3,np.nan,4])\n",
    "s.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用乘法时，缺失值为1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24.0"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.prod()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用累计函数时，缺失值自动略过"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    2.0\n",
       "1    5.0\n",
       "2    NaN\n",
       "3    9.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     2.0\n",
       "1     6.0\n",
       "2     NaN\n",
       "3    24.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.cumprod()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         NaN\n",
       "1    0.500000\n",
       "2    0.000000\n",
       "3    0.333333\n",
       "dtype: float64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.pct_change()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.groupby方法中的缺失值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "自动忽略为缺失值的组"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>one</th>\n",
       "      <th>two</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A</td>\n",
       "      <td>0.013885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B</td>\n",
       "      <td>1.736015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C</td>\n",
       "      <td>1.939304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>D</td>\n",
       "      <td>-1.120659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.560868</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   one       two\n",
       "0    A  0.013885\n",
       "1    B  1.736015\n",
       "2    C  1.939304\n",
       "3    D -1.120659\n",
       "4  NaN -0.560868"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_g = pd.DataFrame({'one':['A','B','C','D',np.nan],'two':np.random.randn(5)})\n",
    "df_g"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'A': [0], 'B': [1], 'C': [2], 'D': [3]}"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_g.groupby('one').groups"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 第2部分：缺失值处理"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. 丢弃缺失值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seriese 使用 dropna 比较简单，对于 DataFrame 来说，可以设置更多的参数。\n",
    "> user_info.dropna(axis=0, how='any', thresh=None, subset=None, inplace=False)\n",
    ">- axis 参数用于控制行或列，跟其他不一样的是，axis=0 （默认）表示操作行，axis=1 表示操作列。\n",
    ">- how 参数可选的值为 any（默认） 或者 all。any 表示一行/列有任意元素为空时即丢弃，all 一行/列所有值都为空时才丢弃。\n",
    ">- subset 参数表示删除时只考虑的索引或列名。\n",
    ">- thresh参数的类型为整数，它的作用是，比如 thresh=3，会在一行/列中至少有 3 个非空值时将其保留。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex       birth\n",
       "name                                  \n",
       "Tom    18.0    北京     None  2000-02-10\n",
       "Bob    30.0    上海     male  2001-12-12\n",
       "Mary    NaN  <NA>   female        None\n",
       "James  40.0    深圳     male  1978-08-08\n",
       "Andy    NaN            NaN         NaT\n",
       "Alice  30.0    晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "axis参数: 0为行，1为列"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age city      sex       birth\n",
       "name                                 \n",
       "Bob    30.0   上海     male  2001-12-12\n",
       "James  40.0   深圳     male  1978-08-08\n",
       "Alice  30.0   晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.dropna(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: []\n",
       "Index: [Tom, Bob, Mary, James, Andy, Alice]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.dropna(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "how参数: any或all,默认为any"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age city      sex       birth\n",
       "name                                 \n",
       "Bob    30.0   上海     male  2001-12-12\n",
       "James  40.0   深圳     male  1978-08-08\n",
       "Alice  30.0   晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# any：只有存在至少缺失值就删除该行或该列\n",
    "user_info.dropna(axis=0,how='any')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex       birth\n",
       "name                                  \n",
       "Tom    18.0    北京     None  2000-02-10\n",
       "Bob    30.0    上海     male  2001-12-12\n",
       "Mary    NaN  <NA>   female        None\n",
       "James  40.0    深圳     male  1978-08-08\n",
       "Andy    NaN            NaN         NaT\n",
       "Alice  30.0    晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 一行数据所有字段都为空值才删除\n",
    "user_info.dropna(axis=0,how='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "subset参数：限定列中查找缺失值是否满足条件"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age city      sex       birth\n",
       "name                                 \n",
       "Bob    30.0   上海     male  2001-12-12\n",
       "James  40.0   深圳     male  1978-08-08\n",
       "Alice  30.0   晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 一行数据中只要 city 或 sex 存在空值即删除\n",
    "user_info.dropna(axis=0, how=\"any\", subset=[\"city\", \"sex\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. 填充缺失值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### fillna(): 填充缺失值\n",
    "除了可以丢弃缺失值外，也可以填充缺失值，最常见的是使用 fillna 完成填充。\n",
    "> fillna(value=None,method=None,axis=None,inplace=False,limit=None,downcast=None)\n",
    "\n",
    "fillna 这名字一看就是用来填充缺失值的。\n",
    "填充缺失值时，常见的一种方式是使用一个标量来填充。例如，这里我将有缺失的年龄都填充为 0。\n",
    "\n",
    "除了可以使用标量来填充之外，还可以使用前一个或后一个有效值来填充。\n",
    "设置参数 method='pad' 或 method='ffill' 可以使用前一个有效值来填充。\n",
    "设置参数 method='bfill' 或 method='backfill' 可以使用后一个有效值来填充。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex       birth\n",
       "name                                  \n",
       "Tom    18.0    北京     None  2000-02-10\n",
       "Bob    30.0    上海     male  2001-12-12\n",
       "Mary    NaN  <NA>   female        None\n",
       "James  40.0    深圳     male  1978-08-08\n",
       "Andy    NaN            NaN         NaT\n",
       "Alice  30.0    晋城  unknown  1988-10-17"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary      0.0\n",
       "James    40.0\n",
       "Andy      0.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary     30.0\n",
       "James    40.0\n",
       "Andy     40.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.fillna(method=\"ffill\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary     40.0\n",
       "James    40.0\n",
       "Andy     30.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.fillna(method=\"backfill\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### interpolate()：插值\n",
    "除了通过 fillna 方法来填充缺失值外，还可以通过 interpolate 方法来填充。默认情况下使用线性差值，可以是设置 method 参数来改变方式。　\n",
    "> interpolate(method: str = \"linear\",axis: Axis = 0,limit: Optional[int] = None,inplace: bool_t = False,limit_direction: Optional[str] = None,limit_area: Optional[str] = None, downcast: Optional[str] = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary     35.0\n",
       "James    40.0\n",
       "Andy     35.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.interpolate()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 线性插值\n",
    "（a）索引无关的线性插值\n",
    "默认状态下，interpolate会对缺失的值进行线性插值"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     1.0\n",
       "1    10.0\n",
       "2    15.0\n",
       "3    -5.0\n",
       "4    -2.0\n",
       "5     NaN\n",
       "6     NaN\n",
       "7    28.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([1,10,15,-5,-2,np.nan,np.nan,28])\n",
    "s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     1.0\n",
       "1    10.0\n",
       "2    15.0\n",
       "3    -5.0\n",
       "4    -2.0\n",
       "5     8.0\n",
       "6    18.0\n",
       "7    28.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.interpolate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "s.interpolate().plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "此时的插值与索引无关"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "62      1.0\n",
       "77     10.0\n",
       "96     15.0\n",
       "119    -5.0\n",
       "121    -2.0\n",
       "212     8.0\n",
       "263    18.0\n",
       "265    28.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s.index = np.sort(np.random.randint(50,300,8))\n",
    "s.interpolate()\n",
    "# 值不变"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "s.interpolate().plot()\n",
    "# 后面三个点不是线性的（如果几乎为线性函数，请重新运行上面的一个代码块，这是随机性导致的）"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(b) 与索引有关的插值  \n",
    "method中的index和time选项可以使插值线性地依赖索引，即插值为索引的线性函数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "s.interpolate(method='index').plot()\n",
    "#可以看到与上面的区别"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果索引是时间，那么可以按照时间长短插值，对于时间序列将在后续章节详细介绍"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2020-05-01     0.0\n",
       "2020-05-07     NaN\n",
       "2020-06-03    10.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_t = pd.Series([0,np.nan,10]\n",
    "        ,index=[pd.Timestamp('2020-05-01'),pd.Timestamp('2020-05-07'),pd.Timestamp('2020-06-03')])\n",
    "s_t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "s_t.interpolate().plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "s_t.interpolate(method='time').plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 高级插值方法\n",
    "此处的高级指的是与线性插值相比较，例如样条插值、多项式插值、阿基玛插值等（需要安装Scipy），方法详情请看[这里](https://pandas.pydata.org/pandas-docs/version/1.0.0/reference/api/pandas.DataFrame.interpolate.html#pandas.DataFrame.interpolate)  \n",
    "关于这部分仅给出一个官方的例子，因为插值方法是数值分析的内容，而不是Pandas中的基本知识："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ser = pd.Series(np.arange(1, 10.1, .25) ** 2 + np.random.randn(37))\n",
    "missing = np.array([4, 13, 14, 15, 16, 17, 18, 20, 29])\n",
    "ser[missing] = np.nan\n",
    "methods = ['linear', 'quadratic', 'cubic']\n",
    "df = pd.DataFrame({m: ser.interpolate(method=m) for m in methods})\n",
    "df.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### interpolate中的限制参数"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "（a）limit表示最多插入多少个"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    2.0\n",
       "2    3.0\n",
       "3    NaN\n",
       "4    5.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([1,np.nan,np.nan,np.nan,5])\n",
    "s.interpolate(limit=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "（b）limit_direction表示插值方向，可选forward,backward,both，默认前向"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    1.0\n",
       "2    1.0\n",
       "3    2.0\n",
       "4    3.0\n",
       "5    4.0\n",
       "6    5.0\n",
       "7    NaN\n",
       "8    NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([np.nan,np.nan,1,np.nan,np.nan,np.nan,5,np.nan,np.nan,])\n",
    "s.interpolate(limit_direction='backward')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "（c）limit_area表示插值区域，可选inside,outside，默认None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    NaN\n",
       "1    NaN\n",
       "2    1.0\n",
       "3    2.0\n",
       "4    3.0\n",
       "5    4.0\n",
       "6    5.0\n",
       "7    NaN\n",
       "8    NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([np.nan,np.nan,1,np.nan,np.nan,np.nan,5,np.nan,np.nan,])\n",
    "s.interpolate(limit_area='inside')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    NaN\n",
       "1    NaN\n",
       "2    1.0\n",
       "3    NaN\n",
       "4    NaN\n",
       "5    NaN\n",
       "6    5.0\n",
       "7    5.0\n",
       "8    5.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.Series([np.nan,np.nan,1,np.nan,np.nan,np.nan,5,np.nan,np.nan,])\n",
    "s.interpolate(limit_area='outside')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. 替换缺失值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "例如，在我们的存储的用户信息中，假定我们限定用户都是青年，出现了年龄为 40 的，我们就可以认为这是一个异常值。再比如，我们都知道性别分为男性（male）和女性（female），在记录用户性别的时候，对于未知的用户性别都记为了 “unknown”,很明显，我们也可以认为“unknown”是缺失值。此外，有的时候会出现空白字符串，这些也可以认为是缺失值。对于上面的这种情况，我们可以使用 replace 方法来替换缺失值。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>40.0</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>1978-08-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex      birth\n",
       "name                                 \n",
       "Tom    18.0    北京     None 2000-02-10\n",
       "Bob    30.0    上海     male 2001-12-12\n",
       "Mary    NaN  <NA>   female        NaT\n",
       "James  40.0    深圳     male 1978-08-08\n",
       "Andy    NaN            NaN        NaT\n",
       "Alice  30.0    晋城  unknown 1988-10-17"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index = pd.Series(data=[\"Tom\", \"Bob\", \"Mary\", \"James\", \"Andy\", \"Alice\"], name=\"name\")\n",
    "data = {\n",
    "    \"age\": [18, 30, None, 40, np.nan, 30],\n",
    "    \"city\": [\"北京\", \"上海\", pd.NA, \"深圳\", \" \", \"晋城\"],\n",
    "    \"sex\": [None, \"male\", \"female\", \"male\", np.nan, \"unknown\"],\n",
    "    \"birth\": [\"2000-02-10\", \"2001-12-12\", pd.NaT, \"1978-08-08\", np.nan, \"1988-10-17\"]\n",
    "}\n",
    "user_info = pd.DataFrame(data=data, index=index)\n",
    "user_info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary      NaN\n",
       "James     NaN\n",
       "Andy      NaN\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.replace(40,np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary      NaN\n",
       "James     NaN\n",
       "Andy      NaN\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.replace({40: np.nan})#制定一个映射字典"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>city</th>\n",
       "      <th>sex</th>\n",
       "      <th>birth</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Tom</th>\n",
       "      <td>18.0</td>\n",
       "      <td>北京</td>\n",
       "      <td>None</td>\n",
       "      <td>2000-02-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bob</th>\n",
       "      <td>30.0</td>\n",
       "      <td>上海</td>\n",
       "      <td>male</td>\n",
       "      <td>2001-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mary</th>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>female</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>James</th>\n",
       "      <td>NaN</td>\n",
       "      <td>深圳</td>\n",
       "      <td>male</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andy</th>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alice</th>\n",
       "      <td>30.0</td>\n",
       "      <td>晋城</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1988-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  city      sex      birth\n",
       "name                                 \n",
       "Tom    18.0    北京     None 2000-02-10\n",
       "Bob    30.0    上海     male 2001-12-12\n",
       "Mary    NaN  <NA>   female        NaT\n",
       "James   NaN    深圳     male        NaT\n",
       "Andy    NaN            NaN        NaT\n",
       "Alice  30.0    晋城  unknown 1988-10-17"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 字典多列筛选\n",
    "user_info.replace({\"age\": 40, \"birth\": pd.Timestamp(\"1978-08-08\")}, np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom        None\n",
       "Bob        male\n",
       "Mary     female\n",
       "James      male\n",
       "Andy        NaN\n",
       "Alice       NaN\n",
       "Name: sex, dtype: object"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.sex.replace(\"unknown\", np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom        北京\n",
       "Bob        上海\n",
       "Mary     <NA>\n",
       "James      深圳\n",
       "Andy      NaN\n",
       "Alice      晋城\n",
       "Name: city, dtype: object"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.city.replace(r'\\s+', np.nan, regex=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "除了我们自己手动丢弃、填充已经替换缺失值之外，我们还可以使用其他对象来填充。\n",
    "例如有两个关于用户年龄的 Series，其中一个有缺失值，另一个没有，我们可以将没有的缺失值的 Series 中的元素传给有缺失值的。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary     20.0\n",
       "James    40.0\n",
       "Andy     20.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "age_new = user_info.age.copy()\n",
    "age_new.fillna(20, inplace=True)\n",
    "age_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name\n",
       "Tom      18.0\n",
       "Bob      30.0\n",
       "Mary     20.0\n",
       "James    40.0\n",
       "Andy     20.0\n",
       "Alice    30.0\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_info.age.combine_first(age_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 简单小结"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "由于数据的来源的复杂性、不确定性，数据中难免会存在字段值不全、缺失等情况，本文介绍了在pandas中，None、np.nan、pd.NaT和pd.NA都会被识别为缺失值，其中None特指Python对象数据类型的缺失值，np.nan表示numpy中的缺失值，np.nan的类型为float,pd.NaT特指为时间类型的缺失值，在 Pandas 1.0 以后为了统一对缺失值表示和处理，实验性地使用一个标量 pd.NA 来代表。接下来介绍了pandas中如何判断缺失值以及处理缺失值的几种方法，学会本篇内容之后，相信你已经对缺失值相关的操作技巧和知识有了一定的了解。"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "272px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}

一键复制编辑原始数据按行查看历史

提交于 2021-01-31 23:27 . 更新第3篇：缺失值

马建仓 AI 助手

尝试更多

代码解读

代码找茬

代码优化

Python

https://gitee.com/liulimin/easy-pandas.git

git@gitee.com:liulimin/easy-pandas.git

liulimin

easy-pandas

轻松掌握Pandas

master

刘利民/轻松掌握Pandas

简介

发行版

贡献者

近期动态

刘利民/轻松掌握Pandas .gitee-modal { width: 500px !important; }

简介

发行版

贡献者

近期动态

搜索帮助

刘利民/轻松掌握Pandas