{"id":432,"date":"2021-11-27T21:54:51","date_gmt":"2021-11-27T13:54:51","guid":{"rendered":"http:\/\/101.43.40.65\/?p=432"},"modified":"2021-11-27T22:35:15","modified_gmt":"2021-11-27T14:35:15","slug":"python-pandas","status":"publish","type":"post","link":"https:\/\/zhang.mba\/index.php\/2021\/11\/27\/21\/54\/51\/432\/python-pandas\/python\/zhangzhiqi\/","title":{"rendered":"Python Pandas"},"content":{"rendered":"

Python Pandas \u4ecb\u7ecd<\/h1>\n

Pandas\u662f\u4e00\u4e2a\u5f00\u6e90\u7684Python\u5e93\uff0c\u4f7f\u7528\u5176\u5f3a\u5927\u7684\u6570\u636e\u7ed3\u6784\u63d0\u4f9b\u9ad8\u6027\u80fd\u7684\u6570\u636e\u5904\u7406\u548c\u5206\u6790\u5de5\u5177\u3002 Pandas\u8fd9\u4e2a\u540d\u5b57\u6e90\u81ea\u9762\u677f\u6570\u636e – \u6765\u81ea\u591a\u7ef4\u6570\u636e\u7684\u8ba1\u91cf\u7ecf\u6d4e\u5b66\u3002<\/p>\n

2008\u5e74\uff0c\u5f00\u53d1\u4eba\u5458Wes McKinney\u5728\u9700\u8981\u9ad8\u6027\u80fd\uff0c\u7075\u6d3b\u7684\u6570\u636e\u5206\u6790\u5de5\u5177\u65f6\u5f00\u59cb\u5f00\u53d1Pandas\u3002<\/p>\n

\u5728Pandas\u4e4b\u524d\uff0cPython\u4e3b\u8981\u7528\u4e8e\u6570\u636e\u7ba1\u7406\u548c\u51c6\u5907\u3002 \u5b83\u5bf9\u6570\u636e\u5206\u6790\u7684\u8d21\u732e\u5f88\u5c0f\u3002 Pandas\u89e3\u51b3\u4e86\u8fd9\u4e2a\u95ee\u9898\u3002 \u4f7f\u7528Pandas\uff0c\u65e0\u8bba\u6570\u636e\u6765\u6e90\u5982\u4f55 – \u52a0\u8f7d\uff0c\u51c6\u5907\uff0c\u64cd\u4f5c\uff0c\u5efa\u6a21\u548c\u5206\u6790\uff0c\u6211\u4eec\u90fd\u53ef\u4ee5\u5b8c\u6210\u6570\u636e\u5904\u7406\u548c\u5206\u6790\u4e2d\u7684\u4e94\u4e2a\u5178\u578b\u6b65\u9aa4\u3002<\/p>\n

Python\u4e0ePandas\u4e00\u8d77\u4f7f\u7528\u7684\u9886\u57df\u5e7f\u6cdb\uff0c\u5305\u62ec\u5b66\u672f\u548c\u5546\u4e1a\u9886\u57df\uff0c\u5305\u62ec\u91d1\u878d\uff0c\u7ecf\u6d4e\u5b66\uff0c\u7edf\u8ba1\u5b66\uff0c\u5206\u6790\u5b66\u7b49\u3002<\/p>\n

Pandas\u7684\u4e3b\u8981\u7279\u70b9<\/p>\n

\u4f7f\u7528\u9ed8\u8ba4\u548c\u81ea\u5b9a\u4e49\u7d22\u5f15\u7684\u5feb\u901f\u9ad8\u6548\u7684DataFrame\u5bf9\u8c61\u3002
\n\u7528\u4e8e\u5c06\u6570\u636e\u4ece\u4e0d\u540c\u6587\u4ef6\u683c\u5f0f\u52a0\u8f7d\u5230\u5185\u5b58\u6570\u636e\u5bf9\u8c61\u7684\u5de5\u5177\u3002
\n\u6570\u636e\u5bf9\u9f50\u548c\u7f3a\u5931\u6570\u636e\u7684\u96c6\u6210\u5904\u7406\u3002
\n\u91cd\u65b0\u8bbe\u7f6e\u548c\u65cb\u8f6c\u65e5\u671f\u96c6\u3002
\n\u5927\u6570\u636e\u96c6\u7684\u57fa\u4e8e\u6807\u7b7e\u7684\u5206\u7247\uff0c\u7d22\u5f15\u548c\u5b50\u96c6\u3002
\n\u6570\u636e\u7ed3\u6784\u4e2d\u7684\u5217\u53ef\u4ee5\u88ab\u5220\u9664\u6216\u63d2\u5165\u3002
\n\u6309\u6570\u636e\u5206\u7ec4\u8fdb\u884c\u805a\u5408\u548c\u8f6c\u6362\u3002
\n\u9ad8\u6027\u80fd\u7684\u6570\u636e\u5408\u5e76\u548c\u8fde\u63a5\u3002
\n\u65f6\u95f4\u5e8f\u5217\u529f\u80fd\u3002<\/p>\n

Python Pandas \u73af\u5883\u8bbe\u7f6e<\/h1>\n

\u6807\u51c6\u7684Python\u53d1\u884c\u7248\u4e0d\u4f1a\u4e0ePandas\u6a21\u5757\u6346\u7ed1\u5728\u4e00\u8d77\u3002 \u4e00\u79cd\u8f7b\u91cf\u7ea7\u7684\u9009\u62e9\u662f\u4f7f\u7528\u6d41\u884c\u7684Python\u5305\u5b89\u88c5\u7a0b\u5e8fpip\u6765\u5b89\u88c5NumPy\u3002<\/p>\n

pip install pandas\r\n<\/pre>\n
\u5982\u679c\u60a8\u5b89\u88c5\u4e86Anaconda Python\u8f6f\u4ef6\u5305\uff0cPandas\u5c06\u9ed8\u8ba4\u5b89\u88c5\u4ee5\u4e0b\u8f6f\u4ef6 –<\/p>\n
Windows<\/h3>\n
    \n
  • Anaconda\uff08\u6765\u81eahttps:\/\/www.continuum.io\uff09\u662fSciPy\u5806\u6808\u7684\u514d\u8d39Python\u53d1\u884c\u7248\u3002 \u5b83\u4e5f\u9002\u7528\u4e8eLinux\u548cMac\u3002<\/li>\n
  • Canopy\uff08https:\/\/www.enthought.com\/products\/canopy\/\uff09\u514d\u8d39\u63d0\u4f9b\uff0c\u5e76\u63d0\u4f9b\u5b8c\u6574\u7684\u7528\u4e8eWindows\uff0cLinux\u548cMac\u7684SciPy\u5806\u6808\u3002<\/li>\n
  • Python\uff08x\uff0cy\uff09\u662f\u4e00\u4e2a\u514d\u8d39\u7684Python\u53d1\u884c\u7248\uff0c\u5e26\u6709\u7528\u4e8eWindows\u64cd\u4f5c\u7cfb\u7edf\u7684SciPy\u5806\u6808\u548cSpyder IDE\u3002 \uff08\u53ef\u4ecehttp:\/\/python-xy.github.io\/\u4e0b\u8f7d\uff09<\/li>\n<\/ul>\n
    Linux<\/h3>\n
    \u5404\u4e2aLinux\u53d1\u884c\u7248\u7684\u8f6f\u4ef6\u5305\u7ba1\u7406\u5668\u7528\u4e8e\u5728SciPy\u5806\u6808\u4e2d\u5b89\u88c5\u4e00\u4e2a\u6216\u591a\u4e2a\u8f6f\u4ef6\u5305\u3002<\/p>\n
    Ubuntu<\/p>\n
    sudo apt-get install python-numpy python-scipy python-matplotlibipythonipythonnotebook\r\npython-pandas python-sympy python-nose\r\n<\/pre>\n
    Fedora<\/p>\n
    sudo yum install numpyscipy python-matplotlibipython python-pandas sympy\r\npython-nose atlas-devel<\/pre>\n
    Python Pandas \u6570\u636e\u7ed3\u6784<\/h1>\n
    Pandas\u5904\u7406\u4ee5\u4e0b\u4e09\u79cd\u6570\u636e\u7ed3\u6784 –<\/p>\n
      \n
    • Series<\/li>\n
    • DataFrame<\/li>\n
    • Panel<\/li>\n<\/ul>\n
      \u8fd9\u4e9b\u6570\u636e\u7ed3\u6784\u5efa\u7acb\u5728Numpy\u6570\u7ec4\u7684\u9876\u90e8\uff0c\u8fd9\u610f\u5473\u7740\u5b83\u4eec\u5f88\u5feb\u3002<\/p>\n
      \u7ef4\u5ea6\u548c\u8bf4\u660e<\/h3>\n
      \u8003\u8651\u8fd9\u4e9b\u6570\u636e\u7ed3\u6784\u7684\u6700\u4f73\u65b9\u5f0f\u662f\u9ad8\u7ef4\u6570\u636e\u7ed3\u6784\u662f\u5176\u8f83\u4f4e\u7ef4\u6570\u636e\u7ed3\u6784\u7684\u5bb9\u5668\u3002 \u4f8b\u5982\uff0cDataFrame\u662fSeries\u7684\u5bb9\u5668\uff0cPanel\u662fDataFrame\u7684\u5bb9\u5668\u3002<\/p>\n\n\n\n\n\n\n
      \u6570\u636e\u7ed3\u6784<\/th>\n\u7ef4\u5ea6<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
      Series<\/td>\n1<\/td>\n1\u7ef4\uff0c\u7c7b\u4f3c\u4e8e\u5b9a\u957f\u7684\u6709\u5e8f\u5b57\u5178\uff0c\u6709Index\u548cvalue\u3002<\/td>\n<\/tr>\n
      Data Frames<\/td>\n2<\/td>\n2\u7ef4, \u53ef\u4ee5\u770b\u6210\u662f\u4ee5Series\u7ec4\u6210\u7684\u5b57\u5178\uff0c\u5177\u6709\u884c\u7d22\u5f15\u548c\u5217\u7d22\u5f15\u3002<\/td>\n<\/tr>\n
      Panel<\/td>\n3<\/td>\n3\u7ef4, Panel\u662f3D\u5bb9\u5668\u7684\u6570\u636e<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
      \u6784\u5efa\u548c\u5904\u7406\u4e24\u4e2a\u6216\u66f4\u591a\u7ef4\u6570\u7ec4\u662f\u4e00\u9879\u5355\u8c03\u4e4f\u5473\u7684\u4efb\u52a1\uff0c\u7528\u6237\u5728\u7f16\u5199\u51fd\u6570\u65f6\u9700\u8981\u8003\u8651\u6570\u636e\u96c6\u7684\u65b9\u5411\u3002 \u4f46\u662f\u4f7f\u7528Pandas\u6570\u636e\u7ed3\u6784\uff0c\u7528\u6237\u7684\u5fc3\u7406\u52aa\u529b\u4f1a\u51cf\u5c11\u3002<\/p>\n
      \u4f8b\u5982\uff0c\u4f7f\u7528\u8868\u683c\u6570\u636e\uff08DataFrame\uff09\u65f6\uff0c\u601d\u8003\u7d22\u5f15\uff08\u884c\uff09\u548c\u5217\u800c\u4e0d\u662f\u8f740\u548c\u8f741\u65f6\u8bed\u4e49\u4e0a\u66f4\u6709\u7528\u3002<\/p>\n
      \u53ef\u53d8\u6027(Mutablility)<\/p>\n
      \u6240\u6709Pandas\u6570\u636e\u7ed3\u6784\u90fd\u662f\u53ef\u53d8\u7684\uff08\u53ef\u4ee5\u66f4\u6539\uff09\uff0c\u9664\u4e86Series\u90fd\u662f\u53ef\u53d8\u5927\u5c0f\u7684\u3002 \u7cfb\u5217\u5927\u5c0f\u4e0d\u53ef\u53d8\u3002
      \n\u6ce8 – DataFrame\u88ab\u5e7f\u6cdb\u4f7f\u7528\u5e76\u4e14\u662f\u6700\u91cd\u8981\u7684\u6570\u636e\u7ed3\u6784\u4e4b\u4e00\u3002 \u9762\u677f\u7528\u5f97\u5c11\u5f97\u591a\u3002<\/p>\n
      Series<\/h3>\n
      Series\u662f\u4e00\u79cd\u5177\u6709\u540c\u8d28\u6570\u636e\u7ed3\u6784\u7684\u4e00\u7ef4\u6570\u7ec4\u3002 \u4f8b\u5982\uff0c\u4ee5\u4e0b\u7cfb\u5217\u662f\u6574\u657010,23,56 …\u7684\u96c6\u5408<\/p>\n\n\n\n
      10<\/td>\n23<\/td>\n56<\/td>\n17<\/td>\n52<\/td>\n61<\/td>\n73<\/td>\n90<\/td>\n26<\/td>\n72<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
      \u5173\u952e\u70b9<\/p>\n
        \n
      • \u540c\u8d28\u6570\u636e<\/li>\n
      • \u5927\u5c0f\u4e0d\u53ef\u53d8<\/li>\n
      • \u6570\u636e\u503c\u53ef\u53d8<\/li>\n<\/ul>\n
        DataFrame<\/h3>\n
        DataFrame\u662f\u4e00\u4e2a\u5177\u6709\u5f02\u6784\u6570\u636e\u7684\u4e8c\u7ef4\u6570\u7ec4\u3002 \u4f8b\u5982\uff0c<\/p>\n\n\n\n\n\n\n\n
        \u59d3\u540d<\/th>\n\u5e74\u9f84<\/th>\n\u6027\u522b<\/th>\n\u8bc4\u5206<\/th>\n<\/tr>\n
        Steve<\/td>\n32<\/td>\nMale<\/td>\n3.45<\/td>\n<\/tr>\n
        Lia<\/td>\n28<\/td>\nFemale<\/td>\n4.6<\/td>\n<\/tr>\n
        Vin<\/td>\n45<\/td>\nMale<\/td>\n3.9<\/td>\n<\/tr>\n
        Katie<\/td>\n38<\/td>\nFemale<\/td>\n2.78<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
        \u8be5\u8868\u683c\u8868\u793a\u4e00\u4e2a\u7ec4\u7ec7\u7684\u9500\u552e\u56e2\u961f\u7684\u603b\u4f53\u7ee9\u6548\u8bc4\u7ea7\u6570\u636e\u3002 \u6570\u636e\u4ee5\u884c\u548c\u5217\u8868\u793a\u3002 \u6bcf\u5217\u4ee3\u8868\u4e00\u4e2a\u5c5e\u6027\uff0c\u6bcf\u884c\u4ee3\u8868\u4e00\u4e2a\u4eba\u3002<\/p>\n
        \u5217\u7684\u6570\u636e\u7c7b\u578b<\/h3>\n
        \u56db\u5217\u7684\u6570\u636e\u7c7b\u578b\u5982\u4e0b –<\/p>\n\n\n\n\n\n\n\n
        \u5217<\/th>\n\u6570\u636e\u7c7b\u578b<\/th>\n<\/tr>\n
        \u59d3\u540d<\/td>\nString<\/td>\n<\/tr>\n
        \u5e74\u9f84<\/td>\nInteger<\/td>\n<\/tr>\n
        \u6027\u522b<\/td>\nString<\/td>\n<\/tr>\n
        \u8bc4\u5206<\/td>\nFloat<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
        \u5173\u952e\u70b9<\/p>\n
          \n
        • \u5f02\u6784\u6570\u636e<\/li>\n
        • \u5927\u5c0f\u53ef\u53d8<\/li>\n
        • \u6570\u636e\u53ef\u53d8<\/li>\n<\/ul>\n
          Panel<\/h3>\n
          Panel\u662f\u5177\u6709\u5f02\u6784\u6570\u636e\u7684\u4e09\u7ef4\u6570\u636e\u7ed3\u6784\u3002 \u56fe\u5f62\u8868\u793a\u5f88\u96be\u8868\u793aPanel\u3002 \u4f46\u662fPanel\u53ef\u4ee5\u4f5c\u4e3aDataFrame\u7684\u5bb9\u5668\u6765\u8bf4\u660e\u3002<\/p>\n
          \u5173\u952e\u70b9<\/p>\n
            \n
          • \u5f02\u6784\u6570\u636e<\/li>\n
          • \u5927\u5c0f\u53ef\u53d8<\/li>\n
          • \u6570\u636e\u53ef\u53d8<\/li>\n<\/ul>\n
            Python Pandas Series<\/h1>\n
            Series\u662f\u4e00\u7ef4\u6807\u7b7e\u6570\u7ec4\uff0c\u80fd\u591f\u5bb9\u7eb3\u4efb\u4f55\u7c7b\u578b\u7684\u6570\u636e\uff08\u6574\u6570\uff0c\u5b57\u7b26\u4e32\uff0c\u6d6e\u70b9\u6570\uff0cpython\u5bf9\u8c61\u7b49\uff09\u3002 \u8f74\u6807\u7b7e\u7edf\u79f0\u4e3a\u7d22\u5f15\u3002<\/p>\n
            pandas.Series<\/h3>\n
            Panas Series\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u6784\u9020\u51fd\u6570\u521b\u5efa –<\/p>\n
            pandas.Series( data, index, dtype, copy)\r\n<\/pre>\n
            \u6784\u9020\u51fd\u6570\u7684\u53c2\u6570\u5982\u4e0b\u6240\u793a –<\/p>\n\n\n\n\n\n\n\n
            \u5e8f\u53f7<\/th>\n\u53c2\u6570 & \u63cf\u8ff0<\/th>\n<\/tr>\n
            1<\/td>\ndata<\/b>
            \n\u6570\u636e\u91c7\u7528\u5404\u79cd\u5f62\u5f0f\uff0c\u5982ndarray\uff0c\u5217\u8868\uff0c\u5e38\u91cf<\/td>\n<\/tr>\n
            2<\/td>\nindex<\/b>
            \n\u7d22\u5f15\u503c\u5fc5\u987b\u662f\u552f\u4e00\u4e14\u53ef\u6563\u5217\u7684\uff0c\u4e0e\u6570\u636e\u957f\u5ea6\u76f8\u540c\u3002 \u5982\u679c\u6ca1\u6709\u7d22\u5f15\u88ab\u4f20\u9012\uff0c\u5219\u9ed8\u8ba4np.arange\uff08n\uff09\u3002<\/td>\n<\/tr>\n
            3<\/td>\ndtype<\/b>
            \ndtype\u7528\u4e8e\u6570\u636e\u7c7b\u578b\u3002 \u5982\u679c\u6ca1\u6709\uff0c\u5219\u4f1a\u63a8\u65ad\u6570\u636e\u7c7b\u578b<\/td>\n<\/tr>\n
            4<\/td>\ncopy<\/b>
            \n\u590d\u5236\u6570\u636e\u3002 \u9ed8\u8ba4\u4e3aFalse<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
            Series\u53ef\u4ee5\u4f7f\u7528\u5404\u79cd\u8f93\u5165\u521b\u5efa\uff0c\u5982 –<\/p>\n
              \n
            • \u6570\u7ec4<\/li>\n
            • \u5b57\u5178<\/li>\n
            • \u6807\u91cf\u503c\u6216\u5e38\u91cf<\/li>\n<\/ul>\n
              \u521b\u5efa\u4e00\u4e2a\u7a7aSeries<\/h3>\n
              \u4e00\u4e2a\u57fa\u672c\u7684Series\uff0c\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a\u7a7aSeries\u3002<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\ns = pd.Series()\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              Series([], dtype: float64)\r\n<\/pre>\n
              \u4ecendarray\u521b\u5efaSeries<\/h3>\n
              \u5982\u679c\u6570\u636e\u662fndarray\uff0c\u5219\u4f20\u9012\u7684\u7d22\u5f15\u5fc5\u987b\u5177\u6709\u76f8\u540c\u7684\u957f\u5ea6\u3002 \u5982\u679c\u6ca1\u6709\u7d22\u5f15\u88ab\u4f20\u9012\uff0c\u5219\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u7d22\u5f15\u5c06\u662frange\uff08n\uff09\uff0c\u5176\u4e2dn\u662f\u6570\u7ec4\u957f\u5ea6\uff0c\u5373[0,1,2,3\u2026. range(len(array))-1]\u3002<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = np.array(['a','b','c','d'])\r\ns = pd.Series(data)\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              0   a\r\n1   b\r\n2   c\r\n3   d\r\ndtype: object\r\n<\/pre>\n
              \u6211\u4eec\u6ca1\u6709\u4f20\u9012\u4efb\u4f55\u7d22\u5f15\uff0c\u56e0\u6b64\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u7d22\u5f15\u7684\u7d22\u5f15\u8303\u56f4\u662f0\u5230len\uff08data\uff09-1\uff0c\u53730\u52303\u3002<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = np.array(['a','b','c','d'])\r\ns = pd.Series(data,index=[100,101,102,103])\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              100  a\r\n101  b\r\n102  c\r\n103  d\r\ndtype: object\r\n<\/pre>\n
              \u6211\u4eec\u5728\u8fd9\u91cc\u901a\u8fc7\u4e86\u7d22\u5f15\u503c\u3002 \u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u5728\u8f93\u51fa\u4e2d\u770b\u5230\u81ea\u5b9a\u4e49\u7684\u7d22\u5f15\u503c\u3002<\/p>\n
              \u4ece\u5b57\u5178\u521b\u5efaSeries<\/h3>\n
              \u4e00\u4e2a\u5b57\u5178\u53ef\u4ee5\u4f5c\u4e3a\u8f93\u5165\u4f20\u9012\uff0c\u5982\u679c\u6ca1\u6709\u6307\u5b9a\u7d22\u5f15\uff0c\u90a3\u4e48\u5b57\u5178\u952e\u5c06\u6309\u7167\u6392\u5e8f\u7684\u987a\u5e8f\u8fdb\u884c\u6784\u5efa\u7d22\u5f15\u3002 \u5982\u679c\u7d22\u5f15\u88ab\u4f20\u9012\uff0c\u7d22\u5f15\u4e2d\u7684\u6807\u7b7e\u5bf9\u5e94\u7684\u6570\u636e\u503c\u5c06\u88ab\u53d6\u51fa\u3002<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = {'a' : 0., 'b' : 1., 'c' : 2.}\r\ns = pd.Series(data)\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              a 0.0\r\nb 1.0\r\nc 2.0\r\ndtype: float64\r\n<\/pre>\n
              \u5b57\u5178\u952e\u7528\u4e8e\u6784\u5efa\u7d22\u5f15\u3002<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = {'a' : 0., 'b' : 1., 'c' : 2.}\r\ns = pd.Series(data,index=['b','c','d','a'])\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              b 1.0\r\nc 2.0\r\nd NaN\r\na 0.0\r\ndtype: float64\r\n<\/pre>\n
              \u7d22\u5f15\u987a\u5e8f\u662f\u6301\u4e45\u7684\uff0c\u7f3a\u5c11\u7684\u5143\u7d20\u7528NaN\uff08\u4e0d\u662f\u6570\u5b57\uff09\u586b\u5145\u3002<\/p>\n
              \n
              \u4ece\u6807\u91cf\u521b\u5efa\u4e00\u4e2aSeries<\/h3>\n
              \u5982\u679c\u6570\u636e\u662f\u6807\u91cf\u503c\uff0c\u5219\u5fc5\u987b\u63d0\u4f9b\u7d22\u5f15\u3002 \u8be5\u503c\u5c06\u88ab\u91cd\u590d\u4ee5\u5339\u914d\u7d22\u5f15\u7684\u957f\u5ea6<\/p>\n
              #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\ns = pd.Series(5, index=[0, 1, 2, 3])\r\nprint s\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              0  5\r\n1  5\r\n2  5\r\n3  5\r\ndtype: int64\r\n<\/pre>\n
              \u4eceSeries\u4f4d\u7f6e\u8bbf\u95ee\u6570\u636e<\/h3>\n
              Series\u4e2d\u7684\u6570\u636e\u53ef\u4ee5\u50cfndarray\u4e2d\u90a3\u6837\u8bbf\u95ee\u3002<\/p>\n
              \u68c0\u7d22\u7b2c\u4e00\u4e2a\u5143\u7d20\u3002 \u6b63\u5982\u6211\u4eec\u5df2\u7ecf\u77e5\u9053\u7684\u90a3\u6837\uff0c\u6570\u7ec4\u4ece\u96f6\u5f00\u59cb\u8ba1\u6570\uff0c\u8fd9\u610f\u5473\u7740\u7b2c\u4e00\u4e2a\u5143\u7d20\u5b58\u50a8\u5728\u7b2c\u96f6\u4e2a\u4f4d\u7f6e\uff0c\u4f9d\u6b64\u7c7b\u63a8\u3002<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve the first element\r\nprint s[0]\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              1\r\n<\/pre>\n
              \u68c0\u7d22\u7cfb\u5217\u4e2d\u7684\u524d\u4e09\u4e2a\u5143\u7d20\u3002 \u5982\u679c\u5c06a\uff1a\u63d2\u5165\u5176\u524d\u9762\uff0c\u5219\u5c06\u4ece\u8be5\u7d22\u5f15\u5f00\u59cb\u63d0\u53d6\u6240\u6709\u9879\u76ee\u3002 \u5982\u679c\u4f7f\u7528\u4e24\u4e2a\u53c2\u6570\uff08\u5728\u5b83\u4eec\u4e4b\u95f4\uff09\uff0c\u5219\u4e24\u4e2a\u7d22\u5f15\u4e4b\u95f4\u7684\u9879\u76ee\uff08\u4e0d\u5305\u62ec\u505c\u6b62\u7d22\u5f15\uff09<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve the first three element\r\nprint s[:3]\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              a  1\r\nb  2\r\nc  3\r\ndtype: int64\r\n<\/pre>\n
              \u68c0\u7d22\u6700\u540e\u4e09\u4e2a\u5143\u7d20\u3002<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve the last three element\r\nprint s[-3:]\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              c  3\r\nd  4\r\ne  5\r\ndtype: int64\r\n<\/pre>\n
              \u4f7f\u7528\u6807\u7b7e\u68c0\u7d22\u6570\u636e\uff08\u7d22\u5f15\uff09<\/h3>\n
              Series\u5c31\u50cf\u4e00\u4e2a\u56fa\u5b9a\u5927\u5c0f\u7684\u5b57\u5178\uff0c\u60a8\u53ef\u4ee5\u901a\u8fc7\u7d22\u5f15\u6807\u7b7e\u83b7\u53d6\u548c\u8bbe\u7f6e\u503c\u3002<\/p>\n
              \u4f7f\u7528\u7d22\u5f15\u6807\u7b7e\u503c\u68c0\u7d22\u5355\u4e2a\u5143\u7d20\u3002<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve a single element\r\nprint s['a']\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              1\r\n<\/pre>\n
              \u4f7f\u7528\u7d22\u5f15\u6807\u7b7e\u503c\u5217\u8868\u68c0\u7d22\u591a\u4e2a\u5143\u7d20\u3002<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve multiple elements\r\nprint s[['a','c','d']]\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              a  1\r\nc  3\r\nd  4\r\ndtype: int64\r\n<\/pre>\n
              \u5982\u679c\u4e0d\u5305\u542b\u6807\u7b7e\uff0c\u5219\u4f1a\u5f15\u53d1\u5f02\u5e38\u3002<\/p>\n
              import pandas as pd\r\ns = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])\r\n\r\n#retrieve multiple elements\r\nprint s['f']\r\n<\/pre>\n
              \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
              \u2026\r\nKeyError: 'f'<\/pre>\n
              Python Pandas DataFrame<\/h1>\n
              \u6570\u636e\u5e27(DataFrame)\u662f\u4e8c\u7ef4\u6570\u636e\u7ed3\u6784\uff0c\u5373\u6570\u636e\u4ee5\u884c\u548c\u5217\u7684\u8868\u683c\u65b9\u5f0f\u6392\u5217\u3002<\/p>\n
              \u6570\u636e\u5e27(DataFrame)\u7684\u529f\u80fd\u7279\u70b9\uff1a<\/p>\n
                \n
              • \u6f5c\u5728\u7684\u5217\u662f\u4e0d\u540c\u7684\u7c7b\u578b<\/li>\n
              • \u5927\u5c0f\u53ef\u53d8<\/li>\n
              • \u6807\u8bb0\u8f74(\u884c\u548c\u5217)<\/li>\n
              • \u53ef\u4ee5\u5bf9\u884c\u548c\u5217\u6267\u884c\u7b97\u672f\u8fd0\u7b97<\/li>\n<\/ul>\n
                \u7ed3\u6784\u4f53<\/strong><\/p>\n
                \u5047\u8bbe\u8981\u521b\u5efa\u4e00\u4e2a\u5305\u542b\u5b66\u751f\u6570\u636e\u7684\u6570\u636e\u5e27\u3002\u53c2\u8003\u4ee5\u4e0b\u56fe\u793a –<\/p>\n
                \"\"<\/div><\/p>\n
                \u53ef\u4ee5\u5c06\u4e0a\u56fe\u8868\u89c6\u4e3aSQL\u8868\u6216\u7535\u5b50\u8868\u683c\u6570\u636e\u8868\u793a\u3002<\/p>\n
                pandas.DataFrame<\/h3>\n
                pandas\u4e2d\u7684DataFrame\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u6784\u9020\u51fd\u6570\u521b\u5efa –<\/p>\n
                pandas.DataFrame( data, index, columns, dtype, copy)\r\n<\/pre>\n
                \u6784\u9020\u51fd\u6570\u7684\u53c2\u6570\u5982\u4e0b –<\/p>\n\n\n\n\n\n\n\n\n
                \u7f16\u53f7<\/th>\n\u53c2\u6570<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                1<\/td>\ndata<\/td>\n\u6570\u636e\u91c7\u53d6\u5404\u79cd\u5f62\u5f0f\uff0c\u5982:ndarray\uff0cseries\uff0cmap\uff0clists\uff0cdict\uff0cconstant\u548c\u53e6\u4e00\u4e2aDataFrame\u3002<\/td>\n<\/tr>\n
                2<\/td>\nindex<\/td>\n\u5bf9\u4e8e\u884c\u6807\u7b7e\uff0c\u8981\u7528\u4e8e\u7ed3\u679c\u5e27\u7684\u7d22\u5f15\u662f\u53ef\u9009\u7f3a\u7701\u503cnp.arrange(n)\uff0c\u5982\u679c\u6ca1\u6709\u4f20\u9012\u7d22\u5f15\u503c\u3002<\/td>\n<\/tr>\n
                3<\/td>\ncolumns<\/td>\n\u5bf9\u4e8e\u5217\u6807\u7b7e\uff0c\u53ef\u9009\u7684\u9ed8\u8ba4\u8bed\u6cd5\u662f – np.arange(n)\u3002 \u8fd9\u53ea\u6709\u5728\u6ca1\u6709\u7d22\u5f15\u4f20\u9012\u7684\u60c5\u51b5\u4e0b\u624d\u662f\u8fd9\u6837\u3002<\/td>\n<\/tr>\n
                4<\/td>\ndtype<\/td>\n\u6bcf\u5217\u7684\u6570\u636e\u7c7b\u578b\u3002<\/td>\n<\/tr>\n
                5<\/td>\ncopy<\/td>\n\u5982\u679c\u9ed8\u8ba4\u503c\u4e3aFalse\uff0c\u5219\u6b64\u547d\u4ee4(\u6216\u4efb\u4f55\u5b83)\u7528\u4e8e\u590d\u5236\u6570\u636e\u3002<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                \u521b\u5efaDataFrame<\/h3>\n
                Pandas\u6570\u636e\u5e27(DataFrame)\u53ef\u4ee5\u4f7f\u7528\u5404\u79cd\u8f93\u5165\u521b\u5efa\uff0c\u5982 –<\/p>\n
                  \n
                • \u5217\u8868<\/li>\n
                • \u5b57\u5178<\/li>\n
                • \u7cfb\u5217<\/li>\n
                • Numpy ndarrays<\/li>\n
                • \u53e6\u4e00\u4e2a\u6570\u636e\u5e27(DataFrame)<\/li>\n<\/ul>\n
                  \u5728\u672c\u7ae0\u7684\u540e\u7eed\u7ae0\u8282\u4e2d\uff0c\u6211\u4eec\u5c06\u770b\u5230\u5982\u4f55\u4f7f\u7528\u8fd9\u4e9b\u8f93\u5165\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  \u521b\u5efa\u4e00\u4e2a\u7a7a\u7684DataFrame<\/h3>\n
                  \u521b\u5efa\u57fa\u672c\u6570\u636e\u5e27\u662f\u7a7a\u6570\u636e\u5e27\u3002
                  \n\u793a\u4f8b<\/strong><\/p>\n
                  #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\ndf = pd.DataFrame()\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  Empty DataFrame\r\nColumns: []\r\nIndex: []\r\n<\/pre>\n
                  \u4ece\u5217\u8868\u521b\u5efaDataFrame<\/h3>\n
                  \u53ef\u4ee5\u4f7f\u7528\u5355\u4e2a\u5217\u8868\u6216\u5217\u8868\u5217\u8868\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  \u5b9e\u4f8b-1<\/strong><\/p>\n
                  import pandas as pd\r\ndata = [1,2,3,4,5]\r\ndf = pd.DataFrame(data)\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                       0\r\n0    1\r\n1    2\r\n2    3\r\n3    4\r\n4    5\r\n<\/pre>\n
                  \u5b9e\u4f8b-2<\/strong><\/p>\n
                  import pandas as pd\r\ndata = [['Alex',10],['Bob',12],['Clarke',13]]\r\ndf = pd.DataFrame(data,columns=['Name','Age'])\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Name      Age\r\n0     Alex      10\r\n1     Bob       12\r\n2     Clarke    13\r\n<\/pre>\n
                  \u5b9e\u4f8b-3<\/strong><\/p>\n
                  import pandas as pd\r\ndata = [['Alex',10],['Bob',12],['Clarke',13]]\r\ndf = pd.DataFrame(data,columns=['Name','Age'],dtype=float)\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Name     Age\r\n0     Alex     10.0\r\n1     Bob      12.0\r\n2     Clarke   13.0\r\n<\/pre>\n
                  \u6ce8\u610f – \u53ef\u4ee5\u89c2\u5bdf\u5230\uff0cdtype\u53c2\u6570\u5c06Age\u5217\u7684\u7c7b\u578b\u66f4\u6539\u4e3a\u6d6e\u70b9\u3002<\/p>\n
                  \u4ecendarrays\/Lists\u7684\u5b57\u5178\u6765\u521b\u5efaDataFrame<\/h3>\n
                  \u6240\u6709\u7684ndarrays\u5fc5\u987b\u5177\u6709\u76f8\u540c\u7684\u957f\u5ea6\u3002\u5982\u679c\u4f20\u9012\u4e86\u7d22\u5f15(index)\uff0c\u5219\u7d22\u5f15\u7684\u957f\u5ea6\u5e94\u7b49\u4e8e\u6570\u7ec4\u7684\u957f\u5ea6\u3002<\/p>\n
                  \u5982\u679c\u6ca1\u6709\u4f20\u9012\u7d22\u5f15\uff0c\u5219\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u7d22\u5f15\u5c06\u4e3arange(n)\uff0c\u5176\u4e2dn\u4e3a\u6570\u7ec4\u957f\u5ea6\u3002<\/p>\n
                  \u5b9e\u4f8b-1<\/strong><\/p>\n
                  import pandas as pd\r\ndata = {'Name':['Tom', 'Jack', 'Steve', 'Ricky'],'Age':[28,34,29,42]}\r\ndf = pd.DataFrame(data)\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Age      Name\r\n0     28        Tom\r\n1     34       Jack\r\n2     29      Steve\r\n3     42      Ricky\r\n<\/pre>\n
                  \u6ce8 – \u89c2\u5bdf\u503c0,1,2,3\u3002\u5b83\u4eec\u662f\u5206\u914d\u7ed9\u6bcf\u4e2a\u4f7f\u7528\u51fd\u6570range(n)\u7684\u9ed8\u8ba4\u7d22\u5f15\u3002<\/p>\n
                  \u793a\u4f8b-2<\/strong><\/p>\n
                  \u4f7f\u7528\u6570\u7ec4\u521b\u5efa\u4e00\u4e2a\u7d22\u5f15\u7684\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  import pandas as pd\r\ndata = {'Name':['Tom', 'Jack', 'Steve', 'Ricky'],'Age':[28,34,29,42]}\r\ndf = pd.DataFrame(data, index=['rank1','rank2','rank3','rank4'])\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                           Age    Name\r\nrank1    28      Tom\r\nrank2    34     Jack\r\nrank3    29    Steve\r\nrank4    42    Ricky\r\n<\/pre>\n
                  \u6ce8\u610f – index\u53c2\u6570\u4e3a\u6bcf\u884c\u5206\u914d\u4e00\u4e2a\u7d22\u5f15\u3002<\/p>\n
                  \u4ece\u5217\u8868\u521b\u5efa\u6570\u636e\u5e27DataFrame<\/h3>\n
                  \u5b57\u5178\u5217\u8868\u53ef\u4f5c\u4e3a\u8f93\u5165\u6570\u636e\u4f20\u9012\u4ee5\u7528\u6765\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\uff0c\u5b57\u5178\u952e\u9ed8\u8ba4\u4e3a\u5217\u540d\u3002<\/p>\n
                  \u5b9e\u4f8b-1<\/strong><\/p>\n
                  \u4ee5\u4e0b\u793a\u4f8b\u663e\u793a\u5982\u4f55\u901a\u8fc7\u4f20\u9012\u5b57\u5178\u5217\u8868\u6765\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  import pandas as pd\r\ndata = [{'a': 1, 'b': 2},{'a': 5, 'b': 10, 'c': 20}]\r\ndf = pd.DataFrame(data)\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                      a    b      c\r\n0   1   2     NaN\r\n1   5   10   20.0\r\n<\/pre>\n
                  \u6ce8\u610f – \u89c2\u5bdf\u5230\uff0cNaN(\u4e0d\u662f\u6570\u5b57)\u88ab\u9644\u52a0\u5728\u7f3a\u5931\u7684\u533a\u57df\u3002<\/p>\n
                  \u793a\u4f8b-2<\/strong><\/p>\n
                  \u4ee5\u4e0b\u793a\u4f8b\u663e\u793a\u5982\u4f55\u901a\u8fc7\u4f20\u9012\u5b57\u5178\u5217\u8868\u548c\u884c\u7d22\u5f15\u6765\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  import pandas as pd\r\ndata = [{'a': 1, 'b': 2},{'a': 5, 'b': 10, 'c': 20}]\r\ndf = pd.DataFrame(data, index=['first', 'second'])\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                          a   b       c\r\nfirst   1   2     NaN\r\nsecond  5   10   20.0\r\n<\/pre>\n
                  \u5b9e\u4f8b-3<\/strong><\/p>\n
                  \u4ee5\u4e0b\u793a\u4f8b\u663e\u793a\u5982\u4f55\u4f7f\u7528\u5b57\u5178\uff0c\u884c\u7d22\u5f15\u548c\u5217\u7d22\u5f15\u5217\u8868\u521b\u5efa\u6570\u636e\u5e27(DataFrame)\u3002<\/p>\n
                  import pandas as pd\r\ndata = [{'a': 1, 'b': 2},{'a': 5, 'b': 10, 'c': 20}]\r\n\r\n#With two column indices, values same as dictionary keys\r\ndf1 = pd.DataFrame(data, index=['first', 'second'], columns=['a', 'b'])\r\n\r\n#With two column indices with one index with other name\r\ndf2 = pd.DataFrame(data, index=['first', 'second'], columns=['a', 'b1'])\r\nprint df1\r\nprint df2\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  #df1 output\r\n\t\t a  b\r\nfirst    1  2\r\nsecond   5  10\r\n\t\r\n#df2 output\r\n\t\t a  b1\r\nfirst    1  NaN\r\nsecond   5  NaN\r\n<\/pre>\n
                  \u6ce8\u610f – \u89c2\u5bdf\uff0cdf2\u4f7f\u7528\u5b57\u5178\u952e\u4ee5\u5916\u7684\u5217\u7d22\u5f15\u521b\u5efaDataFrame; \u56e0\u6b64\uff0c\u9644\u52a0\u4e86NaN\u5230\u4f4d\u7f6e\u4e0a\u3002 \u800cdf1\u662f\u4f7f\u7528\u5217\u7d22\u5f15\u521b\u5efa\u7684\uff0c\u4e0e\u5b57\u5178\u952e\u76f8\u540c\uff0c\u6240\u4ee5\u4e5f\u9644\u52a0\u4e86NaN\u3002<\/p>\n
                  \u4ece\u7cfb\u5217\u7684\u5b57\u5178\u6765\u521b\u5efaDataFrame<\/h3>\n
                  \u5b57\u5178\u7684\u7cfb\u5217\u53ef\u4ee5\u4f20\u9012\u4ee5\u5f62\u6210\u4e00\u4e2aDataFrame\u3002 \u6240\u5f97\u5230\u7684\u7d22\u5f15\u662f\u901a\u8fc7\u7684\u6240\u6709\u7cfb\u5217\u7d22\u5f15\u7684\u5e76\u96c6\u3002<\/p>\n
                  \u793a\u4f8b<\/strong><\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint df`\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        one    two\r\na     1.0    1\r\nb     2.0    2\r\nc     3.0    3\r\nd     NaN    4\r\n<\/pre>\n
                  \u6ce8\u610f – \u5bf9\u4e8e\u7b2c\u4e00\u4e2a\u7cfb\u5217\uff0c\u89c2\u5bdf\u5230\u6ca1\u6709\u4f20\u9012\u6807\u7b7e’d’\uff0c\u4f46\u5728\u7ed3\u679c\u4e2d\uff0c\u5bf9\u4e8ed\u6807\u7b7e\uff0c\u9644\u52a0\u4e86NaN\u3002<\/p>\n
                  \u73b0\u5728\u901a\u8fc7\u5b9e\u4f8b\u6765\u4e86\u89e3\u5217\u9009\u62e9\uff0c\u6dfb\u52a0\u548c\u5220\u9664\u3002<\/p>\n
                  \u5217\u9009\u62e9<\/h4>\n
                  \u4e0b\u9762\u5c06\u901a\u8fc7\u4ece\u6570\u636e\u5e27(DataFrame)\u4e2d\u9009\u62e9\u4e00\u5217\u3002<\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint df ['one']\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  a     1.0\r\nb     2.0\r\nc     3.0\r\nd     NaN\r\nName: one, dtype: float64\r\n<\/pre>\n
                  \u5217\u6dfb\u52a0<\/h4>\n
                  \u4e0b\u9762\u5c06\u901a\u8fc7\u5411\u73b0\u6709\u6570\u636e\u6846\u6dfb\u52a0\u4e00\u4e2a\u65b0\u5217\u6765\u7406\u89e3\u8fd9\u4e00\u70b9\u3002<\/p>\n
                  \u793a\u4f8b<\/strong><\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n\t'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\t\r\ndf = pd.DataFrame(d)\r\n\r\n# Adding a new column to an existing DataFrame object with column label by passing new series\r\n\r\nprint (\"Adding a new column by passing as Series:\")\r\ndf['three']=pd.Series([10,20,30],index=['a','b','c'])\r\nprint df\r\n\r\nprint (\"Adding a new column using the existing columns in DataFrame:\")\r\ndf['four']=df['one']+df['three']\r\n\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  Adding a new column by passing as Series:\r\none   two   three\r\na    1.0    1    10.0\r\nb    2.0    2    20.0\r\nc    3.0    3    30.0\r\nd    NaN    4    NaN\r\n\t\r\nAdding a new column using the existing columns in DataFrame:\r\none   two   three    four\r\na     1.0    1    10.0     11.0\r\nb     2.0    2    20.0     22.0\r\nc     3.0    3    30.0     33.0\r\nd     NaN    4     NaN     NaN\r\n<\/pre>\n
                  \u5217\u5220\u9664<\/h4>\n
                  \u5217\u53ef\u4ee5\u5220\u9664\u6216\u5f39\u51fa; \u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u4e86\u89e3\u4e00\u4e0b\u3002<\/p>\n
                  \u4f8b\u5b50<\/strong><\/p>\n
                  # Using the previous DataFrame, we will delete a column\r\n# using del function\r\nimport pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd']),\r\n'three' : pd.Series([10,20,30], index=['a','b','c'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint (\"Our dataframe is:\")\r\nprint df\r\n\r\n# using del function\r\nprint (\"Deleting the first column using DEL function:\")\r\ndel df['one']\r\nprint df\r\n\r\n# using pop function\r\nprint (\"Deleting another column using POP function:\")\r\ndf.pop('two')\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  Our dataframe is:\r\none   three  two\r\na     1.0    10.0   1\r\nb     2.0    20.0   2\r\nc     3.0    30.0   3\r\nd     NaN     NaN   4\r\n\r\nDeleting the first column using DEL function:\r\nthree    two\r\na     10.0     1\r\nb     20.0     2\r\nc     30.0     3\r\nd     NaN      4\r\n\r\nDeleting another column using POP function:\r\nthree\r\na  10.0\r\nb  20.0\r\nc  30.0\r\nd  NaN\r\n<\/pre>\n
                  \u884c\u9009\u62e9\uff0c\u6dfb\u52a0\u548c\u5220\u9664<\/h4>\n
                  \u73b0\u5728\u5c06\u901a\u8fc7\u4e0b\u9762\u5b9e\u4f8b\u6765\u4e86\u89e3\u884c\u9009\u62e9\uff0c\u6dfb\u52a0\u548c\u5220\u9664\u3002\u6211\u4eec\u4ece\u9009\u62e9\u7684\u6982\u5ff5\u5f00\u59cb\u3002<\/p>\n
                  \u6807\u7b7e\u9009\u62e9<\/strong><\/p>\n
                  \u53ef\u4ee5\u901a\u8fc7\u5c06\u884c\u6807\u7b7e\u4f20\u9012\u7ed9loc()\u51fd\u6570\u6765\u9009\u62e9\u884c\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint df.loc['b']\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  one 2.0\r\ntwo 2.0\r\nName: b, dtype: float64\r\n<\/pre>\n
                  \u7ed3\u679c\u662f\u4e00\u7cfb\u5217\u6807\u7b7e\u4f5c\u4e3aDataFrame\u7684\u5217\u540d\u79f0\u3002 \u800c\u4e14\uff0c\u7cfb\u5217\u7684\u540d\u79f0\u662f\u68c0\u7d22\u7684\u6807\u7b7e\u3002<\/p>\n
                  \u6309\u6574\u6570\u4f4d\u7f6e\u9009\u62e9<\/strong><\/p>\n
                  \u53ef\u4ee5\u901a\u8fc7\u5c06\u6574\u6570\u4f4d\u7f6e\u4f20\u9012\u7ed9iloc()\u51fd\u6570\u6765\u9009\u62e9\u884c\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint df.iloc[2]\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                  one   3.0\r\ntwo   3.0\r\nName: c, dtype: float64\r\n<\/pre>\n
                  \u884c\u5207\u7247<\/h4>\n
                  \u53ef\u4ee5\u4f7f\u7528:\u8fd0\u7b97\u7b26\u9009\u62e9\u591a\u884c\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                  import pandas as pd\r\n\r\nd = {'one' : pd.Series([1, 2, 3], index=['a', 'b', 'c']),\r\n'two' : pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])}\r\n\r\ndf = pd.DataFrame(d)\r\nprint df[2:4]\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        one    two\r\nc     3.0     3\r\nd     NaN     4\r\n<\/pre>\n
                  \u9644\u52a0\u884c<\/strong><\/p>\n
                  \u4f7f\u7528append()\u51fd\u6570\u5c06\u65b0\u884c\u6dfb\u52a0\u5230DataFrame\u3002 \u6b64\u529f\u80fd\u5c06\u9644\u52a0\u884c\u7ed3\u675f\u3002<\/p>\n
                  import pandas as pd\r\n\r\ndf = pd.DataFrame([[1, 2], [3, 4]], columns = ['a','b'])\r\ndf2 = pd.DataFrame([[5, 6], [7, 8]], columns = ['a','b'])\r\n\r\ndf = df.append(df2)\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                     a  b\r\n0  1  2\r\n1  3  4\r\n0  5  6\r\n1  7  8\r\n<\/pre>\n
                  \u5220\u9664\u884c<\/strong><\/p>\n
                  \u4f7f\u7528\u7d22\u5f15\u6807\u7b7e\u4eceDataFrame\u4e2d\u5220\u9664\u6216\u5220\u9664\u884c\u3002 \u5982\u679c\u6807\u7b7e\u91cd\u590d\uff0c\u5219\u4f1a\u5220\u9664\u591a\u884c\u3002<\/p>\n
                  \u5982\u679c\u6709\u6ce8\u610f\uff0c\u5728\u4e0a\u8ff0\u793a\u4f8b\u4e2d\uff0c\u6709\u6807\u7b7e\u662f\u91cd\u590d\u7684\u3002\u8fd9\u91cc\u518d\u591a\u653e\u4e00\u4e2a\u6807\u7b7e\uff0c\u770b\u770b\u6709\u591a\u5c11\u884c\u88ab\u5220\u9664\u3002<\/p>\n
                  import pandas as pd\r\n\r\ndf = pd.DataFrame([[1, 2], [3, 4]], columns = ['a','b'])\r\ndf2 = pd.DataFrame([[5, 6], [7, 8]], columns = ['a','b'])\r\n\r\ndf = df.append(df2)\r\n\r\n# Drop rows with label 0\r\ndf = df.drop(0)\r\n\r\nprint df\r\n<\/pre>\n
                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                    a b\r\n1 3 4\r\n1 7 8\r\n<\/pre>\n
                  \u5728\u4e0a\u9762\u7684\u4f8b\u5b50\u4e2d\uff0c\u4e00\u5171\u6709\u4e24\u884c\u88ab\u5220\u9664\uff0c\u56e0\u4e3a\u8fd9\u4e24\u884c\u5305\u542b\u76f8\u540c\u7684\u6807\u7b7e0\u3002<\/p>\n
                  Python Pandas Panel<\/h1>\n
                  Panel\u662f3D\u5bb9\u5668\u7684\u6570\u636e\u3002Panel\u6570\u636e\u4e00\u8bcd\u6765\u6e90\u4e8e\u8ba1\u91cf\u7ecf\u6d4e\u5b66\uff0c\u90e8\u5206\u6e90\u4e8e\u540d\u79f0\uff1aPandas – pan(el)-da(ta)-s\u3002<\/p>\n
                  3\u8f74(axis)\u8fd9\u4e2a\u540d\u79f0\u65e8\u5728\u7ed9\u51fa\u63cf\u8ff0\u6d89\u53caPanel\u6570\u636e\u7684\u64cd\u4f5c\u7684\u4e00\u4e9b\u8bed\u4e49\u3002\u5b83\u4eec\u662f –<\/p>\n
                    \n
                  • items – axis 0\uff0c\u6bcf\u4e2a\u9879\u76ee\u5bf9\u5e94\u4e8e\u5185\u90e8\u5305\u542b\u7684\u6570\u636e\u5e27(DataFrame)\u3002<\/li>\n
                  • major_axis – axis 1\uff0c\u5b83\u662f\u6bcf\u4e2a\u6570\u636e\u5e27(DataFrame)\u7684\u7d22\u5f15(\u884c)\u3002<\/li>\n
                  • minor_axis – axis 2\uff0c\u5b83\u662f\u6bcf\u4e2a\u6570\u636e\u5e27(DataFrame)\u7684\u5217\u3002<\/li>\n<\/ul>\n
                    pandas.Panel()<\/h3>\n
                    \u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u6784\u9020\u51fd\u6570\u521b\u5efaPanel –<\/p>\n
                    pandas.Panel(data, items, major_axis, minor_axis, dtype, copy)\r\n<\/pre>\n
                    \u6784\u9020\u51fd\u6570\u7684\u53c2\u6570\u5982\u4e0b –<\/p>\n\n\n\n\n\n\n\n\n\n
                    \u53c2\u6570<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                    data<\/td>\n\u6570\u636e\u91c7\u53d6\u5404\u79cd\u5f62\u5f0f\uff0c\u5982\uff1andarray\uff0cseries\uff0cmap\uff0clists\uff0cdict\uff0cconstant\u548c\u53e6\u4e00\u4e2a\u6570\u636e\u5e27(DataFrame)<\/td>\n<\/tr>\n
                    items<\/td>\naxis=0<\/td>\n<\/tr>\n
                    major_axis<\/td>\naxis=1<\/td>\n<\/tr>\n
                    minor_axis<\/td>\naxis=2<\/td>\n<\/tr>\n
                    dtype<\/td>\n\u6bcf\u5217\u7684\u6570\u636e\u7c7b\u578b<\/td>\n<\/tr>\n
                    copy<\/td>\n\u590d\u5236\u6570\u636e\uff0c\u9ed8\u8ba4 – false<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                    \u521b\u5efaPanel<\/h3>\n
                    \u53ef\u4ee5\u4f7f\u7528\u591a\u79cd\u65b9\u5f0f\u521b\u5efaPanel –<\/p>\n
                      \n
                    • \u4ecendarrays\u521b\u5efa<\/li>\n
                    • \u4eceDataFrames\u7684dict\u521b\u5efa<\/li>\n<\/ul>\n
                      \u4ece3D ndarray\u521b\u5efa<\/h4>\n
                      # creating an empty panel\r\nimport pandas as pd\r\nimport numpy as np\r\n\r\ndata = np.random.rand(2,4,5)\r\np = pd.Panel(data)\r\nprint p\r\n<\/pre>\n
                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                      <class 'pandas.core.panel.Panel'>\r\nDimensions: 2 (items) x 4 (major_axis) x 5 (minor_axis)\r\nItems axis: 0 to 1\r\nMajor_axis axis: 0 to 3\r\nMinor_axis axis: 0 to 4\r\n<\/pre>\n
                      \u6ce8\u610f – \u89c2\u5bdf\u7a7a\u9762\u677f\u548c\u4e0a\u9762\u677f\u7684\u5c3a\u5bf8\u5927\u5c0f\uff0c\u6240\u6709\u5bf9\u8c61\u90fd\u4e0d\u540c\u3002<\/p>\n
                      \u4eceDataFrame\u5bf9\u8c61\u7684dict\u521b\u5efa\u9762\u677f<\/h4>\n
                      #creating an empty panel\r\nimport pandas as pd\r\nimport numpy as np\r\n\r\ndata = {'Item1' : pd.DataFrame(np.random.randn(4, 3)),\r\n'Item2' : pd.DataFrame(np.random.randn(4, 2))}\r\np = pd.Panel(data)\r\nprint p\r\n<\/pre>\n
                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                      <class 'pandas.core.panel.Panel'>\r\nDimensions: 2 (items) x 4 (major_axis) x 5 (minor_axis)\r\nItems axis: 0 to 1\r\nMajor_axis axis: 0 to 3\r\nMinor_axis axis: 0 to 4\r\n<\/pre>\n
                      \u521b\u5efa\u4e00\u4e2a\u7a7a\u9762\u677f<\/h4>\n
                      \u53ef\u4ee5\u4f7f\u7528Panel\u7684\u6784\u9020\u51fd\u6570\u521b\u5efa\u4e00\u4e2a\u7a7a\u9762\u677f\uff0c\u5982\u4e0b\u6240\u793a\uff1a<\/p>\n
                      #creating an empty panel\r\nimport pandas as pd\r\np = pd.Panel()\r\nprint p\r\n<\/pre>\n
                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                      <class 'pandas.core.panel.Panel'>\r\nDimensions: 0 (items) x 0 (major_axis) x 0 (minor_axis)\r\nItems axis: None\r\nMajor_axis axis: None\r\nMinor_axis axis: None\r\n<\/pre>\n
                      \u4ece\u9762\u677f\u4e2d\u9009\u62e9\u6570\u636e<\/h3>\n
                      \u8981\u4ece\u9762\u677f\u4e2d\u9009\u62e9\u6570\u636e\uff0c\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u65b9\u5f0f –<\/p>\n
                        \n
                      • Items<\/li>\n
                      • Major_axis<\/li>\n
                      • Minor_axis<\/li>\n<\/ul>\n
                        \u4f7f\u7528Items<\/strong><\/p>\n
                        # creating an empty panel\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = {'Item1' : pd.DataFrame(np.random.randn(4, 3)),\r\n'Item2' : pd.DataFrame(np.random.randn(4, 2))}\r\np = pd.Panel(data)\r\nprint p['Item1']\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                    0          1          2\r\n0    0.488224  -0.128637   0.930817\r\n1    0.417497   0.896681   0.576657\r\n2   -2.775266   0.571668   0.290082\r\n3   -0.400538  -0.144234   1.110535\r\n<\/pre>\n
                        \u4e0a\u9762\u793a\u4f8b\u6709\u4e24\u4e2a\u6570\u636e\u9879\uff0c\u8fd9\u91cc\u53ea\u68c0\u7d22item1\u3002\u7ed3\u679c\u662f\u5177\u67094\u884c\u548c3\u5217\u7684\u6570\u636e\u5e27(DataFrame)\uff0c\u5b83\u4eec\u662fMajor_axis\u548cMinor_axis\u7ef4\u3002<\/p>\n
                        \u4f7f\u7528major_axis<\/strong><\/p>\n
                        \u53ef\u4ee5\u4f7f\u7528panel.major_axis(index)\u65b9\u6cd5\u8bbf\u95ee\u6570\u636e\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        # creating an empty panel\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = {'Item1' : pd.DataFrame(np.random.randn(4, 3)),\r\n'Item2' : pd.DataFrame(np.random.randn(4, 2))}\r\np = pd.Panel(data)\r\nprint p.major_xs(1)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                   Item1       Item2\r\n\t0   0.417497    0.748412\r\n\t1   0.896681   -0.557322\r\n\t2   0.576657       NaN\r\n<\/pre>\n
                        \u4f7f\u7528minor_axis<\/strong><\/p>\n
                        \u53ef\u4ee5\u4f7f\u7528panel.minor_axis(index)\u65b9\u6cd5\u8bbf\u95ee\u6570\u636e\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        # creating an empty panel\r\nimport pandas as pd\r\nimport numpy as np\r\ndata = {'Item1' : pd.DataFrame(np.random.randn(4, 3)),\r\n'Item2' : pd.DataFrame(np.random.randn(4, 2))}\r\np = pd.Panel(data)\r\nprint p.minor_xs(1)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                    Item1       Item2\r\n\t0   -0.128637   -1.047032\r\n\t1    0.896681   -0.557322\r\n\t2    0.571668    0.431953\r\n\t3   -0.144234    1.302466\r\n<\/pre>\n
                        \u6ce8\u610f – \u89c2\u5bdf\u5c3a\u5bf8\u5927\u4e0d\u7684\u53d8\u5316\u3002<\/p>\n
                        Python Pandas\u57fa\u672c\u529f\u80fd<\/h1>\n
                        \u5230\u76ee\u524d\u4e3a\u6b62\uff0c\u6211\u4eec\u4e86\u89e3\u4e86\u4e09\u79cdPandas\u6570\u636e\u7ed3\u6784\u4ee5\u53ca\u5982\u4f55\u521b\u5efa\u5b83\u4eec\u3002\u63a5\u4e0b\u6765\u5c06\u4e3b\u8981\u5173\u6ce8\u6570\u636e\u5e27(DataFrame)\u5bf9\u8c61\uff0c\u56e0\u4e3a\u5b83\u5728\u5b9e\u65f6\u6570\u636e\u5904\u7406\u4e2d\u975e\u5e38\u91cd\u8981\uff0c\u5e76\u4e14\u8fd8\u8ba8\u8bba\u5176\u4ed6\u6570\u636e\u7ed3\u6784\u3002<\/p>\n
                        \u7cfb\u5217\u57fa\u672c\u529f\u80fd<\/h3>\n\n\n\n\n\n\n\n\n\n\n\n
                        \u7f16\u53f7<\/th>\n\u5c5e\u6027\u6216\u65b9\u6cd5<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                        1<\/td>\naxes<\/td>\n\u8fd4\u56de\u884c\u8f74\u6807\u7b7e\u5217\u8868\u3002<\/td>\n<\/tr>\n
                        2<\/td>\ndtype<\/td>\n\u8fd4\u56de\u5bf9\u8c61\u7684\u6570\u636e\u7c7b\u578b(dtype)\u3002<\/td>\n<\/tr>\n
                        3<\/td>\nempty<\/td>\n\u5982\u679c\u7cfb\u5217\u4e3a\u7a7a\uff0c\u5219\u8fd4\u56deTrue\u3002<\/td>\n<\/tr>\n
                        4<\/td>\nndim<\/td>\n\u8fd4\u56de\u5e95\u5c42\u6570\u636e\u7684\u7ef4\u6570\uff0c\u9ed8\u8ba4\u5b9a\u4e49\uff1a1\u3002<\/td>\n<\/tr>\n
                        5<\/td>\nsize<\/td>\n\u8fd4\u56de\u57fa\u7840\u6570\u636e\u4e2d\u7684\u5143\u7d20\u6570\u3002<\/td>\n<\/tr>\n
                        6<\/td>\nvalues<\/td>\n\u5c06\u7cfb\u5217\u4f5c\u4e3andarray\u8fd4\u56de\u3002<\/td>\n<\/tr>\n
                        7<\/td>\nhead()<\/td>\n\u8fd4\u56de\u524dn\u884c\u3002<\/td>\n<\/tr>\n
                        8<\/td>\ntail()<\/td>\n\u8fd4\u56de\u6700\u540en\u884c\u3002<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                        \u73b0\u5728\u521b\u5efa\u4e00\u4e2a\u7cfb\u5217\u5e76\u6f14\u793a\u5982\u4f55\u4f7f\u7528\u4e0a\u9762\u6240\u6709\u5217\u51fa\u7684\u5c5e\u6027\u64cd\u4f5c\u3002<\/p>\n
                        \u793a\u4f8b<\/strong><\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 100 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint s\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u8f93\u51fa\u7ed3\u679c –<\/p>\n
                        0   0.967853\r\n1  -0.148368\r\n2  -1.395906\r\n3  -1.758394\r\ndtype: float64\r\n<\/pre>\n
                        axes\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u7cfb\u5217\u7684\u6807\u7b7e\u5217\u8868\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 100 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint (\"The axes are:\")\r\nprint s.axes\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u8f93\u51fa\u7ed3\u679c –<\/p>\n
                        The axes are:\r\n[RangeIndex(start=0, stop=4, step=1)]\r\n<\/pre>\n
                        \u4e0a\u8ff0\u7ed3\u679c\u662f\u4ece0\u52305\u7684\u503c\u5217\u8868\u7684\u7d27\u51d1\u683c\u5f0f\uff0c\u5373\uff1a[0,1,2,3,4]\u3002<\/p>\n
                        empty\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u5e03\u5c14\u503c\uff0c\u8868\u793a\u5bf9\u8c61\u662f\u5426\u4e3a\u7a7a\u3002\u8fd4\u56deTrue\u5219\u8868\u793a\u5bf9\u8c61\u4e3a\u7a7a\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 100 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint (\"Is the Object empty?\")\r\nprint s.empty\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u8f93\u51fa\u7ed3\u679c –<\/p>\n
                        Is the Object empty?\r\nFalse\r\n<\/pre>\n
                        ndim\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u5bf9\u8c61\u7684\u7ef4\u6570\u3002\u6839\u636e\u5b9a\u4e49\uff0c\u4e00\u4e2a\u7cfb\u5217\u662f\u4e00\u4e2a1D\u6570\u636e\u7ed3\u6784\uff0c\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 4 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint s\r\n\r\nprint (\"The dimensions of the object:\")\r\nprint s.ndim\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        0   0.175898\r\n1   0.166197\r\n2  -0.609712\r\n3  -1.377000\r\ndtype: float64\r\n\r\nThe dimensions of the object:\r\n1\r\n<\/pre>\n
                        size\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u7cfb\u5217\u7684\u5927\u5c0f(\u957f\u5ea6)\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 4 random numbers\r\ns = pd.Series(np.random.randn(2))\r\nprint s\r\nprint (\"The size of the object:\")\r\nprint s.size\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        0   3.078058\r\n1  -1.207803\r\ndtype: float64\r\n\r\nThe size of the object:\r\n2\r\n<\/pre>\n
                        values\u793a\u4f8b<\/strong><\/p>\n
                        \u4ee5\u6570\u7ec4\u5f62\u5f0f\u8fd4\u56de\u7cfb\u5217\u4e2d\u7684\u5b9e\u9645\u6570\u636e\u503c\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 4 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint s\r\n\r\nprint (\"The actual data series is:\")\r\nprint s.values\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        0   1.787373\r\n1  -0.605159\r\n2   0.180477\r\n3  -0.140922\r\ndtype: float64\r\n\r\nThe actual data series is:\r\n[ 1.78737302 -0.60515881 0.18047664 -0.1409218 ]\r\n<\/pre>\n
                        head()\u548ctail()\u65b9\u6cd5\u793a\u4f8b<\/strong><\/p>\n
                        \u8981\u67e5\u770bSeries\u6216DataFrame\u5bf9\u8c61\u7684\u5c0f\u6837\u672c\uff0c\u8bf7\u4f7f\u7528head()\u548ctail()\u65b9\u6cd5\u3002<\/p>\n
                        head()\u8fd4\u56de\u524dn\u884c(\u89c2\u5bdf\u7d22\u5f15\u503c)\u3002\u8981\u663e\u793a\u7684\u5143\u7d20\u7684\u9ed8\u8ba4\u6570\u91cf\u4e3a5\uff0c\u4f46\u53ef\u4ee5\u4f20\u9012\u81ea\u5b9a\u4e49\u8fd9\u4e2a\u6570\u5b57\u503c\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 4 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint (\"The original series is:\")\r\nprint s\r\n\r\nprint (\"The first two rows of the data series:\")\r\nprint s.head(2)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        The original series is:\r\n0   0.720876\r\n1  -0.765898\r\n2   0.479221\r\n3  -0.139547\r\ndtype: float64\r\n\r\nThe first two rows of the data series:\r\n0   0.720876\r\n1  -0.765898\r\ndtype: float64\r\n<\/pre>\n
                        tail()\u8fd4\u56de\u6700\u540en\u884c(\u89c2\u5bdf\u7d22\u5f15\u503c)\u3002 \u8981\u663e\u793a\u7684\u5143\u7d20\u7684\u9ed8\u8ba4\u6570\u91cf\u4e3a5\uff0c\u4f46\u53ef\u4ee5\u4f20\u9012\u81ea\u5b9a\u4e49\u6570\u5b57\u503c\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a series with 4 random numbers\r\ns = pd.Series(np.random.randn(4))\r\nprint (\"The original series is:\")\r\nprint s\r\n\r\nprint (\"The last two rows of the data series:\")\r\nprint s.tail(2)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        The original series is:\r\n0 -0.655091\r\n1 -0.881407\r\n2 -0.608592\r\n3 -2.341413\r\ndtype: float64\r\n\r\nThe last two rows of the data series:\r\n2 -0.608592\r\n3 -2.341413\r\ndtype: float64\r\n<\/pre>\n
                        DataFrame\u57fa\u672c\u529f\u80fd<\/h3>\n
                        \u4e0b\u9762\u6765\u770b\u770b\u6570\u636e\u5e27(DataFrame)\u7684\u57fa\u672c\u529f\u80fd\u6709\u54ea\u4e9b\uff1f\u4e0b\u8868\u5217\u51fa\u4e86DataFrame\u57fa\u672c\u529f\u80fd\u7684\u91cd\u8981\u5c5e\u6027\u6216\u65b9\u6cd5\u3002<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
                        \u7f16\u53f7<\/th>\n\u5c5e\u6027\u6216\u65b9\u6cd5<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                        1<\/td>\nT<\/td>\n\u8f6c\u7f6e\u884c\u548c\u5217\u3002<\/td>\n<\/tr>\n
                        2<\/td>\naxes<\/td>\n\u8fd4\u56de\u4e00\u4e2a\u5217\uff0c\u884c\u8f74\u6807\u7b7e\u548c\u5217\u8f74\u6807\u7b7e\u4f5c\u4e3a\u552f\u4e00\u7684\u6210\u5458\u3002<\/td>\n<\/tr>\n
                        3<\/td>\ndtypes<\/td>\n\u8fd4\u56de\u6b64\u5bf9\u8c61\u4e2d\u7684\u6570\u636e\u7c7b\u578b(dtypes)\u3002<\/td>\n<\/tr>\n
                        4<\/td>\nempty<\/td>\n\u5982\u679cNDFrame\u5b8c\u5168\u4e3a\u7a7a[\u65e0\u9879\u76ee]\uff0c\u5219\u8fd4\u56de\u4e3aTrue; \u5982\u679c\u4efb\u4f55\u8f74\u7684\u957f\u5ea6\u4e3a0\u3002<\/td>\n<\/tr>\n
                        5<\/td>\nndim<\/td>\n\u8f74\/\u6570\u7ec4\u7ef4\u5ea6\u5927\u5c0f\u3002<\/td>\n<\/tr>\n
                        6<\/td>\nshape<\/td>\n\u8fd4\u56de\u8868\u793aDataFrame\u7684\u7ef4\u5ea6\u7684\u5143\u7ec4\u3002<\/td>\n<\/tr>\n
                        7<\/td>\nsize<\/td>\nNDFrame\u4e2d\u7684\u5143\u7d20\u6570\u3002<\/td>\n<\/tr>\n
                        8<\/td>\nvalues<\/td>\nNDFrame\u7684Numpy\u8868\u793a\u3002<\/td>\n<\/tr>\n
                        9<\/td>\nhead()<\/td>\n\u8fd4\u56de\u5f00\u5934\u524dn\u884c\u3002<\/td>\n<\/tr>\n
                        10<\/td>\ntail()<\/td>\n\u8fd4\u56de\u6700\u540en\u884c\u3002<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                        \u4e0b\u9762\u6765\u770b\u770b\u5982\u4f55\u521b\u5efa\u4e00\u4e2aDataFrame\u5e76\u4f7f\u7528\u4e0a\u8ff0\u5c5e\u6027\u548c\u65b9\u6cd5\u3002<\/p>\n
                        \u793a\u4f8b<\/strong><\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our data series is:\")\r\nprint df\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our data series is:\r\n  Age   Name    Rating\r\n0   25    Tom     4.23\r\n1   26    James   3.24\r\n2   25    Ricky   3.98\r\n3   23    Vin     2.56\r\n4   30    Steve   3.20\r\n5   29    Minsu   4.60\r\n6   23    Jack    3.80\r\n<\/pre>\n
                        T(\u8f6c\u7f6e)\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56deDataFrame\u7684\u8f6c\u7f6e\u3002\u884c\u548c\u5217\u5c06\u4ea4\u6362\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n# Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n# Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"The transpose of the data series is:\")\r\nprint df.T\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        The transpose of the data series is:\r\n0     1       2      3      4      5       6\r\nAge      25    26      25     23     30     29      23\r\nName     Tom   James   Ricky  Vin    Steve  Minsu   Jack\r\nRating   4.23  3.24    3.98   2.56   3.2    4.6     3.8\r\n<\/pre>\n
                        axes\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u884c\u8f74\u6807\u7b7e\u548c\u5217\u8f74\u6807\u7b7e\u5217\u8868\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Row axis labels and column axis labels are:\")\r\nprint df.axes\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Row axis labels and column axis labels are:\r\n\r\n[RangeIndex(start=0, stop=7, step=1), Index([u'Age', u'Name', u'Rating'],\r\ndtype='object')]\r\n<\/pre>\n
                        dtypes\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u6bcf\u5217\u7684\u6570\u636e\u7c7b\u578b\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"The data types of each column are:\")\r\nprint df.dtypes\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        The data types of each column are:\r\nAge     int64\r\nName    object\r\nRating  float64\r\ndtype: object\r\n<\/pre>\n
                        empty\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u5e03\u5c14\u503c\uff0c\u8868\u793a\u5bf9\u8c61\u662f\u5426\u4e3a\u7a7a; \u8fd4\u56deTrue\u8868\u793a\u5bf9\u8c61\u4e3a\u7a7a\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Is the object empty?\")\r\nprint df.empty\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Is the object empty?\r\nFalse\r\n<\/pre>\n
                        ndim\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u5bf9\u8c61\u7684\u7ef4\u6570\u3002\u6839\u636e\u5b9a\u4e49\uff0cDataFrame\u662f\u4e00\u4e2a2D\u5bf9\u8c61\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our object is:\")\r\nprint df\r\nprint (\"The dimension of the object is:\")\r\nprint df.ndim\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our object is:\r\nAge    Name     Rating\r\n0     25     Tom      4.23\r\n1     26     James    3.24\r\n2     25     Ricky    3.98\r\n3     23     Vin      2.56\r\n4     30     Steve    3.20\r\n5     29     Minsu    4.60\r\n6     23     Jack     3.80\r\n\r\nThe dimension of the object is:\r\n2\r\n<\/pre>\n
                        shape\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56de\u8868\u793aDataFrame\u7684\u7ef4\u5ea6\u7684\u5143\u7ec4\u3002 \u5143\u7ec4(a\uff0cb)\uff0c\u5176\u4e2da\u8868\u793a\u884c\u6570\uff0cb\u8868\u793a\u5217\u6570\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our object is:\")\r\nprint df\r\nprint (\"The shape of the object is:\")\r\nprint df.shape\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our object is:\r\nAge   Name    Rating\r\n0  25    Tom     4.23\r\n1  26    James   3.24\r\n2  25    Ricky   3.98\r\n3  23    Vin     2.56\r\n4  30    Steve   3.20\r\n5  29    Minsu   4.60\r\n6  23    Jack    3.80\r\n\r\nThe shape of the object is:\r\n(7, 3)\r\n<\/pre>\n
                        size\u793a\u4f8b<\/strong><\/p>\n
                        \u8fd4\u56deDataFrame\u4e2d\u7684\u5143\u7d20\u6570\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our object is:\")\r\nprint df\r\nprint (\"The total number of elements in our object is:\")\r\nprint df.size\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our object is:\r\nAge   Name    Rating\r\n0   25    Tom     4.23\r\n1   26    James   3.24\r\n2   25    Ricky   3.98\r\n3   23    Vin     2.56\r\n4   30    Steve   3.20\r\n5   29    Minsu   4.60\r\n6   23    Jack    3.80\r\n\r\nThe total number of elements in our object is:\r\n21\r\n<\/pre>\n
                        values\u793a\u4f8b<\/strong><\/p>\n
                        \u5c06DataFrame\u4e2d\u7684\u5b9e\u9645\u6570\u636e\u4f5c\u4e3aNDarray\u8fd4\u56de\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our object is:\")\r\nprint df\r\nprint (\"The actual data in our data frame is:\")\r\nprint df.values\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our object is:\r\nAge   Name    Rating\r\n0   25    Tom     4.23\r\n1   26    James   3.24\r\n2   25    Ricky   3.98\r\n3   23    Vin     2.56\r\n4   30    Steve   3.20\r\n5   29    Minsu   4.60\r\n6   23    Jack    3.80\r\nThe actual data in our data frame is:\r\n[[25 'Tom' 4.23]\r\n[26 'James' 3.24]\r\n[25 'Ricky' 3.98]\r\n[23 'Vin' 2.56]\r\n[30 'Steve' 3.2]\r\n[29 'Minsu' 4.6]\r\n[23 'Jack' 3.8]]\r\n<\/pre>\n
                        head()\u548ctail()\u793a\u4f8b<\/strong><\/p>\n
                        \u8981\u67e5\u770bDataFrame\u5bf9\u8c61\u7684\u5c0f\u6837\u672c\uff0c\u53ef\u4f7f\u7528head()\u548ctail()\u65b9\u6cd5\u3002head()\u8fd4\u56de\u524dn\u884c(\u89c2\u5bdf\u7d22\u5f15\u503c)\u3002\u663e\u793a\u5143\u7d20\u7684\u9ed8\u8ba4\u6570\u91cf\u4e3a5\uff0c\u4f46\u53ef\u4ee5\u4f20\u9012\u81ea\u5b9a\u4e49\u6570\u5b57\u503c\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our data frame is:\")\r\nprint df\r\nprint (\"The first two rows of the data frame is:\")\r\nprint df.head(2)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our data frame is:\r\nAge   Name    Rating\r\n0   25    Tom     4.23\r\n1   26    James   3.24\r\n2   25    Ricky   3.98\r\n3   23    Vin     2.56\r\n4   30    Steve   3.20\r\n5   29    Minsu   4.60\r\n6   23    Jack    3.80\r\n\r\nThe first two rows of the data frame is:\r\nAge   Name   Rating\r\n0  25    Tom    4.23\r\n1  26    James  3.24\r\n<\/pre>\n
                        tail()\u8fd4\u56de\u6700\u540en\u884c(\u89c2\u5bdf\u7d22\u5f15\u503c)\u3002\u663e\u793a\u5143\u7d20\u7684\u9ed8\u8ba4\u6570\u91cf\u4e3a5\uff0c\u4f46\u53ef\u4ee5\u4f20\u9012\u81ea\u5b9a\u4e49\u6570\u5b57\u503c\u3002<\/p>\n
                        import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack']),\r\n'Age':pd.Series([25,26,25,23,30,29,23]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint (\"Our data frame is:\")\r\nprint df\r\nprint (\"The last two rows of the data frame is:\")\r\nprint df.tail(2)\r\n<\/pre>\n
                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                        Our data frame is:\r\nAge   Name    Rating\r\n0   25    Tom     4.23\r\n1   26    James   3.24\r\n2   25    Ricky   3.98\r\n3   23    Vin     2.56\r\n4   30    Steve   3.20\r\n5   29    Minsu   4.60\r\n6   23    Jack    3.80\r\n\r\nThe last two rows of the data frame is:\r\nAge   Name    Rating\r\n5   29    Minsu    4.6\r\n6   23    Jack     3.8<\/pre>\n
                        Python Pandas\u63cf\u8ff0\u6027\u7edf\u8ba1<\/h1>\n
                        \u6709\u5f88\u591a\u65b9\u6cd5\u7528\u6765\u96c6\u4f53\u8ba1\u7b97DataFrame\u7684\u63cf\u8ff0\u6027\u7edf\u8ba1\u4fe1\u606f\u548c\u5176\u4ed6\u76f8\u5173\u64cd\u4f5c\u3002 \u5176\u4e2d\u5927\u591a\u6570\u662fsum()\uff0cmean()\u7b49\u805a\u5408\u51fd\u6570\uff0c\u4f46\u5176\u4e2d\u4e00\u4e9b\uff0c\u5982sumsum()\uff0c\u4ea7\u751f\u4e00\u4e2a\u76f8\u540c\u5927\u5c0f\u7684\u5bf9\u8c61\u3002 \u4e00\u822c\u6765\u8bf4\uff0c\u8fd9\u4e9b\u65b9\u6cd5\u91c7\u7528\u8f74\u53c2\u6570\uff0c\u5c31\u50cfndarray.{sum\uff0cstd\uff0c…}\uff0c\u4f46\u8f74\u53ef\u4ee5\u901a\u8fc7\u540d\u79f0\u6216\u6574\u6570\u6765\u6307\u5b9a\uff1a<\/p>\n
                          \n
                        • \u6570\u636e\u5e27(DataFrame) – \u201cindex\u201d(axis=0\uff0c\u9ed8\u8ba4)\uff0ccolumns(axis=1)<\/li>\n<\/ul>\n
                          \u4e0b\u9762\u521b\u5efa\u4e00\u4e2a\u6570\u636e\u5e27(DataFrame)\uff0c\u5e76\u4f7f\u7528\u6b64\u5bf9\u8c61\u8fdb\u884c\u6f14\u793a\u672c\u7ae0\u4e2d\u6240\u6709\u64cd\u4f5c\u3002<\/p>\n
                          \u793a\u4f8b<\/strong><\/p>\n
                          import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df\r\n<\/pre>\n
                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                             Age  Name   Rating\r\n0   25   Tom     4.23\r\n1   26   James   3.24\r\n2   25   Ricky   3.98\r\n3   23   Vin     2.56\r\n4   30   Steve   3.20\r\n5   29   Minsu   4.60\r\n6   23   Jack    3.80\r\n7   34   Lee     3.78\r\n8   40   David   2.98\r\n9   30   Gasper  4.80\r\n10  51   Betina  4.10\r\n11  46   Andres  3.65\r\n<\/pre>\n
                          sum()\u65b9\u6cd5<\/strong><\/p>\n
                          \u8fd4\u56de\u6240\u8bf7\u6c42\u8f74\u7684\u503c\u7684\u603b\u548c\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u8f74\u4e3a\u7d22\u5f15(axis=0)\u3002<\/p>\n
                          import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.sum()\r\n<\/pre>\n
                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                          Age                                                    382\r\nName     TomJamesRickyVinSteveMinsuJackLeeDavidGasperBe...\r\nRating                                               44.92\r\ndtype: object\r\n<\/pre>\n
                          \u6bcf\u4e2a\u5355\u72ec\u7684\u5217\u5355\u72ec\u6dfb\u52a0(\u9644\u52a0\u5b57\u7b26\u4e32)\u3002<\/p>\n
                          axis=1\u793a\u4f8b<\/strong><\/p>\n
                          \u6b64\u8bed\u6cd5\u5c06\u7ed9\u51fa\u5982\u4e0b\u6240\u793a\u7684\u8f93\u51fa\uff0c\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                          import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.sum(1)\r\n<\/pre>\n
                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                          0    29.23\r\n1    29.24\r\n2    28.98\r\n3    25.56\r\n4    33.20\r\n5    33.60\r\n6    26.80\r\n7    37.78\r\n8    42.98\r\n9    34.80\r\n10   55.10\r\n11   49.65\r\ndtype: float64\r\n<\/pre>\n
                          mean()\u793a\u4f8b<\/strong>
                          \n\u8fd4\u56de\u5e73\u5747\u503c\uff0c\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                          import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.mean()\r\n<\/pre>\n
                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                          Age       31.833333\r\nRating     3.743333\r\ndtype: float64\r\n<\/pre>\n
                          std()\u793a\u4f8b<\/strong><\/p>\n
                          \u8fd4\u56de\u6570\u5b57\u5217\u7684Bressel\u6807\u51c6\u504f\u5dee\u3002<\/p>\n
                          import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.std()\r\n<\/pre>\n
                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                          Age       9.232682\r\nRating    0.661628\r\ndtype: float64\r\n<\/pre>\n
                          \u51fd\u6570\u548c\u8bf4\u660e<\/h3>\n
                          \u4e0b\u9762\u6765\u4e86\u89e3Python Pandas\u4e2d\u63cf\u8ff0\u6027\u7edf\u8ba1\u4fe1\u606f\u7684\u51fd\u6570\uff0c\u4e0b\u8868\u5217\u51fa\u4e86\u91cd\u8981\u51fd\u6570 –<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
                          \u7f16\u53f7<\/th>\n\u51fd\u6570<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                          1<\/td>\ncount()<\/td>\n\u975e\u7a7a\u89c2\u6d4b\u6570\u91cf<\/td>\n<\/tr>\n
                          2<\/td>\nsum()<\/td>\n\u6240\u6709\u503c\u4e4b\u548c<\/td>\n<\/tr>\n
                          3<\/td>\nmean()<\/td>\n\u6240\u6709\u503c\u7684\u5e73\u5747\u503c<\/td>\n<\/tr>\n
                          4<\/td>\nmedian()<\/td>\n\u6240\u6709\u503c\u7684\u4e2d\u4f4d\u6570<\/td>\n<\/tr>\n
                          5<\/td>\nmode()<\/td>\n\u503c\u7684\u6a21\u503c<\/td>\n<\/tr>\n
                          6<\/td>\nstd()<\/td>\n\u503c\u7684\u6807\u51c6\u504f\u5dee<\/td>\n<\/tr>\n
                          7<\/td>\nmin()<\/td>\n\u6240\u6709\u503c\u4e2d\u7684\u6700\u5c0f\u503c<\/td>\n<\/tr>\n
                          8<\/td>\nmax()<\/td>\n\u6240\u6709\u503c\u4e2d\u7684\u6700\u5927\u503c<\/td>\n<\/tr>\n
                          9<\/td>\nabs()<\/td>\n\u7edd\u5bf9\u503c<\/td>\n<\/tr>\n
                          10<\/td>\nprod()<\/td>\n\u6570\u7ec4\u5143\u7d20\u7684\u4e58\u79ef<\/td>\n<\/tr>\n
                          11<\/td>\ncumsum()<\/td>\n\u7d2f\u8ba1\u603b\u548c<\/td>\n<\/tr>\n
                          12<\/td>\ncumprod()<\/td>\n\u7d2f\u8ba1\u4e58\u79ef<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                          \u6ce8 – \u7531\u4e8eDataFrame\u662f\u5f02\u6784\u6570\u636e\u7ed3\u6784\u3002\u901a\u7528\u64cd\u4f5c\u4e0d\u9002\u7528\u4e8e\u6240\u6709\u51fd\u6570\u3002<\/p>\n
                            \n
                          • \u7c7b\u4f3c\u4e8e\uff1asum()\uff0ccumsum()\u51fd\u6570\u80fd\u4e0e\u6570\u5b57\u548c\u5b57\u7b26(\u6216)\u5b57\u7b26\u4e32\u6570\u636e\u5143\u7d20\u4e00\u8d77\u5de5\u4f5c\uff0c\u4e0d\u4f1a\u4ea7\u751f\u4efb\u4f55\u9519\u8bef\u3002\u5b57\u7b26\u805a\u5408\u4ece\u6765\u90fd\u6bd4\u8f83\u5c11\u88ab\u4f7f\u7528\uff0c\u867d\u7136\u8fd9\u4e9b\u51fd\u6570\u4e0d\u4f1a\u5f15\u53d1\u4efb\u4f55\u5f02\u5e38\u3002<\/li>\n
                          • \u7531\u4e8e\u8fd9\u6837\u7684\u64cd\u4f5c\u65e0\u6cd5\u6267\u884c\uff0c\u56e0\u6b64\uff0c\u5f53DataFrame\u5305\u542b\u5b57\u7b26\u6216\u5b57\u7b26\u4e32\u6570\u636e\u65f6\uff0c\u50cfabs()\uff0ccumprod()\u8fd9\u6837\u7684\u51fd\u6570\u4f1a\u629b\u51fa\u5f02\u5e38\u3002<\/li>\n<\/ul>\n
                            \u6c47\u603b\u6570\u636e<\/h3>\n
                            describe()\u51fd\u6570\u662f\u7528\u6765\u8ba1\u7b97\u6709\u5173DataFrame\u5217\u7684\u7edf\u8ba1\u4fe1\u606f\u7684\u6458\u8981\u3002<\/p>\n
                            import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.describe()\r\n<\/pre>\n
                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                           Age         Rating\r\ncount    12.000000      12.000000\r\nmean     31.833333       3.743333\r\nstd       9.232682       0.661628\r\nmin      23.000000       2.560000\r\n25%      25.000000       3.230000\r\n50%      29.500000       3.790000\r\n75%      35.500000       4.132500\r\nmax      51.000000       4.800000\r\n<\/pre>\n
                            \u8be5\u51fd\u6570\u7ed9\u51fa\u4e86\u5e73\u5747\u503c\uff0c\u6807\u51c6\u5dee\u548cIQR\u503c\u3002 \u800c\u4e14\uff0c\u51fd\u6570\u6392\u9664\u5b57\u7b26\u5217\uff0c\u5e76\u7ed9\u51fa\u5173\u4e8e\u6570\u5b57\u5217\u7684\u6458\u8981\u3002 include\u662f\u7528\u4e8e\u4f20\u9012\u5173\u4e8e\u4ec0\u4e48\u5217\u9700\u8981\u8003\u8651\u7528\u4e8e\u603b\u7ed3\u7684\u5fc5\u8981\u4fe1\u606f\u7684\u53c2\u6570\u3002\u83b7\u53d6\u503c\u5217\u8868; \u9ed8\u8ba4\u60c5\u51b5\u4e0b\u662f\u201d\u6570\u5b57\u503c\u201d\u3002<\/p>\n
                              \n
                            • object – \u6c47\u603b\u5b57\u7b26\u4e32\u5217<\/li>\n
                            • number – \u6c47\u603b\u6570\u5b57\u5217<\/li>\n
                            • all – \u5c06\u6240\u6709\u5217\u6c47\u603b\u5728\u4e00\u8d77(\u4e0d\u5e94\u5c06\u5176\u4f5c\u4e3a\u5217\u8868\u503c\u4f20\u9012)<\/li>\n<\/ul>\n
                              \u73b0\u5728\uff0c\u5728\u7a0b\u5e8f\u4e2d\u4f7f\u7528\u4ee5\u4e0b\u8bed\u53e5\u5e76\u68c0\u67e5\u8f93\u51fa –<\/p>\n
                              import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df.describe(include=['object'])\r\n<\/pre>\n
                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                              Name\r\ncount       12\r\nunique      12\r\ntop      Ricky\r\nfreq         1\r\n<\/pre>\n
                              \u73b0\u5728\uff0c\u4f7f\u7528\u4ee5\u4e0b\u8bed\u53e5\u5e76\u67e5\u770b\u8f93\u51fa –<\/p>\n
                              import pandas as pd\r\nimport numpy as np\r\n\r\n#Create a Dictionary of series\r\nd = {'Name':pd.Series(['Tom','James','Ricky','Vin','Steve','Minsu','Jack',\r\n'Lee','David','Gasper','Betina','Andres']),\r\n'Age':pd.Series([25,26,25,23,30,29,23,34,40,30,51,46]),\r\n'Rating':pd.Series([4.23,3.24,3.98,2.56,3.20,4.6,3.8,3.78,2.98,4.80,4.10,3.65])}\r\n\r\n#Create a DataFrame\r\ndf = pd.DataFrame(d)\r\nprint df. describe(include='all')\r\n<\/pre>\n
                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        Age          Name       Rating\r\ncount   12.000000        12    12.000000\r\nunique        NaN        12          NaN\r\ntop           NaN     Ricky          NaN\r\nfreq          NaN         1          NaN\r\nmean    31.833333       NaN     3.743333\r\nstd      9.232682       NaN     0.661628\r\nmin     23.000000       NaN     2.560000\r\n25%     25.000000       NaN     3.230000\r\n50%     29.500000       NaN     3.790000\r\n75%     35.500000       NaN     4.132500\r\nmax     51.000000       NaN     4.800000<\/pre>\n
                              Python Pandas\u51fd\u6570\u5e94\u7528<\/h1>\n
                              \u8981\u5c06\u81ea\u5df1\u6216\u5176\u4ed6\u5e93\u7684\u51fd\u6570\u5e94\u7528\u4e8ePandas\u5bf9\u8c61\uff0c\u5e94\u8be5\u4e86\u89e3\u4e09\u79cd\u91cd\u8981\u7684\u65b9\u6cd5\u3002\u4ee5\u4e0b\u8ba8\u8bba\u4e86\u8fd9\u4e9b\u65b9\u6cd5\u3002 \u4f7f\u7528\u9002\u5f53\u7684\u65b9\u6cd5\u53d6\u51b3\u4e8e\u51fd\u6570\u662f\u5426\u671f\u671b\u5728\u6574\u4e2aDataFrame\uff0c\u884c\u6216\u5217\u6216\u5143\u7d20\u4e0a\u8fdb\u884c\u64cd\u4f5c\u3002<\/p>\n
                                \n
                              • \u8868\u660e\u667a\u51fd\u6570\u5e94\u7528\uff1apipe()<\/li>\n
                              • \u884c\u6216\u5217\u51fd\u6570\u5e94\u7528\uff1aapply()<\/li>\n
                              • \u5143\u7d20\u51fd\u6570\u5e94\u7528\uff1aapplymap()<\/li>\n<\/ul>\n
                                \u8868\u683c\u51fd\u6570\u5e94\u7528<\/h3>\n
                                \u53ef\u4ee5\u901a\u8fc7\u5c06\u51fd\u6570\u548c\u9002\u5f53\u6570\u91cf\u7684\u53c2\u6570\u4f5c\u4e3a\u7ba1\u9053\u53c2\u6570\u6765\u6267\u884c\u81ea\u5b9a\u4e49\u64cd\u4f5c\u3002 \u56e0\u6b64\uff0c\u5bf9\u6574\u4e2aDataFrame\u6267\u884c\u64cd\u4f5c\u3002<\/p>\n
                                \u4f8b\u5982\uff0c\u4e3aDataFrame\u4e2d\u7684\u6240\u6709\u5143\u7d20\u76f8\u52a0\u4e00\u4e2a\u503c2\u3002 \u7136\u540e\uff0c<\/p>\n
                                \u52a0\u6cd5\u5668\u51fd\u6570<\/strong><\/p>\n
                                \u52a0\u6cd5\u5668\u51fd\u6570\u5c06\u4e24\u4e2a\u6570\u503c\u4f5c\u4e3a\u53c2\u6570\u6dfb\u52a0\u5e76\u8fd4\u56de\u603b\u548c\u3002<\/p>\n
                                def adder(ele1,ele2):\r\nreturn ele1+ele2\r\n<\/pre>\n
                                \u73b0\u5728\u5c06\u4f7f\u7528\u81ea\u5b9a\u4e49\u51fd\u6570\u5bf9DataFrame\u8fdb\u884c\u64cd\u4f5c\u3002<\/p>\n
                                df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.pipe(adder,2)\r\n<\/pre>\n
                                \u4e0b\u9762\u6765\u770b\u770b\u5b8c\u6574\u7684\u7a0b\u5e8f –<\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\n\r\ndef adder(ele1,ele2):\r\nreturn ele1+ele2\r\n\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.pipe(adder,2)\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        col1       col2       col3\r\n0   2.176704   2.219691   1.509360\r\n1   2.222378   2.422167   3.953921\r\n2   2.241096   1.135424   2.696432\r\n3   2.355763   0.376672   1.182570\r\n4   2.308743   2.714767   2.130288\r\n<\/pre>\n
                                \u884c\u6216\u5217\u667a\u80fd\u51fd\u6570\u5e94\u7528<\/h3>\n
                                \u53ef\u4ee5\u4f7f\u7528apply()\u65b9\u6cd5\u6cbfDataFrame\u6216Panel\u7684\u8f74\u5e94\u7528\u4efb\u610f\u51fd\u6570\uff0c\u5b83\u4e0e\u63cf\u8ff0\u6027\u7edf\u8ba1\u65b9\u6cd5\u4e00\u6837\uff0c\u91c7\u7528\u53ef\u9009\u7684\u8f74\u53c2\u6570\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u64cd\u4f5c\u6309\u5217\u6267\u884c\uff0c\u5c06\u6bcf\u5217\u5217\u4e3a\u6570\u7ec4\u3002<\/p>\n
                                \u793a\u4f8b<\/strong><\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.apply(np.mean)\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        col1       col2        col3\r\n0   0.343569  -1.013287    1.131245\r\n\r\n1   0.508922  -0.949778   -1.600569\r\n\r\n2  -1.182331  -0.420703   -1.725400\r\n\r\n3   0.860265   2.069038   -0.537648\r\n\r\n4   0.876758  -0.238051    0.473992\r\n<\/pre>\n
                                \u901a\u8fc7\u4f20\u9012axis\u53c2\u6570\uff0c\u53ef\u4ee5\u5728\u884c\u4e0a\u6267\u884c\u64cd\u4f5c\u3002<\/p>\n
                                \u793a\u4f8b-2<\/strong><\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.apply(np.mean,axis=1)\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                       col1         col2         col3\r\n\r\n0  0.543255    -1.613418    -0.500731\r\n\r\n1  0.976543    -1.135835    -0.719153\r\n\r\n2  0.184282    -0.721153    -2.876206\r\n\r\n3  0.447738     0.268062    -1.937888\r\n\r\n4 -0.677673     0.177455     1.397360\r\n<\/pre>\n
                                \u793a\u4f8b-3<\/strong><\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.apply(lambda x: x.max() - x.min())\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                         col1        col2      col3\r\n\r\n0   -0.585206   -0.104938   1.424115\r\n\r\n1   -0.326036   -1.444798   0.196849\r\n\r\n2   -2.033478    1.682253   1.223152\r\n\r\n3   -0.107015    0.499846   0.084127\r\n\r\n4   -1.046964   -1.935617  -0.009919\r\n<\/pre>\n
                                \u5143\u7d20\u667a\u80fd\u51fd\u6570\u5e94\u7528<\/h3>\n
                                \u5e76\u4e0d\u662f\u6240\u6709\u7684\u51fd\u6570\u90fd\u53ef\u4ee5\u5411\u91cf\u5316(\u4e5f\u4e0d\u662f\u8fd4\u56de\u53e6\u4e00\u4e2a\u6570\u7ec4\u7684NumPy\u6570\u7ec4\uff0c\u4e5f\u4e0d\u662f\u4efb\u4f55\u503c)\uff0c\u5728DataFrame\u4e0a\u7684\u65b9\u6cd5applymap()\u548c\u7c7b\u4f3c\u5730\u5728Series\u4e0a\u7684map()\u63a5\u53d7\u4efb\u4f55Python\u51fd\u6570\uff0c\u5e76\u4e14\u8fd4\u56de\u5355\u4e2a\u503c\u3002<\/p>\n
                                \u793a\u4f8b-1<\/strong><\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\n\r\n# My custom function\r\ndf['col1'].map(lambda x:x*100)\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                         col1      col2       col3\r\n\r\n0    0.629348  0.088467  -1.790702\r\n\r\n1   -0.592595  0.184113  -1.524998\r\n\r\n2   -0.419298  0.262369  -0.178849\r\n\r\n3   -1.036930  1.103169   0.941882\r\n\r\n4   -0.573333 -0.031056   0.315590\r\n<\/pre>\n
                                \u793a\u4f8b-2<\/strong><\/p>\n
                                import pandas as pd\r\nimport numpy as np\r\n\r\n# My custom function\r\ndf = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])\r\ndf.applymap(lambda x:x*100)\r\nprint df\r\n<\/pre>\n
                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                output is as follows:\r\ncol1         col2         col3\r\n0   17.670426    21.969052    -49.064031\r\n1   22.237846    42.216693     195.392124\r\n2   24.109576   -86.457646     69.643171\r\n3   35.576312   -162.332803   -81.743023\r\n4   30.874333    71.476717     13.028751<\/pre>\n
                                Python Pandas\u91cd\u5efa\u7d22\u5f15<\/h1>\n
                                \u91cd\u65b0\u7d22\u5f15\u4f1a\u66f4\u6539DataFrame\u7684\u884c\u6807\u7b7e\u548c\u5217\u6807\u7b7e\u3002\u91cd\u65b0\u7d22\u5f15\u610f\u5473\u7740\u7b26\u5408\u6570\u636e\u4ee5\u5339\u914d\u7279\u5b9a\u8f74\u4e0a\u7684\u4e00\u7ec4\u7ed9\u5b9a\u7684\u6807\u7b7e\u3002<\/p>\n
                                \u53ef\u4ee5\u901a\u8fc7\u7d22\u5f15\u6765\u5b9e\u73b0\u591a\u4e2a\u64cd\u4f5c –<\/p>\n
                                  \n
                                • \u91cd\u65b0\u6392\u5e8f\u73b0\u6709\u6570\u636e\u4ee5\u5339\u914d\u4e00\u7ec4\u65b0\u7684\u6807\u7b7e\u3002<\/li>\n
                                • \u5728\u6ca1\u6709\u6807\u7b7e\u6570\u636e\u7684\u6807\u7b7e\u4f4d\u7f6e\u63d2\u5165\u7f3a\u5931\u503c(NA)\u6807\u8bb0\u3002<\/li>\n<\/ul>\n
                                  \u793a\u4f8b<\/strong><\/p>\n
                                  import pandas as pd\r\nimport numpy as np\r\n\r\nN=20\r\n\r\ndf = pd.DataFrame({\r\n'A': pd.date_range(start='2016-01-01',periods=N,freq='D'),\r\n'x': np.linspace(0,stop=N-1,num=N),\r\n'y': np.random.rand(N),\r\n'C': np.random.choice(['Low','Medium','High'],N).tolist(),\r\n'D': np.random.normal(100, 10, size=(N)).tolist()\r\n})\r\n\r\n#reindex the DataFrame\r\ndf_reindexed = df.reindex(index=[0,2,5], columns=['A', 'C', 'B'])\r\n\r\nprint (df_reindexed)\r\n<\/pre>\n
                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              A    C     B\r\n0  2016-01-01  Low   NaN\r\n2  2016-01-03  High  NaN\r\n5  2016-01-06  Low   NaN\r\n<\/pre>\n
                                  \u91cd\u5efa\u7d22\u5f15\u4e0e\u5176\u4ed6\u5bf9\u8c61\u5bf9\u9f50<\/h3>\n
                                  \u6709\u65f6\u53ef\u80fd\u5e0c\u671b\u91c7\u53d6\u4e00\u4e2a\u5bf9\u8c61\u548c\u91cd\u65b0\u7d22\u5f15\uff0c\u5176\u8f74\u88ab\u6807\u8bb0\u4e3a\u4e0e\u53e6\u4e00\u4e2a\u5bf9\u8c61\u76f8\u540c\u3002 \u8003\u8651\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e00\u70b9\u3002<\/p>\n
                                  \u793a\u4f8b<\/strong><\/p>\n
                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf1 = pd.DataFrame(np.random.randn(10,3),columns=['col1','col2','col3'])\r\ndf2 = pd.DataFrame(np.random.randn(7,3),columns=['col1','col2','col3'])\r\n\r\ndf1 = df1.reindex_like(df2)\r\nprint df1\r\n<\/pre>\n
                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            col1         col2         col3\r\n0    -2.467652    -1.211687    -0.391761\r\n1    -0.287396     0.522350     0.562512\r\n2    -0.255409    -0.483250     1.866258\r\n3    -1.150467    -0.646493    -0.222462\r\n4     0.152768    -2.056643     1.877233\r\n5    -1.155997     1.528719    -1.343719\r\n6    -1.015606    -1.245936    -0.295275\r\n<\/pre>\n
                                  \u6ce8\u610f – \u5728\u8fd9\u91cc\uff0cdf1\u6570\u636e\u5e27(DataFrame)\u88ab\u66f4\u6539\u5e76\u91cd\u65b0\u7f16\u53f7\uff0c\u5982df2\u3002 \u5217\u540d\u79f0\u5e94\u8be5\u5339\u914d\uff0c\u5426\u5219\u5c06\u4e3a\u6574\u4e2a\u5217\u6807\u7b7e\u6dfb\u52a0NAN\u3002<\/p>\n
                                  \u586b\u5145\u65f6\u91cd\u65b0\u52a0\u6ce8<\/h3>\n
                                  reindex()\u91c7\u7528\u53ef\u9009\u53c2\u6570\u65b9\u6cd5\uff0c\u5b83\u662f\u4e00\u4e2a\u586b\u5145\u65b9\u6cd5\uff0c\u5176\u503c\u5982\u4e0b\uff1a<\/p>\n
                                    \n
                                  • pad\/ffill – \u5411\u524d\u586b\u5145\u503c<\/li>\n
                                  • bfill\/backfill – \u5411\u540e\u586b\u5145\u503c<\/li>\n
                                  • nearest – \u4ece\u6700\u8fd1\u7684\u7d22\u5f15\u503c\u586b\u5145<\/li>\n<\/ul>\n
                                    \u793a\u4f8b<\/strong><\/p>\n
                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf1 = pd.DataFrame(np.random.randn(6,3),columns=['col1','col2','col3'])\r\ndf2 = pd.DataFrame(np.random.randn(2,3),columns=['col1','col2','col3'])\r\n\r\n# Padding NAN's\r\nprint df2.reindex_like(df1)\r\n\r\n# Now Fill the NAN's with preceding Values\r\nprint (\"Data Frame with Forward Fill:\")\r\nprint df2.reindex_like(df1,method='ffill')\r\n<\/pre>\n
                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\u65f6\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1        col2       col3\r\n0    1.311620   -0.707176   0.599863\r\n1   -0.423455   -0.700265   1.133371\r\n2         NaN         NaN        NaN\r\n3         NaN         NaN        NaN\r\n4         NaN         NaN        NaN\r\n5         NaN         NaN        NaN\r\n\r\nData Frame with Forward Fill:\r\ncol1        col2        col3\r\n0    1.311620   -0.707176    0.599863\r\n1   -0.423455   -0.700265    1.133371\r\n2   -0.423455   -0.700265    1.133371\r\n3   -0.423455   -0.700265    1.133371\r\n4   -0.423455   -0.700265    1.133371\r\n5   -0.423455   -0.700265    1.133371\r\n<\/pre>\n
                                    \u6ce8 – \u6700\u540e\u56db\u884c\u88ab\u586b\u5145\u4e86\u3002<\/p>\n
                                    \u91cd\u5efa\u7d22\u5f15\u65f6\u7684\u586b\u5145\u9650\u5236<\/h3>\n
                                    \u9650\u5236\u53c2\u6570\u5728\u91cd\u5efa\u7d22\u5f15\u65f6\u63d0\u4f9b\u5bf9\u586b\u5145\u7684\u989d\u5916\u63a7\u5236\u3002\u9650\u5236\u6307\u5b9a\u8fde\u7eed\u5339\u914d\u7684\u6700\u5927\u8ba1\u6570\u3002\u8003\u8651\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e2a\u6982\u5ff5 –<\/p>\n
                                    \u793a\u4f8b<\/strong><\/p>\n
                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf1 = pd.DataFrame(np.random.randn(6,3),columns=['col1','col2','col3'])\r\ndf2 = pd.DataFrame(np.random.randn(2,3),columns=['col1','col2','col3'])\r\n\r\n# Padding NAN's\r\nprint df2.reindex_like(df1)\r\n\r\n# Now Fill the NAN's with preceding Values\r\nprint (\"Data Frame with Forward Fill limiting to 1:\")\r\nprint df2.reindex_like(df1,method='ffill',limit=1)\r\n<\/pre>\n
                                    \u5728\u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\u65f6\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1        col2        col3\r\n0    0.247784    2.128727    0.702576\r\n1   -0.055713   -0.021732   -0.174577\r\n2         NaN         NaN         NaN\r\n3         NaN         NaN         NaN\r\n4         NaN         NaN         NaN\r\n5         NaN         NaN         NaN\r\n\r\nData Frame with Forward Fill limiting to 1:\r\ncol1        col2        col3\r\n0    0.247784    2.128727    0.702576\r\n1   -0.055713   -0.021732   -0.174577\r\n2   -0.055713   -0.021732   -0.174577\r\n3         NaN         NaN         NaN\r\n4         NaN         NaN         NaN\r\n5         NaN         NaN         NaN\r\n<\/pre>\n
                                    \u6ce8\u610f – \u53ea\u6709\u7b2c7\u884c\u7531\u524d6\u884c\u586b\u5145\u3002 \u7136\u540e\uff0c\u5176\u5b83\u884c\u6309\u539f\u6837\u4fdd\u7559\u3002<\/p>\n
                                    \u91cd\u547d\u540d<\/h3>\n
                                    rename()\u65b9\u6cd5\u5141\u8bb8\u57fa\u4e8e\u4e00\u4e9b\u6620\u5c04(\u5b57\u5178\u6216\u8005\u7cfb\u5217)\u6216\u4efb\u610f\u51fd\u6570\u6765\u91cd\u65b0\u6807\u8bb0\u4e00\u4e2a\u8f74\u3002
                                    \n\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e00\u6982\u5ff5\u3002<\/p>\n
                                    \u793a\u4f8b<\/strong><\/p>\n
                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf1 = pd.DataFrame(np.random.randn(6,3),columns=['col1','col2','col3'])\r\nprint df1\r\n\r\nprint (\"After renaming the rows and columns:\")\r\nprint df1.rename(columns={'col1' : 'c1', 'col2' : 'c2'},\r\nindex = {0 : 'apple', 1 : 'banana', 2 : 'durian'})\r\n<\/pre>\n
                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1        col2        col3\r\n0    0.486791    0.105759    1.540122\r\n1   -0.990237    1.007885   -0.217896\r\n2   -0.483855   -1.645027   -1.194113\r\n3   -0.122316    0.566277   -0.366028\r\n4   -0.231524   -0.721172   -0.112007\r\n5    0.438810    0.000225    0.435479\r\n\r\nAfter renaming the rows and columns:\r\nc1          c2        col3\r\napple     0.486791    0.105759    1.540122\r\nbanana   -0.990237    1.007885   -0.217896\r\ndurian   -0.483855   -1.645027   -1.194113\r\n3        -0.122316    0.566277   -0.366028\r\n4        -0.231524   -0.721172   -0.112007\r\n5         0.438810    0.000225    0.435479\r\n<\/pre>\n
                                    rename()\u65b9\u6cd5\u63d0\u4f9b\u4e86\u4e00\u4e2ainplace\u547d\u540d\u53c2\u6570\uff0c\u9ed8\u8ba4\u4e3aFalse\u5e76\u590d\u5236\u5e95\u5c42\u6570\u636e\u3002 \u6307\u5b9a\u4f20\u9012inplace = True\u5219\u8868\u793a\u5c06\u6570\u636e\u91cd\u547d\u540d\u3002<\/p>\n
                                    Python Pandas\u8fed\u4ee3<\/h1>\n
                                    Pandas\u5bf9\u8c61\u4e4b\u95f4\u7684\u57fa\u672c\u8fed\u4ee3\u7684\u884c\u4e3a\u53d6\u51b3\u4e8e\u7c7b\u578b\u3002\u5f53\u8fed\u4ee3\u4e00\u4e2a\u7cfb\u5217\u65f6\uff0c\u5b83\u88ab\u89c6\u4e3a\u6570\u7ec4\u5f0f\uff0c\u57fa\u672c\u8fed\u4ee3\u4ea7\u751f\u8fd9\u4e9b\u503c\u3002\u5176\u4ed6\u6570\u636e\u7ed3\u6784\uff0c\u5982\uff1aDataFrame\u548cPanel\uff0c\u9075\u5faa\u7c7b\u4f3c\u60ef\u4f8b\u8fed\u4ee3\u5bf9\u8c61\u7684\u952e\u3002<\/p>\n
                                    \u7b80\u800c\u8a00\u4e4b\uff0c\u57fa\u672c\u8fed\u4ee3(\u5bf9\u4e8ei\u5728\u5bf9\u8c61\u4e2d)\u4ea7\u751f –<\/p>\n
                                      \n
                                    • Series – \u503c<\/li>\n
                                    • DataFrame – \u5217\u6807\u7b7e<\/li>\n
                                    • Pannel – \u9879\u76ee\u6807\u7b7e<\/li>\n<\/ul>\n
                                      \u8fed\u4ee3DataFrame<\/h3>\n
                                      \u8fed\u4ee3DataFrame\u63d0\u4f9b\u5217\u540d\u3002\u73b0\u5728\u6765\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e2a\u6982\u5ff5\u3002<\/p>\n
                                      import pandas as pd\r\nimport numpy as np\r\n\r\nN=20\r\n\r\ndf = pd.DataFrame({\r\n\t'A': pd.date_range(start='2016-01-01',periods=N,freq='D'),\r\n\t'x': np.linspace(0,stop=N-1,num=N),\r\n\t'y': np.random.rand(N),\r\n\t'C': np.random.choice(['Low','Medium','High'],N).tolist(),\r\n\t'D': np.random.normal(100, 10, size=(N)).tolist()\r\n})\r\n\r\nfor col in df:\r\n\tprint (col)\r\n<\/pre>\n
                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                      A\r\nC\r\nD\r\nx\r\ny\r\n<\/pre>\n
                                      \u8981\u904d\u5386\u6570\u636e\u5e27(DataFrame)\u4e2d\u7684\u884c\uff0c\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u51fd\u6570 –<\/p>\n
                                        \n
                                      • iteritems() – \u8fed\u4ee3(key\uff0cvalue)\u5bf9<\/li>\n
                                      • iterrows() – \u5c06\u884c\u8fed\u4ee3\u4e3a(\u7d22\u5f15\uff0c\u7cfb\u5217)\u5bf9<\/li>\n
                                      • itertuples() – \u4ee5namedtuples\u7684\u5f62\u5f0f\u8fed\u4ee3\u884c<\/li>\n<\/ul>\n
                                        iteritems()\u793a\u4f8b<\/strong><\/p>\n
                                        \u5c06\u6bcf\u4e2a\u5217\u4f5c\u4e3a\u952e\uff0c\u5c06\u503c\u4e0e\u503c\u4f5c\u4e3a\u952e\u548c\u5217\u503c\u8fed\u4ee3\u4e3aSeries\u5bf9\u8c61\u3002<\/p>\n
                                        import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(4,3),columns=['col1','col2','col3'])\r\nfor key,value in df.iteritems():\r\nprint (key,value)\r\n<\/pre>\n
                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        col1 0    0.802390\r\n1    0.324060\r\n2    0.256811\r\n3    0.839186\r\nName: col1, dtype: float64\r\n\r\ncol2 0    1.624313\r\n1   -1.033582\r\n2    1.796663\r\n3    1.856277\r\nName: col2, dtype: float64\r\n\r\ncol3 0   -0.022142\r\n1   -0.230820\r\n2    1.160691\r\n3   -0.830279\r\nName: col3, dtype: float64\r\n<\/pre>\n
                                        \u89c2\u5bdf\u4e00\u4e0b\uff0c\u5355\u72ec\u8fed\u4ee3\u6bcf\u4e2a\u5217\u4f5c\u4e3a\u7cfb\u5217\u4e2d\u7684\u952e\u503c\u5bf9\u3002<\/p>\n
                                        iterrows()\u793a\u4f8b<\/strong><\/p>\n
                                        iterrows()\u8fd4\u56de\u8fed\u4ee3\u5668\uff0c\u4ea7\u751f\u6bcf\u4e2a\u7d22\u5f15\u503c\u4ee5\u53ca\u5305\u542b\u6bcf\u884c\u6570\u636e\u7684\u5e8f\u5217\u3002<\/p>\n
                                        import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(4,3),columns = ['col1','col2','col3'])\r\nfor row_index,row in df.iterrows():\r\n\tprint (row_index,row)\r\n<\/pre>\n
                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        0  col1    1.529759\r\ncol2    0.762811\r\ncol3   -0.634691\r\nName: 0, dtype: float64\r\n\r\n1  col1   -0.944087\r\ncol2    1.420919\r\ncol3   -0.507895\r\nName: 1, dtype: float64\r\n\r\n2  col1   -0.077287\r\ncol2   -0.858556\r\ncol3   -0.663385\r\nName: 2, dtype: float64\r\n3  col1    -1.638578\r\ncol2     0.059866\r\ncol3     0.493482\r\nName: 3, dtype: float64\r\n<\/pre>\n
                                        \u6ce8\u610f – \u7531\u4e8eiterrows()\u904d\u5386\u884c\uff0c\u56e0\u6b64\u4e0d\u4f1a\u8de8\u8be5\u884c\u4fdd\u7559\u6570\u636e\u7c7b\u578b\u30020,1,2\u662f\u884c\u7d22\u5f15\uff0ccol1\uff0ccol2\uff0ccol3\u662f\u5217\u7d22\u5f15\u3002<\/p>\n
                                        itertuples()\u793a\u4f8b<\/strong><\/p>\n
                                        itertuples()\u65b9\u6cd5\u5c06\u4e3aDataFrame\u4e2d\u7684\u6bcf\u4e00\u884c\u8fd4\u56de\u4e00\u4e2a\u4ea7\u751f\u4e00\u4e2a\u547d\u540d\u5143\u7ec4\u7684\u8fed\u4ee3\u5668\u3002\u5143\u7ec4\u7684\u7b2c\u4e00\u4e2a\u5143\u7d20\u5c06\u662f\u884c\u7684\u76f8\u5e94\u7d22\u5f15\u503c\uff0c\u800c\u5269\u4f59\u7684\u503c\u662f\u884c\u503c\u3002<\/p>\n
                                        \u793a\u4f8b<\/strong><\/p>\n
                                        import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(4,3),columns = ['col1','col2','col3'])\r\nfor row in df.itertuples():\r\n\tprint (row)\r\n<\/pre>\n
                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                        Pandas(Index=0, col1=1.5297586201375899, col2=0.76281127433814944, col3=-\r\n0.6346908238310438)\r\n\r\nPandas(Index=1, col1=-0.94408735763808649, col2=1.4209186418359423, col3=-\r\n0.50789517967096232)\r\n\r\nPandas(Index=2, col1=-0.07728664756791935, col2=-0.85855574139699076, col3=-\r\n0.6633852507207626)\r\n\r\nPandas(Index=3, col1=0.65734942534106289, col2=-0.95057710432604969,\r\ncol3=0.80344487462316527)\r\n<\/pre>\n
                                        \u6ce8\u610f – \u4e0d\u8981\u5c1d\u8bd5\u5728\u8fed\u4ee3\u65f6\u4fee\u6539\u4efb\u4f55\u5bf9\u8c61\u3002\u8fed\u4ee3\u662f\u7528\u4e8e\u8bfb\u53d6\uff0c\u8fed\u4ee3\u5668\u8fd4\u56de\u539f\u59cb\u5bf9\u8c61(\u89c6\u56fe)\u7684\u526f\u672c\uff0c\u56e0\u6b64\u66f4\u6539\u5c06\u4e0d\u4f1a\u53cd\u6620\u5728\u539f\u59cb\u5bf9\u8c61\u4e0a\u3002<\/p>\n
                                        \u793a\u4f8b\u4ee3\u7801<\/strong><\/p>\n
                                        import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(4,3),columns = ['col1','col2','col3'])\r\n\r\nfor index, row in df.iterrows():\r\n\trow['a'] = 10\r\nprint (df)\r\n<\/pre>\n
                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                col1       col2       col3\r\n0  -1.739815   0.735595  -0.295589\r\n1   0.635485   0.106803   1.527922\r\n2  -0.939064   0.547095   0.038585\r\n3  -1.016509  -0.116580  -0.523158\r\n<\/pre>\n
                                        \u6ce8\u610f\u89c2\u5bdf\u7ed3\u679c\uff0c\u4fee\u6539\u53d8\u5316\u5e76\u672a\u53cd\u6620\u51fa\u6765\u3002<\/p>\n
                                        Python Pandas\u6392\u5e8f<\/h1>\n
                                        Pandas\u6709\u4e24\u79cd\u6392\u5e8f\u65b9\u5f0f\uff0c\u5b83\u4eec\u5206\u522b\u662f –<\/p>\n
                                          \n
                                        • \u6309\u6807\u7b7e<\/li>\n
                                        • \u6309\u5b9e\u9645\u503c<\/li>\n<\/ul>\n
                                          \u4e0b\u9762\u6765\u770b\u770b\u4e00\u4e2a\u8f93\u51fa\u7684\u4f8b\u5b50\u3002<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df=pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],colu\r\nmns=['col2','col1'])\r\nprint (unsorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                 col2      col1\r\n1  1.069838  0.096230\r\n4 -0.542406 -0.219829\r\n6 -0.071661  0.392091\r\n2  1.399976 -0.472169\r\n3  0.428372 -0.624630\r\n5  0.471875  0.966560\r\n9 -0.131851 -1.254495\r\n8  1.180651  0.199548\r\n0  0.906202  0.418524\r\n7  0.124800  2.011962\r\n<\/pre>\n
                                          \u5728unsorted_df\u6570\u636e\u503c\u4e2d\uff0c\u6807\u7b7e\u548c\u503c\u672a\u6392\u5e8f\u3002\u4e0b\u9762\u6765\u770b\u770b\u5982\u4f55\u6309\u6807\u7b7e\u6765\u6392\u5e8f\u3002<\/p>\n
                                          \u6309\u6807\u7b7e\u6392\u5e8f<\/h3>\n
                                          \u4f7f\u7528sort_index()\u65b9\u6cd5\uff0c\u901a\u8fc7\u4f20\u9012axis\u53c2\u6570\u548c\u6392\u5e8f\u987a\u5e8f\uff0c\u53ef\u4ee5\u5bf9DataFrame\u8fdb\u884c\u6392\u5e8f\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u6309\u7167\u5347\u5e8f\u5bf9\u884c\u6807\u7b7e\u8fdb\u884c\u6392\u5e8f\u3002<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df=unsorted_df.sort_index()\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                 col2      col1\r\n0  0.431384 -0.401538\r\n1  0.111887 -0.222582\r\n2 -0.166893 -0.237506\r\n3  0.476472  0.508397\r\n4  0.670838  0.406476\r\n5  2.065969 -0.324510\r\n6 -0.441630  1.060425\r\n7  0.735145  0.972447\r\n8 -0.051904 -1.112292\r\n9  0.134108  0.759698\r\n<\/pre>\n
                                          \u6392\u5e8f\u987a\u5e8f<\/h3>\n
                                          \u901a\u8fc7\u5c06\u5e03\u5c14\u503c\u4f20\u9012\u7ed9\u5347\u5e8f\u53c2\u6570\uff0c\u53ef\u4ee5\u63a7\u5236\u6392\u5e8f\u987a\u5e8f\u3002 \u6765\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u4e00\u4e0b\u3002<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df = unsorted_df.sort_index(ascending=False)\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                 col2      col1\r\n9  0.750452  1.754815\r\n8  0.945238  2.079394\r\n7  0.345238 -0.162737\r\n6 -0.512060  0.887094\r\n5  1.163144  0.595402\r\n4 -0.063584 -0.185536\r\n3 -0.275438 -2.286831\r\n2 -1.504792 -1.222394\r\n1  1.031234 -1.848174\r\n0 -0.615083  0.784086\r\n<\/pre>\n
                                          \u6309\u5217\u6392\u5217<\/h3>\n
                                          \u901a\u8fc7\u4f20\u9012axis\u53c2\u6570\u503c\u4e3a0\u62161\uff0c\u53ef\u4ee5\u5bf9\u5217\u6807\u7b7e\u8fdb\u884c\u6392\u5e8f\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0caxis = 0\uff0c\u9010\u884c\u6392\u5217\u3002\u6765\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e2a\u6982\u5ff5\u3002<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df=unsorted_df.sort_index(axis=1)\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                 col1      col2\r\n1 -0.997962  0.736707\r\n4  1.196464  0.703710\r\n6 -0.387800  1.207803\r\n2  1.614043  0.356389\r\n3 -0.057181 -0.551742\r\n5  1.034451 -0.731490\r\n9 -0.564355  0.892203\r\n8 -0.763526  0.684207\r\n0 -1.213615  1.268649\r\n7  0.316543 -1.450784\r\n<\/pre>\n
                                          \u6309\u503c\u6392\u5e8f<\/h3>\n
                                          \u50cf\u7d22\u5f15\u6392\u5e8f\u4e00\u6837\uff0csort_values()\u662f\u6309\u503c\u6392\u5e8f\u7684\u65b9\u6cd5\u3002\u5b83\u63a5\u53d7\u4e00\u4e2aby\u53c2\u6570\uff0c\u5b83\u5c06\u4f7f\u7528\u8981\u4e0e\u5176\u6392\u5e8f\u503c\u7684DataFrame\u7684\u5217\u540d\u79f0\u3002<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by='col1')\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1  col2\r\n1     1     3\r\n2     1     2\r\n3     1     4\r\n0     2     1\r\n<\/pre>\n
                                          \u6ce8\u610f\uff1a \u89c2\u5bdf\u4e0a\u9762\u7684\u8f93\u51fa\u7ed3\u679c\uff0ccol1\u503c\u88ab\u6392\u5e8f\uff0c\u76f8\u5e94\u7684col2\u503c\u548c\u884c\u7d22\u5f15\u5c06\u968fcol1\u4e00\u8d77\u6539\u53d8\u3002\u56e0\u6b64\uff0c\u5b83\u4eec\u770b\u8d77\u6765\u6ca1\u6709\u6392\u5e8f\u3002<\/p>\n
                                          \u901a\u8fc7by\u53c2\u6570\u6307\u5b9a\u9700\u8981\u5217\u503c\uff0c\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by=['col1','col2'])\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1  col2\r\n2     1     2\r\n1     1     3\r\n3     1     4\r\n0     2     1\r\n<\/pre>\n
                                          \u6392\u5e8f\u7b97\u6cd5<\/h3>\n
                                          sort_values()\u63d0\u4f9b\u4e86\u4ecemergeesort\uff0cheapsort\u548cquicksort\u4e2d\u9009\u62e9\u7b97\u6cd5\u7684\u4e00\u4e2a\u914d\u7f6e\u3002Mergesort\u662f\u552f\u4e00\u7a33\u5b9a\u7684\u7b97\u6cd5\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                          import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by='col1' ,kind='mergesort')\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                             col1  col2\r\n1     1     3\r\n2     1     2\r\n3     1     4\r\n0     2     1<\/pre>\n
                                          Python Pandas\u6392\u5e8f<\/h3>\n
                                          Pandas\u6709\u4e24\u79cd\u6392\u5e8f\u65b9\u5f0f\uff0c\u5b83\u4eec\u5206\u522b\u662f –<\/p>\n
                                            \n
                                          • \u6309\u6807\u7b7e<\/li>\n
                                          • \u6309\u5b9e\u9645\u503c<\/li>\n<\/ul>\n
                                            \u4e0b\u9762\u6765\u770b\u770b\u4e00\u4e2a\u8f93\u51fa\u7684\u4f8b\u5b50\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df=pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],colu\r\nmns=['col2','col1'])\r\nprint (unsorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                   col2      col1\r\n1  1.069838  0.096230\r\n4 -0.542406 -0.219829\r\n6 -0.071661  0.392091\r\n2  1.399976 -0.472169\r\n3  0.428372 -0.624630\r\n5  0.471875  0.966560\r\n9 -0.131851 -1.254495\r\n8  1.180651  0.199548\r\n0  0.906202  0.418524\r\n7  0.124800  2.011962\r\n<\/pre>\n
                                            \u5728unsorted_df\u6570\u636e\u503c\u4e2d\uff0c\u6807\u7b7e\u548c\u503c\u672a\u6392\u5e8f\u3002\u4e0b\u9762\u6765\u770b\u770b\u5982\u4f55\u6309\u6807\u7b7e\u6765\u6392\u5e8f\u3002<\/p>\n
                                            \u6309\u6807\u7b7e\u6392\u5e8f<\/h3>\n
                                            \u4f7f\u7528sort_index()\u65b9\u6cd5\uff0c\u901a\u8fc7\u4f20\u9012axis\u53c2\u6570\u548c\u6392\u5e8f\u987a\u5e8f\uff0c\u53ef\u4ee5\u5bf9DataFrame\u8fdb\u884c\u6392\u5e8f\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u6309\u7167\u5347\u5e8f\u5bf9\u884c\u6807\u7b7e\u8fdb\u884c\u6392\u5e8f\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df=unsorted_df.sort_index()\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                   col2      col1\r\n0  0.431384 -0.401538\r\n1  0.111887 -0.222582\r\n2 -0.166893 -0.237506\r\n3  0.476472  0.508397\r\n4  0.670838  0.406476\r\n5  2.065969 -0.324510\r\n6 -0.441630  1.060425\r\n7  0.735145  0.972447\r\n8 -0.051904 -1.112292\r\n9  0.134108  0.759698\r\n<\/pre>\n
                                            \u6392\u5e8f\u987a\u5e8f<\/h3>\n
                                            \u901a\u8fc7\u5c06\u5e03\u5c14\u503c\u4f20\u9012\u7ed9\u5347\u5e8f\u53c2\u6570\uff0c\u53ef\u4ee5\u63a7\u5236\u6392\u5e8f\u987a\u5e8f\u3002 \u6765\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u4e00\u4e0b\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df = unsorted_df.sort_index(ascending=False)\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                   col2      col1\r\n9  0.750452  1.754815\r\n8  0.945238  2.079394\r\n7  0.345238 -0.162737\r\n6 -0.512060  0.887094\r\n5  1.163144  0.595402\r\n4 -0.063584 -0.185536\r\n3 -0.275438 -2.286831\r\n2 -1.504792 -1.222394\r\n1  1.031234 -1.848174\r\n0 -0.615083  0.784086\r\n<\/pre>\n
                                            \u6309\u5217\u6392\u5217<\/h3>\n
                                            \u901a\u8fc7\u4f20\u9012axis\u53c2\u6570\u503c\u4e3a0\u62161\uff0c\u53ef\u4ee5\u5bf9\u5217\u6807\u7b7e\u8fdb\u884c\u6392\u5e8f\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0caxis = 0\uff0c\u9010\u884c\u6392\u5217\u3002\u6765\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50\u6765\u7406\u89e3\u8fd9\u4e2a\u6982\u5ff5\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame(np.random.randn(10,2),index=[1,4,6,2,3,5,9,8,0,7],columns = ['col2','col1'])\r\n\r\nsorted_df=unsorted_df.sort_index(axis=1)\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                   col1      col2\r\n1 -0.997962  0.736707\r\n4  1.196464  0.703710\r\n6 -0.387800  1.207803\r\n2  1.614043  0.356389\r\n3 -0.057181 -0.551742\r\n5  1.034451 -0.731490\r\n9 -0.564355  0.892203\r\n8 -0.763526  0.684207\r\n0 -1.213615  1.268649\r\n7  0.316543 -1.450784\r\n<\/pre>\n
                                            \u6309\u503c\u6392\u5e8f<\/h3>\n
                                            \u50cf\u7d22\u5f15\u6392\u5e8f\u4e00\u6837\uff0csort_values()\u662f\u6309\u503c\u6392\u5e8f\u7684\u65b9\u6cd5\u3002\u5b83\u63a5\u53d7\u4e00\u4e2aby\u53c2\u6570\uff0c\u5b83\u5c06\u4f7f\u7528\u8981\u4e0e\u5176\u6392\u5e8f\u503c\u7684DataFrame\u7684\u5217\u540d\u79f0\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by='col1')\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                               col1  col2\r\n1     1     3\r\n2     1     2\r\n3     1     4\r\n0     2     1\r\n<\/pre>\n
                                            \u6ce8\u610f\uff1a \u89c2\u5bdf\u4e0a\u9762\u7684\u8f93\u51fa\u7ed3\u679c\uff0ccol1\u503c\u88ab\u6392\u5e8f\uff0c\u76f8\u5e94\u7684col2\u503c\u548c\u884c\u7d22\u5f15\u5c06\u968fcol1\u4e00\u8d77\u6539\u53d8\u3002\u56e0\u6b64\uff0c\u5b83\u4eec\u770b\u8d77\u6765\u6ca1\u6709\u6392\u5e8f\u3002<\/p>\n
                                            \u901a\u8fc7by\u53c2\u6570\u6307\u5b9a\u9700\u8981\u5217\u503c\uff0c\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by=['col1','col2'])\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                               col1  col2\r\n2     1     2\r\n1     1     3\r\n3     1     4\r\n0     2     1\r\n<\/pre>\n
                                            \u6392\u5e8f\u7b97\u6cd5<\/h3>\n
                                            sort_values()\u63d0\u4f9b\u4e86\u4ecemergeesort\uff0cheapsort\u548cquicksort\u4e2d\u9009\u62e9\u7b97\u6cd5\u7684\u4e00\u4e2a\u914d\u7f6e\u3002Mergesort\u662f\u552f\u4e00\u7a33\u5b9a\u7684\u7b97\u6cd5\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\nunsorted_df = pd.DataFrame({'col1':[2,1,1,1],'col2':[1,3,2,4]})\r\nsorted_df = unsorted_df.sort_values(by='col1' ,kind='mergesort')\r\n\r\nprint (sorted_df)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            col1  col2\r\n1     1     3\r\n2     1     2\r\n3     1     4\r\n0     2     1<\/pre>\n
                                            Python Pandas\u5b57\u7b26\u4e32\u548c\u6587\u672c\u6570\u636e<\/h1>\n
                                            \u5728\u672c\u7ae0\u4e2d\uff0c\u6211\u4eec\u5c06\u4f7f\u7528\u57fa\u672c\u7cfb\u5217\/\u7d22\u5f15\u6765\u8ba8\u8bba\u5b57\u7b26\u4e32\u64cd\u4f5c\u3002\u5728\u968f\u540e\u7684\u7ae0\u8282\u4e2d\uff0c\u5c06\u5b66\u4e60\u5982\u4f55\u5c06\u8fd9\u4e9b\u5b57\u7b26\u4e32\u51fd\u6570\u5e94\u7528\u4e8e\u6570\u636e\u5e27(DataFrame<\/em>)\u3002<\/p>\n
                                            Pandas<\/em>\u63d0\u4f9b\u4e86\u4e00\u7ec4\u5b57\u7b26\u4e32\u51fd\u6570\uff0c\u53ef\u4ee5\u65b9\u4fbf\u5730\u5bf9\u5b57\u7b26\u4e32\u6570\u636e\u8fdb\u884c\u64cd\u4f5c\u3002 \u6700\u91cd\u8981\u7684\u662f\uff0c\u8fd9\u4e9b\u51fd\u6570\u5ffd\u7565(\u6216\u6392\u9664)\u4e22\u5931\/NaN\u503c\u3002<\/p>\n
                                            \u51e0\u4e4e\u8fd9\u4e9b\u65b9\u6cd5\u90fd\u4f7f\u7528Python\u5b57\u7b26\u4e32\u51fd\u6570\u3002 \u56e0\u6b64\uff0c\u5c06Series\u5bf9\u8c61\u8f6c\u6362\u4e3aString\u5bf9\u8c61\uff0c\u7136\u540e\u6267\u884c\u8be5\u64cd\u4f5c\u3002<\/p>\n
                                            \u4e0b\u9762\u6765\u770b\u770b\u6bcf\u4e2a\u64cd\u4f5c\u7684\u6267\u884c\u548c\u8bf4\u660e\u3002<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
                                            \u7f16\u53f7<\/th>\n\u51fd\u6570<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                                            1<\/td>\nlower()<\/td>\n\u5c06Series\/Index\u4e2d\u7684\u5b57\u7b26\u4e32\u8f6c\u6362\u4e3a\u5c0f\u5199\u3002<\/td>\n<\/tr>\n
                                            2<\/td>\nupper()<\/td>\n\u5c06Series\/Index\u4e2d\u7684\u5b57\u7b26\u4e32\u8f6c\u6362\u4e3a\u5927\u5199\u3002<\/td>\n<\/tr>\n
                                            3<\/td>\nlen()<\/td>\n\u8ba1\u7b97\u5b57\u7b26\u4e32\u957f\u5ea6\u3002<\/td>\n<\/tr>\n
                                            4<\/td>\nstrip()<\/td>\n\u5e2e\u52a9\u4ece\u4e24\u4fa7\u7684\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u7684\u6bcf\u4e2a\u5b57\u7b26\u4e32\u4e2d\u5220\u9664\u7a7a\u683c(\u5305\u62ec\u6362\u884c\u7b26)\u3002<\/td>\n<\/tr>\n
                                            5<\/td>\nsplit(‘ ‘)<\/td>\n\u7528\u7ed9\u5b9a\u7684\u6a21\u5f0f\u62c6\u5206\u6bcf\u4e2a\u5b57\u7b26\u4e32\u3002<\/td>\n<\/tr>\n
                                            6<\/td>\ncat(sep=’ ‘)<\/td>\n\u4f7f\u7528\u7ed9\u5b9a\u7684\u5206\u9694\u7b26\u8fde\u63a5\u7cfb\u5217\/\u7d22\u5f15\u5143\u7d20\u3002<\/td>\n<\/tr>\n
                                            7<\/td>\nget_dummies()<\/td>\n\u8fd4\u56de\u5177\u6709\u5355\u70ed\u7f16\u7801\u503c\u7684\u6570\u636e\u5e27(DataFrame)\u3002<\/td>\n<\/tr>\n
                                            8<\/td>\ncontains(pattern)<\/td>\n\u5982\u679c\u5143\u7d20\u4e2d\u5305\u542b\u5b50\u5b57\u7b26\u4e32\uff0c\u5219\u8fd4\u56de\u6bcf\u4e2a\u5143\u7d20\u7684\u5e03\u5c14\u503cTrue\uff0c\u5426\u5219\u4e3aFalse\u3002<\/td>\n<\/tr>\n
                                            9<\/td>\nreplace(a,b)<\/td>\n\u5c06\u503ca\u66ff\u6362\u4e3a\u503cb\u3002<\/td>\n<\/tr>\n
                                            10<\/td>\nrepeat(value)<\/td>\n\u91cd\u590d\u6bcf\u4e2a\u5143\u7d20\u6307\u5b9a\u7684\u6b21\u6570\u3002<\/td>\n<\/tr>\n
                                            11<\/td>\ncount(pattern)<\/td>\n\u8fd4\u56de\u6a21\u5f0f\u4e2d\u6bcf\u4e2a\u5143\u7d20\u7684\u51fa\u73b0\u603b\u6570\u3002<\/td>\n<\/tr>\n
                                            12<\/td>\nstartswith(pattern)<\/td>\n\u5982\u679c\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u7684\u5143\u7d20\u4ee5\u6a21\u5f0f\u5f00\u59cb\uff0c\u5219\u8fd4\u56detrue\u3002<\/td>\n<\/tr>\n
                                            13<\/td>\nendswith(pattern)<\/td>\n\u5982\u679c\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u7684\u5143\u7d20\u4ee5\u6a21\u5f0f\u7ed3\u675f\uff0c\u5219\u8fd4\u56detrue\u3002<\/td>\n<\/tr>\n
                                            14<\/td>\nfind(pattern)<\/td>\n\u8fd4\u56de\u6a21\u5f0f\u7b2c\u4e00\u6b21\u51fa\u73b0\u7684\u4f4d\u7f6e\u3002<\/td>\n<\/tr>\n
                                            15<\/td>\nfindall(pattern)<\/td>\n\u8fd4\u56de\u6a21\u5f0f\u7684\u6240\u6709\u51fa\u73b0\u7684\u5217\u8868\u3002<\/td>\n<\/tr>\n
                                            16<\/td>\nswapcase<\/td>\n\u53d8\u6362\u5b57\u6bcd\u5927\u5c0f\u5199\u3002<\/td>\n<\/tr>\n
                                            17<\/td>\nislower()<\/td>\n\u68c0\u67e5\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u6bcf\u4e2a\u5b57\u7b26\u4e32\u4e2d\u7684\u6240\u6709\u5b57\u7b26\u662f\u5426\u5c0f\u5199\uff0c\u8fd4\u56de\u5e03\u5c14\u503c<\/td>\n<\/tr>\n
                                            18<\/td>\nisupper()<\/td>\n\u68c0\u67e5\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u6bcf\u4e2a\u5b57\u7b26\u4e32\u4e2d\u7684\u6240\u6709\u5b57\u7b26\u662f\u5426\u5927\u5199\uff0c\u8fd4\u56de\u5e03\u5c14\u503c<\/td>\n<\/tr>\n
                                            19<\/td>\nisnumeric()<\/td>\n\u68c0\u67e5\u7cfb\u5217\/\u7d22\u5f15\u4e2d\u6bcf\u4e2a\u5b57\u7b26\u4e32\u4e2d\u7684\u6240\u6709\u5b57\u7b26\u662f\u5426\u4e3a\u6570\u5b57\uff0c\u8fd4\u56de\u5e03\u5c14\u503c\u3002<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                            \u73b0\u5728\u521b\u5efa\u4e00\u4e2a\u7cfb\u5217\uff0c\u770b\u770b\u4e0a\u8ff0\u6240\u6709\u51fd\u6570\u662f\u5982\u4f55\u5de5\u4f5c\u7684\u3002<\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t', np.nan, '1234','SteveMinsu'])\r\n\r\nprint (s)\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             Tom\r\n1    William Rick\r\n2            John\r\n3         Alber@t\r\n4             NaN\r\n5            1234\r\n6      SteveMinsu\r\ndtype: object\r\n<\/pre>\n
                                            1. lower()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t', np.nan, '1234','SteveMinsu'])\r\n\r\nprint (s.str.lower())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             tom\r\n1    william rick\r\n2            john\r\n3         alber@t\r\n4             NaN\r\n5            1234\r\n6      steveminsu\r\ndtype: object\r\n<\/pre>\n
                                            2. upper()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t', np.nan, '1234','SteveMinsu'])\r\n\r\nprint (s.str.upper())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             TOM\r\n1    WILLIAM RICK\r\n2            JOHN\r\n3         ALBER@T\r\n4             NaN\r\n5            1234\r\n6      STEVESMITH\r\ndtype: object\r\n<\/pre>\n
                                            3. len()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t', np.nan, '1234','SteveMinsu'])\r\nprint (s.str.len())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0     3.0\r\n1    12.0\r\n2     4.0\r\n3     7.0\r\n4     NaN\r\n5     4.0\r\n6    10.0\r\ndtype: float64\r\n<\/pre>\n
                                            4. strip()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s)\r\nprint (\"=========== After Stripping ================\")\r\nprint (s.str.strip())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             Tom\r\n1     William Rick\r\n2             John\r\n3          Alber@t\r\ndtype: object\r\n=========== After Stripping ================\r\n0             Tom\r\n1    William Rick\r\n2            John\r\n3         Alber@t\r\ndtype: object\r\n<\/pre>\n
                                            5. split(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s)\r\nprint (\"================= Split Pattern: ==================\")\r\nprint (s.str.split(' '))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             Tom\r\n1     William Rick\r\n2             John\r\n3          Alber@t\r\ndtype: object\r\n================= Split Pattern: ==================\r\n0              [Tom, ]\r\n1    [, William, Rick]\r\n2               [John]\r\n3            [Alber@t]\r\ndtype: object\r\n<\/pre>\n
                                            6. cat(sep=pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\n\r\nprint (s.str.cat(sep=' <=> '))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            Tom  <=>  William Rick <=> John <=> Alber@t\r\n<\/pre>\n
                                            7. get_dummies()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\n\r\nprint (s.str.get_dummies())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                 William Rick  Alber@t  John  Tom\r\n0              0        0     0     1\r\n1              1        0     0     0\r\n2              0        0     1     0\r\n3              0        1     0     0\r\n<\/pre>\n
                                            8. contains()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s.str.contains(' '))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0     True\r\n1     True\r\n2    False\r\n3    False\r\ndtype: bool\r\n<\/pre>\n
                                            9. replace(a,b)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s)\r\nprint (\"After replacing @ with $: ============== \")\r\nprint (s.str.replace('@','$'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             Tom\r\n1     William Rick\r\n2             John\r\n3          Alber@t\r\ndtype: object\r\nAfter replacing @ with $: ==============\r\n0             Tom\r\n1     William Rick\r\n2             John\r\n3          Alber$t\r\ndtype: object\r\n<\/pre>\n
                                            10. repeat(value)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\n\r\nprint (s.str.repeat(2))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0                      Tom Tom\r\n1     William Rick William Rick\r\n2                      JohnJohn\r\n3                Alber@tAlber@t\r\ndtype: object\r\n<\/pre>\n
                                            11. count(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\n\r\nprint (\"The number of 'm's in each string:\")\r\nprint (s.str.count('m'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            The number of 'm's in each string:\r\n0    1\r\n1    1\r\n2    0\r\n3    0\r\ndtype: int64\r\n<\/pre>\n
                                            12. startswith(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\n\r\nprint (\"Strings that start with 'T':\")\r\nprint (s.str. startswith ('T'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            Strings that start with 'T':\r\n0     True\r\n1    False\r\n2    False\r\n3    False\r\ndtype: bool\r\n<\/pre>\n
                                            13. endswith(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (\"Strings that end with 't':\")\r\nprint (s.str.endswith('t'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            Strings that end with 't':\r\n0    False\r\n1    False\r\n2    False\r\n3     True\r\ndtype: bool\r\n<\/pre>\n
                                            14. find(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s.str.find('e'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0   -1\r\n1   -1\r\n2   -1\r\n3    3\r\ndtype: int64\r\n<\/pre>\n
                                            \u6ce8\u610f\uff1a-1\u8868\u793a\u5143\u7d20\u4e2d\u6ca1\u6709\u8fd9\u6837\u7684\u6a21\u5f0f\u53ef\u7528\u3002<\/p>\n
                                            15. findall(pattern)\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom ', ' William Rick', 'John', 'Alber@t'])\r\nprint (s.str.findall('e'))\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0     []\r\n1     []\r\n2     []\r\n3    [e]\r\ndtype: object\r\n<\/pre>\n
                                            \u7a7a\u5217\u8868([])\u8868\u793a\u5143\u7d20\u4e2d\u6ca1\u6709\u8fd9\u6837\u7684\u6a21\u5f0f\u53ef\u7528\u3002<\/p>\n
                                            16. swapcase()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t'])\r\nprint (s.str.swapcase())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0             tOM\r\n1    wILLIAM rICK\r\n2            jOHN\r\n3         aLBER@T\r\ndtype: object\r\n<\/pre>\n
                                            17. islower()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['Tom', 'William Rick', 'John', 'Alber@t'])\r\nprint (s.str.islower())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0    False\r\n1    False\r\n2    False\r\n3    False\r\ndtype: bool\r\n<\/pre>\n
                                            18. isupper()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\n\r\ns = pd.Series(['TOM', 'William Rick', 'John', 'Alber@t'])\r\n\r\nprint (s.str.isupper())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0    True\r\n1    False\r\n2    False\r\n3    False\r\ndtype: bool\r\n<\/pre>\n
                                            19. isnumeric()\u51fd\u6570\u793a\u4f8b<\/strong><\/p>\n
                                            import pandas as pd\r\ns = pd.Series(['Tom', '1199','William Rick', 'John', 'Alber@t'])\r\nprint (s.str.isnumeric())\r\n<\/pre>\n
                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                            0    False\r\n1     True\r\n2    False\r\n3    False\r\n4    False\r\ndtype: bool<\/pre>\n
                                            Python Pandas\u9009\u9879\u548c\u81ea\u5b9a\u4e49<\/h1>\n
                                            Pandas\u63d0\u4f9bAPI\u6765\u81ea\u5b9a\u4e49\u5176\u884c\u4e3a\u7684\u67d0\u4e9b\u65b9\u9762\uff0c\u5927\u591a\u4f7f\u7528\u6765\u663e\u793a\u3002<\/p>\n
                                            API\u7531\u4e94\u4e2a\u76f8\u5173\u51fd\u6570\u7ec4\u6210\u3002\u5b83\u4eec\u5206\u522b\u662f –<\/p>\n
                                              \n
                                            • get_option()<\/li>\n
                                            • set_option()<\/li>\n
                                            • reset_option()<\/li>\n
                                            • describe_option()<\/li>\n
                                            • option_context()<\/li>\n<\/ul>\n
                                              \u73b0\u5728\u6765\u4e86\u89e3\u51fd\u6570\u662f\u5982\u4f55\u5de5\u4f5c\u7684\u3002<\/p>\n
                                              get_option(param)<\/h3>\n
                                              get_option(param)\u9700\u8981\u4e00\u4e2a\u53c2\u6570\uff0c\u5e76\u8fd4\u56de\u4e0b\u9762\u8f93\u51fa\u4e2d\u7ed9\u51fa\u7684\u503c –<\/p>\n
                                              get_option\u9700\u8981\u4e00\u4e2a\u53c2\u6570\uff0c\u5e76\u8fd4\u56de\u4e0b\u9762\u8f93\u51fa\u4e2d\u7ed9\u51fa\u7684\u503c –<\/p>\n
                                              display.max_rows<\/strong><\/p>\n
                                              \u663e\u793a\u9ed8\u8ba4\u503c\u3002\u89e3\u91ca\u5668\u8bfb\u53d6\u6b64\u503c\u5e76\u663e\u793a\u6b64\u503c\u4f5c\u4e3a\u663e\u793a\u4e0a\u9650\u7684\u884c\u3002<\/p>\n
                                              import pandas as pd\r\nprint (\"display.max_rows = \", pd.get_option(\"display.max_rows\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              display.max_rows =  60\r\n<\/pre>\n
                                              display.max_columns<\/strong><\/p>\n
                                              \u663e\u793a\u9ed8\u8ba4\u503c\uff0c\u89e3\u91ca\u5668\u8bfb\u53d6\u6b64\u503c\u5e76\u663e\u793a\u6b64\u503c\u4f5c\u4e3a\u663e\u793a\u4e0a\u9650\u7684\u884c\u3002<\/p>\n
                                              import pandas as pd\r\nprint (\"display.max_columns = \", pd.get_option(\"display.max_columns\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              display.max_columns =  20\r\n<\/pre>\n
                                              \u8fd9\u91cc\uff0c60\u548c20\u662f\u9ed8\u8ba4\u914d\u7f6e\u53c2\u6570\u503c\u3002<\/p>\n
                                              set_option(param,value)<\/h3>\n
                                              set_option\u9700\u8981\u4e24\u4e2a\u53c2\u6570\uff0c\u5e76\u5c06\u8be5\u503c\u8bbe\u7f6e\u4e3a\u6307\u5b9a\u7684\u53c2\u6570\u503c\uff0c\u5982\u4e0b\u6240\u793a\uff1a<\/p>\n
                                              display.max_rows<\/strong><\/p>\n
                                              \u4f7f\u7528set_option()\uff0c\u53ef\u4ee5\u66f4\u6539\u8981\u663e\u793a\u7684\u9ed8\u8ba4\u884c\u6570\u3002<\/p>\n
                                              import pandas as pd\r\n\r\nprint (\"before set display.max_rows = \", pd.get_option(\"display.max_rows\"))\r\n\r\npd.set_option(\"display.max_rows\",80)\r\nprint (\"after set display.max_rows = \", pd.get_option(\"display.max_rows\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              before set display.max_rows =  60\r\nafter set display.max_rows =  80\r\n<\/pre>\n
                                              display.max_columns<\/strong><\/p>\n
                                              \u4f7f\u7528set_option()\uff0c\u53ef\u4ee5\u66f4\u6539\u8981\u663e\u793a\u7684\u9ed8\u8ba4\u884c\u6570\u3002<\/p>\n
                                              import pandas as pd\r\n\r\nprint (\"before set display.max_columns = \", pd.get_option(\"display.max_columns\"))\r\n\r\npd.set_option(\"display.max_columns\",32)\r\nprint (\"after set display.max_columns = \", pd.get_option(\"display.max_columns\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              before set display.max_columns =  20\r\nafter set display.max_columns =  32\r\n<\/pre>\n
                                              reset_option(param)<\/h3>\n
                                              reset_option\u63a5\u53d7\u4e00\u4e2a\u53c2\u6570\uff0c\u5e76\u5c06\u8be5\u503c\u8bbe\u7f6e\u4e3a\u9ed8\u8ba4\u503c\u3002<\/p>\n
                                              display.max_rows<\/strong><\/p>\n
                                              \u4f7f\u7528reset_option()\uff0c\u53ef\u4ee5\u5c06\u8be5\u503c\u66f4\u6539\u56de\u663e\u793a\u7684\u9ed8\u8ba4\u884c\u6570\u3002<\/p>\n
                                              import pandas as pd\r\n\r\npd.set_option(\"display.max_rows\",32)\r\nprint (\"after set display.max_rows = \", pd.get_option(\"display.max_rows\"))\r\n\r\npd.reset_option(\"display.max_rows\")\r\nprint (\"reset display.max_rows = \", pd.get_option(\"display.max_rows\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              after set display.max_rows =  32\r\nreset display.max_rows =  60\r\n<\/pre>\n
                                              describe_option(param)<\/h3>\n
                                              describe_option\u6253\u5370\u53c2\u6570\u7684\u63cf\u8ff0\u3002<\/p>\n
                                              display.max_rows<\/strong><\/p>\n
                                              \u4f7f\u7528reset_option()\uff0c\u53ef\u4ee5\u5c06\u8be5\u503c\u66f4\u6539\u56de\u663e\u793a\u7684\u9ed8\u8ba4\u884c\u6570\u3002<\/p>\n
                                              import pandas as pd\r\n\t\r\npd.describe_option(\"display.max_rows\")\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              display.max_rows : int\r\nIf max_rows is exceeded, switch to truncate view. Depending on\r\n`large_repr`, objects are either centrally truncated or printed as\r\na summary view. 'None' value means unlimited.\r\n\r\nIn case python\/IPython is running in a terminal and `large_repr`\r\nequals 'truncate' this can be set to 0 and pandas will auto-detect\r\nthe height of the terminal and print a truncated object which fits\r\nthe screen height. The IPython notebook, IPython qtconsole, or\r\nIDLE do not run in a terminal and hence it is not possible to do\r\ncorrect auto-detection.\r\n[default: 60] [currently: 60]\r\n<\/pre>\n
                                              option_context()<\/h3>\n
                                              option_context\u4e0a\u4e0b\u6587\u7ba1\u7406\u5668\u7528\u4e8e\u4e34\u65f6\u8bbe\u7f6e\u8bed\u53e5\u4e2d\u7684\u9009\u9879\u3002\u5f53\u9000\u51fa\u4f7f\u7528\u5757\u65f6\uff0c\u9009\u9879\u503c\u5c06\u81ea\u52a8\u6062\u590d –<\/p>\n
                                              display.max_rows<\/strong>
                                              \n\u4f7f\u7528option_context()\uff0c\u53ef\u4ee5\u4e34\u65f6\u8bbe\u7f6e\u8be5\u503c\u3002<\/p>\n
                                              import pandas as pd\r\nwith pd.option_context(\"display.max_rows\",10):\r\nprint(pd.get_option(\"display.max_rows\"))\r\nprint(pd.get_option(\"display.max_rows\"))\r\n<\/pre>\n
                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                              10\r\n10\r\n<\/pre>\n
                                              \u8bf7\u53c2\u9605\u7b2c\u4e00\u548c\u7b2c\u4e8c\u4e2a\u6253\u5370\u8bed\u53e5\u4e4b\u95f4\u7684\u533a\u522b\u3002\u7b2c\u4e00\u4e2a\u8bed\u53e5\u6253\u5370\u7531option_context()\u8bbe\u7f6e\u7684\u503c\uff0c\u8be5\u503c\u5728\u4e0a\u4e0b\u6587\u4e2d\u662f\u4e34\u65f6\u7684\u3002\u5728\u4f7f\u7528\u4e0a\u4e0b\u6587\u4e4b\u540e\uff0c\u7b2c\u4e8c\u4e2a\u6253\u5370\u8bed\u53e5\u6253\u5370\u914d\u7f6e\u7684\u503c\u3002<\/p>\n
                                              \u5e38\u7528\u53c2\u6570\uff0c\u8bf7\u53c2\u8003\u4e0b\u8868 –<\/p>\n\n\n\n\n\n\n\n\n
                                              \u7f16\u53f7<\/th>\n\u53c2\u6570<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                                              1<\/td>\ndisplay.max_rows<\/td>\n\u8981\u663e\u793a\u7684\u6700\u5927\u884c\u6570<\/td>\n<\/tr>\n
                                              2<\/td>\ndisplay.max_columns<\/td>\n\u8981\u663e\u793a\u7684\u6700\u5927\u5217\u6570<\/td>\n<\/tr>\n
                                              3<\/td>\ndisplay.expand_frame_repr<\/td>\n\u663e\u793a\u6570\u636e\u5e27\u4ee5\u62c9\u4f38\u9875\u9762<\/td>\n<\/tr>\n
                                              4<\/td>\ndisplay.max_colwidth<\/td>\n\u663e\u793a\u6700\u5927\u5217\u5bbd<\/td>\n<\/tr>\n
                                              5<\/td>\ndisplay.precision<\/td>\n\u663e\u793a\u5341\u8fdb\u5236\u6570\u7684\u7cbe\u5ea6<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                              Python Pandas\u7d22\u5f15\u548c\u9009\u62e9\u6570\u636e<\/h1>\n
                                              \u5728\u672c\u7ae0\u4e2d\uff0c\u6211\u4eec\u5c06\u8ba8\u8bba\u5982\u4f55\u5207\u5272\u548c\u4e22\u5f03\u65e5\u671f\uff0c\u5e76\u83b7\u53d6Pandas\u4e2d\u5927\u5bf9\u8c61\u7684\u5b50\u96c6\u3002<\/p>\n
                                              Python\u548cNumPy\u7d22\u5f15\u8fd0\u7b97\u7b26”[]”\u548c\u5c5e\u6027\u8fd0\u7b97\u7b26”.”\u3002 \u53ef\u4ee5\u5728\u5e7f\u6cdb\u7684\u7528\u4f8b\u4e2d\u5feb\u901f\u8f7b\u677e\u5730\u8bbf\u95eePandas\u6570\u636e\u7ed3\u6784\u3002\u7136\u800c\uff0c\u7531\u4e8e\u8981\u8bbf\u95ee\u7684\u6570\u636e\u7c7b\u578b\u4e0d\u662f\u9884\u5148\u77e5\u9053\u7684\uff0c\u6240\u4ee5\u76f4\u63a5\u4f7f\u7528\u6807\u51c6\u8fd0\u7b97\u7b26\u5177\u6709\u4e00\u4e9b\u4f18\u5316\u9650\u5236\u3002\u5bf9\u4e8e\u751f\u4ea7\u73af\u5883\u7684\u4ee3\u7801\uff0c\u6211\u4eec\u5efa\u8bae\u5229\u7528\u672c\u7ae0\u4ecb\u7ecd\u7684\u4f18\u5316Pandas\u6570\u636e\u8bbf\u95ee\u65b9\u6cd5\u3002<\/p>\n
                                              Pandas\u73b0\u5728\u652f\u6301\u4e09\u79cd\u7c7b\u578b\u7684\u591a\u8f74\u7d22\u5f15; \u8fd9\u4e09\u79cd\u7c7b\u578b\u5728\u4e0b\u8868\u4e2d\u63d0\u5230 –<\/p>\n\n\n\n\n\n\n
                                              \u7f16\u53f7<\/th>\n\u7d22\u5f15<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                                              1<\/td>\n.loc()<\/td>\n\u57fa\u4e8e\u6807\u7b7e<\/td>\n<\/tr>\n
                                              2<\/td>\n.iloc()<\/td>\n\u57fa\u4e8e\u6574\u6570<\/td>\n<\/tr>\n
                                              3<\/td>\n.ix()<\/td>\n\u57fa\u4e8e\u6807\u7b7e\u548c\u6574\u6570<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                              .loc()<\/h3>\n
                                              Pandas\u63d0\u4f9b\u4e86\u5404\u79cd\u65b9\u6cd5\u6765\u5b8c\u6210\u57fa\u4e8e\u6807\u7b7e\u7684\u7d22\u5f15\u3002 \u5207\u7247\u65f6\uff0c\u4e5f\u5305\u62ec\u8d77\u59cb\u8fb9\u754c\u3002\u6574\u6570\u662f\u6709\u6548\u7684\u6807\u7b7e\uff0c\u4f46\u5b83\u4eec\u662f\u6307\u6807\u7b7e\u800c\u4e0d\u662f\u4f4d\u7f6e\u3002<\/p>\n
                                              .loc()\u5177\u6709\u591a\u79cd\u8bbf\u95ee\u65b9\u5f0f\uff0c\u5982 –<\/p>\n
                                                \n
                                              • \u5355\u4e2a\u6807\u91cf\u6807\u7b7e<\/li>\n
                                              • \u6807\u7b7e\u5217\u8868<\/li>\n
                                              • \u5207\u7247\u5bf9\u8c61<\/li>\n
                                              • \u4e00\u4e2a\u5e03\u5c14\u6570\u7ec4<\/li>\n<\/ul>\n
                                                loc\u9700\u8981\u4e24\u4e2a\u5355\/\u5217\u8868\/\u8303\u56f4\u8fd0\u7b97\u7b26\uff0c\u7528”,”\u5206\u9694\u3002\u7b2c\u4e00\u4e2a\u8868\u793a\u884c\uff0c\u7b2c\u4e8c\u4e2a\u8868\u793a\u5217\u3002<\/p>\n
                                                \u793a\u4f8b1<\/strong><\/p>\n
                                                #import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4),\r\nindex = ['a','b','c','d','e','f','g','h'], columns = ['A', 'B', 'C', 'D'])\r\n\r\n#select all rows for a specific column\r\nprint (df.loc[:,'A'])\r\n<\/pre>\n
                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                a    0.015860\r\nb   -0.014135\r\nc    0.446061\r\nd    1.801269\r\ne   -1.404779\r\nf   -0.044016\r\ng    0.996651\r\nh    0.764672\r\nName: A, dtype: float64\r\n<\/pre>\n
                                                \u793a\u4f8b2<\/strong><\/p>\n
                                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4),\r\nindex = ['a','b','c','d','e','f','g','h'], columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Select all rows for multiple columns, say list[]\r\nprint (df.loc[:,['A','C']])\r\n<\/pre>\n
                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                          A         C\r\na -0.529735 -1.067299\r\nb -2.230089 -1.798575\r\nc  0.685852  0.333387\r\nd  1.061853  0.131853\r\ne  0.990459  0.189966\r\nf  0.057314 -0.370055\r\ng  0.453960 -0.624419\r\nh  0.666668 -0.433971\r\n<\/pre>\n
                                                \u793a\u4f8b3<\/strong><\/p>\n
                                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4),\r\nindex = ['a','b','c','d','e','f','g','h'], columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Select few rows for multiple columns, say list[]\r\nprint (df.loc[['a','b','f','h'],['A','C']])\r\n# Select all rows for multiple columns, say list[]\r\nprint (df.loc[:,['A','C']])\r\n<\/pre>\n
                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                          A         C\r\na -1.959731  0.720956\r\nb  1.318976  0.199987\r\nf -1.117735 -0.181116\r\nh -0.147029  0.027369\r\nA         C\r\na -1.959731  0.720956\r\nb  1.318976  0.199987\r\nc  0.839221 -1.611226\r\nd  0.722810  1.649130\r\ne -0.524845 -0.037824\r\nf -1.117735 -0.181116\r\ng -0.642907  0.443261\r\nh -0.147029  0.027369\r\n<\/pre>\n
                                                \u793a\u4f8b4<\/strong><\/p>\n
                                                # import the pandas library and aliasing as pd\r\nimport pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4),\r\nindex = ['a','b','c','d','e','f','g','h'], columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Select range of rows for all columns\r\nprint (df.loc['a':'h'])\r\n<\/pre>\n
                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                          A         B         C         D\r\na  1.556186  1.765712  1.060657  0.810279\r\nb  1.377965 -0.183283 -0.224379  0.963105\r\nc -0.530016  0.167183 -0.066459  0.074198\r\nd -1.515189 -1.453529 -1.559400  1.072148\r\ne -0.487399  0.436143 -1.045622 -0.029507\r\nf  0.552548  0.410745  0.570222 -0.628133\r\ng  0.865293 -0.638388  0.388827 -0.469282\r\nh -0.690596  1.765139 -0.492070 -0.176074\r\n<\/pre>\n
                                                \u793a\u4f8b5<\/strong><\/p>\n
                                                import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4),\r\nindex = ['a','b','c','d','e','f','g','h'], columns = ['A', 'B', 'C', 'D'])\r\n\r\n# for getting values with a boolean array\r\nprint (df.loc['a']>0)\r\n<\/pre>\n
                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                A    False\r\nB     True\r\nC    False\r\nD     True\r\nName: a, dtype: bool\r\n<\/pre>\n
                                                .iloc()<\/h3>\n
                                                Pandas\u63d0\u4f9b\u4e86\u5404\u79cd\u65b9\u6cd5\uff0c\u4ee5\u83b7\u5f97\u7eaf\u6574\u6570\u7d22\u5f15\u3002\u50cfpython\u548cnumpy\u4e00\u6837\uff0c\u7b2c\u4e00\u4e2a\u4f4d\u7f6e\u662f\u57fa\u4e8e0\u7684\u7d22\u5f15\u3002<\/p>\n
                                                \u5404\u79cd\u8bbf\u95ee\u65b9\u5f0f\u5982\u4e0b –<\/p>\n
                                                  \n
                                                • \u6574\u6570<\/li>\n
                                                • \u6574\u6570\u5217\u8868<\/li>\n
                                                • \u7cfb\u5217\u503c<\/li>\n<\/ul>\n
                                                  \u793a\u4f8b1<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\n# select all rows for a specific column\r\nprint (df.iloc[:4])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            A         B         C         D\r\n0  0.277146  0.274234  0.860555 -1.312323\r\n1 -1.064776  2.082030  0.695930  2.409340\r\n2  0.033953 -1.155217  0.113045 -0.028330\r\n3  0.241075 -2.156415  0.939586 -1.670171\r\n<\/pre>\n
                                                  \u793a\u4f8b2<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Integer slicing\r\nprint (df.iloc[:4])\r\nprint (df.iloc[1:5, 2:4])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            A         B         C         D\r\n0  1.346210  0.251839  0.975964  0.319049\r\n1  0.459074  0.038155  0.893615  0.659946\r\n2 -1.097043  0.017080  0.869331 -1.443731\r\n3  1.008033 -0.189436 -0.483688 -1.167312\r\nC         D\r\n1  0.893615  0.659946\r\n2  0.869331 -1.443731\r\n3 -0.483688 -1.167312\r\n4  1.566395 -1.292206\r\n<\/pre>\n
                                                  \u793a\u4f8b3<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Slicing through list of values\r\nprint (df.iloc[[1, 3, 5], [1, 3]])\r\nprint (df.iloc[1:3, :])\r\nprint (df.iloc[:,1:3])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            B         D\r\n1  0.081257  0.009109\r\n3  1.037680 -1.467327\r\n5  1.106721  0.320468\r\nA         B         C         D\r\n1 -0.133711  0.081257 -0.031869  0.009109\r\n2  0.895576 -0.513450 -0.048573  0.698965\r\nB         C\r\n0  0.442735 -0.949859\r\n1  0.081257 -0.031869\r\n2 -0.513450 -0.048573\r\n3  1.037680 -0.801157\r\n4 -0.547456 -0.255016\r\n5  1.106721  0.688142\r\n6 -0.466452  0.219914\r\n7  1.583112  0.982030\r\n<\/pre>\n
                                                  .ix()<\/h3>\n
                                                  \u9664\u4e86\u57fa\u4e8e\u7eaf\u6807\u7b7e\u548c\u6574\u6570\u4e4b\u5916\uff0cPandas\u8fd8\u63d0\u4f9b\u4e86\u4e00\u79cd\u4f7f\u7528.ix()\u8fd0\u7b97\u7b26\u8fdb\u884c\u9009\u62e9\u548c\u5b50\u96c6\u5316\u5bf9\u8c61\u7684\u6df7\u5408\u65b9\u6cd5\u3002<\/p>\n
                                                  \u793a\u4f8b1<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\n# Integer slicing\r\nprint (df.ix[:4])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            A         B         C         D\r\n0 -1.449975 -0.002573  1.349962  0.539765\r\n1 -1.249462 -0.800467  0.483950  0.187853\r\n2  1.361273 -1.893519  0.307613 -0.119003\r\n3 -0.103433 -1.058175 -0.587307 -0.114262\r\n4 -0.612298  0.873136 -0.607457  1.047772\r\n<\/pre>\n
                                                  \u793a\u4f8b2<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n# Index slicing\r\nprint (df.ix[:,'A'])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  0    1.539915\r\n1    1.359477\r\n2    0.239694\r\n3    0.563254\r\n4    2.123950\r\n5    0.341554\r\n6   -0.075717\r\n7   -0.606742\r\nName: A, dtype: float64\r\n<\/pre>\n
                                                  \u4f7f\u7528\u7b26\u53f7<\/h3>\n
                                                  \u4f7f\u7528\u591a\u8f74\u7d22\u5f15\u4ecePandas\u5bf9\u8c61\u83b7\u53d6\u503c\u53ef\u4f7f\u7528\u4ee5\u4e0b\u7b26\u53f7 –<\/p>\n\n\n\n\n\n\n
                                                  \u5bf9\u8c61<\/th>\n\u7d22\u5f15<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                                                  Series<\/td>\ns.loc[indexer]<\/td>\n\u6807\u91cf\u503c<\/td>\n<\/tr>\n
                                                  DataFrame<\/td>\ndf.loc[row_index,col_index]<\/td>\n\u6807\u91cf\u5bf9\u8c61<\/td>\n<\/tr>\n
                                                  Panel<\/td>\np.loc[item_index,major_index, minor_index]<\/td>\np.loc[item_index,major_index, minor_index]<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                  \u6ce8\u610f – .iloc()\u548c.ix()\u5e94\u7528\u76f8\u540c\u7684\u7d22\u5f15\u9009\u9879\u548c\u8fd4\u56de\u503c\u3002<\/p>\n
                                                  \u73b0\u5728\u6765\u770b\u770b\u5982\u4f55\u5728DataFrame\u5bf9\u8c61\u4e0a\u6267\u884c\u6bcf\u4e2a\u64cd\u4f5c\u3002\u8fd9\u91cc\u4f7f\u7528\u57fa\u672c\u7d22\u5f15\u8fd0\u7b97\u7b26[] –<\/p>\n
                                                  \u793a\u4f8b1<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\nprint (df['A'])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  0    0.028277\r\n1   -1.037595\r\n2   -0.563495\r\n3   -1.196961\r\n4   -0.805250\r\n5   -0.911648\r\n6   -0.355171\r\n7   -0.232612\r\nName: A, dtype: float64\r\n<\/pre>\n
                                                  \u793a\u4f8b2<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\nprint (df[['A','B']])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            A         B\r\n0 -0.767339 -0.729411\r\n1 -0.563540 -0.639142\r\n2  0.873589 -2.166382\r\n3  0.900330  0.253875\r\n4 -0.520105  0.064438\r\n5 -1.452176 -0.440864\r\n6 -0.291556 -0.861924\r\n7 -1.464235  0.313168\r\n<\/pre>\n
                                                  \u793a\u4f8b3<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\nprint (df[2:2])\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  Empty DataFrame\r\nColumns: [A, B, C, D]\r\nIndex: []\r\n<\/pre>\n
                                                  \u5c5e\u6027\u8bbf\u95ee<\/h3>\n
                                                  \u53ef\u4ee5\u4f7f\u7528\u5c5e\u6027\u8fd0\u7b97\u7b26.\u6765\u9009\u62e9\u5217\u3002<\/p>\n
                                                  \u793a\u4f8b<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(8, 4), columns = ['A', 'B', 'C', 'D'])\r\n\r\nprint (df.A)\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  0    0.104820\r\n1   -1.206600\r\n2    0.469083\r\n3   -0.821226\r\n4   -1.238865\r\n5    1.083185\r\n6   -0.827833\r\n7   -0.199558\r\nName: A, dtype: float64<\/pre>\n
                                                  Python Pandas\u7edf\u8ba1\u51fd\u6570<\/h1>\n
                                                  \u7edf\u8ba1\u65b9\u6cd5\u6709\u52a9\u4e8e\u7406\u89e3\u548c\u5206\u6790\u6570\u636e\u7684\u884c\u4e3a\u3002\u73b0\u5728\u6211\u4eec\u5c06\u5b66\u4e60\u4e00\u4e9b\u7edf\u8ba1\u51fd\u6570\uff0c\u53ef\u4ee5\u5c06\u8fd9\u4e9b\u51fd\u6570\u5e94\u7528\u5230Pandas<\/em>\u7684\u5bf9\u8c61\u4e0a\u3002<\/p>\n
                                                  pct_change()\u51fd\u6570<\/h3>\n
                                                  \u7cfb\u5217\uff0cDatFrames\u548cPanel\u90fd\u6709pct_change()\u51fd\u6570\u3002\u6b64\u51fd\u6570\u5c06\u6bcf\u4e2a\u5143\u7d20\u4e0e\u5176\u524d\u4e00\u4e2a\u5143\u7d20\u8fdb\u884c\u6bd4\u8f83\uff0c\u5e76\u8ba1\u7b97\u53d8\u5316\u767e\u5206\u6bd4\u3002<\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ns = pd.Series([1,2,3,4,5,4])\r\nprint (s.pct_change())\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 2))\r\nprint (df.pct_change())\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  0        NaN\r\n1   1.000000\r\n2   0.500000\r\n3   0.333333\r\n4   0.250000\r\n5  -0.200000\r\ndtype: float64\r\n\r\n0          1\r\n0         NaN        NaN\r\n1  -15.151902   0.174730\r\n2  -0.746374   -1.449088\r\n3  -3.582229   -3.165836\r\n4   15.601150  -1.860434\r\n<\/pre>\n
                                                  \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0cpct_change()\u5bf9\u5217\u8fdb\u884c\u64cd\u4f5c; \u5982\u679c\u60f3\u5e94\u7528\u5230\u884c\u4e0a\uff0c\u90a3\u4e48\u53ef\u4f7f\u7528axis = 1\u53c2\u6570\u3002<\/p>\n
                                                  \u534f\u65b9\u5dee<\/h3>\n
                                                  \u534f\u65b9\u5dee\u9002\u7528\u4e8e\u7cfb\u5217\u6570\u636e\u3002Series\u5bf9\u8c61\u6709\u4e00\u4e2a\u65b9\u6cd5cov\u7528\u6765\u8ba1\u7b97\u5e8f\u5217\u5bf9\u8c61\u4e4b\u95f4\u7684\u534f\u65b9\u5dee\u3002NA\u5c06\u88ab\u81ea\u52a8\u6392\u9664\u3002<\/p>\n
                                                  Cov\u7cfb\u5217\u793a\u4f8b<\/strong><\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ns1 = pd.Series(np.random.randn(10))\r\ns2 = pd.Series(np.random.randn(10))\r\nprint (s1.cov(s2))\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  0.0667296739178\r\n<\/pre>\n
                                                  \u5f53\u5e94\u7528\u4e8eDataFrame\u65f6\uff0c\u534f\u65b9\u5dee\u65b9\u6cd5\u8ba1\u7b97\u6240\u6709\u5217\u4e4b\u95f4\u7684\u534f\u65b9\u5dee(cov)\u503c\u3002<\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\nframe = pd.DataFrame(np.random.randn(10, 5), columns=['a', 'b', 'c', 'd', 'e'])\r\nprint (frame['a'].cov(frame['b']))\r\nprint (frame.cov())\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  -0.406796939839\r\n\t\t  a         b         c         d         e\r\na  0.784886 -0.406797  0.181312  0.513549 -0.597385\r\nb -0.406797  0.987106 -0.662898 -0.492781  0.388693\r\nc  0.181312 -0.662898  1.450012  0.484724 -0.476961\r\nd  0.513549 -0.492781  0.484724  1.571194 -0.365274\r\ne -0.597385  0.388693 -0.476961 -0.365274  0.785044\r\n<\/pre>\n
                                                  \u6ce8 – \u89c2\u5bdf\u7b2c\u4e00\u4e2a\u8bed\u53e5\u4e2da\u548cb\u5217\u4e4b\u95f4\u7684cov\u7ed3\u679c\u503c\uff0c\u4e0e\u7531DataFrame\u4e0a\u7684cov\u8fd4\u56de\u7684\u503c\u76f8\u540c\u3002<\/p>\n
                                                  \u76f8\u5173\u6027<\/h3>\n
                                                  \u76f8\u5173\u6027\u663e\u793a\u4e86\u4efb\u4f55\u4e24\u4e2a\u6570\u503c(\u7cfb\u5217)\u4e4b\u95f4\u7684\u7ebf\u6027\u5173\u7cfb\u3002\u6709\u591a\u79cd\u65b9\u6cd5\u6765\u8ba1\u7b97pearson(\u9ed8\u8ba4)\uff0cspearman\u548ckendall\u4e4b\u95f4\u7684\u76f8\u5173\u6027\u3002<\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\nframe = pd.DataFrame(np.random.randn(10, 5), columns=['a', 'b', 'c', 'd', 'e'])\r\n\r\nprint (frame['a'].corr(frame['b']))\r\nprint (frame.corr())\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  -0.613999376618\r\n\t\t  a         b         c         d         e\r\na  1.000000 -0.613999 -0.040741 -0.227761 -0.192171\r\nb -0.613999  1.000000  0.012303  0.273584  0.591826\r\nc -0.040741  0.012303  1.000000 -0.391736 -0.470765\r\nd -0.227761  0.273584 -0.391736  1.000000  0.364946\r\ne -0.192171  0.591826 -0.470765  0.364946  1.000000\r\n<\/pre>\n
                                                  \u5982\u679cDataFrame\u4e2d\u5b58\u5728\u4efb\u4f55\u975e\u6570\u5b57\u5217\uff0c\u5219\u4f1a\u81ea\u52a8\u6392\u9664\u3002<\/p>\n
                                                  \u6570\u636e\u6392\u540d<\/h3>\n
                                                  \u6570\u636e\u6392\u540d\u4e3a\u5143\u7d20\u6570\u7ec4\u4e2d\u7684\u6bcf\u4e2a\u5143\u7d20\u751f\u6210\u6392\u540d\u3002\u5728\u5173\u7cfb\u7684\u60c5\u51b5\u4e0b\uff0c\u5206\u914d\u5e73\u5747\u7b49\u7ea7\u3002<\/p>\n
                                                  import pandas as pd\r\nimport numpy as np\r\ns = pd.Series(np.random.np.random.randn(5), index=list('abcde'))\r\n\r\ns['d'] = s['b'] # so there's a tie\r\n\r\nprint (s.rank())\r\n<\/pre>\n
                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                  a    4.0\r\nb    1.5\r\nc    3.0\r\nd    1.5\r\ne    5.0\r\ndtype: float64\r\n<\/pre>\n
                                                  Rank\u53ef\u9009\u5730\u4f7f\u7528\u4e00\u4e2a\u9ed8\u8ba4\u4e3atrue\u7684\u5347\u5e8f\u53c2\u6570; \u5f53\u9519\u8bef\u65f6\uff0c\u6570\u636e\u88ab\u53cd\u5411\u6392\u5e8f\uff0c\u4e5f\u5c31\u662f\u8f83\u5927\u7684\u503c\u88ab\u5206\u914d\u8f83\u5c0f\u7684\u6392\u5e8f\u3002<\/p>\n
                                                  Rank\u652f\u6301\u4e0d\u540c\u7684tie-breaking\u65b9\u6cd5\uff0c\u7528\u65b9\u6cd5\u53c2\u6570\u6307\u5b9a –<\/p>\n
                                                    \n
                                                  • average – \u5e76\u5217\u7ec4\u5e73\u5747\u6392\u5e8f\u7b49\u7ea7<\/li>\n
                                                  • min – \u7ec4\u4e2d\u6700\u4f4e\u7684\u6392\u5e8f\u7b49\u7ea7<\/li>\n
                                                  • max – \u7ec4\u4e2d\u6700\u9ad8\u7684\u6392\u5e8f\u7b49\u7ea7<\/li>\n
                                                  • first – \u6309\u7167\u5b83\u4eec\u51fa\u73b0\u5728\u6570\u7ec4\u4e2d\u7684\u987a\u5e8f\u5206\u914d\u961f\u5217<\/li>\n<\/ul>\n
                                                    Python Pandas\u7a97\u53e3\u51fd\u6570<\/h1>\n
                                                    \u4e3a\u4e86\u5904\u7406\u6570\u5b57\u6570\u636e\uff0cPandas\u63d0\u4f9b\u4e86\u51e0\u4e2a\u53d8\u4f53\uff0c\u5982\u6eda\u52a8\uff0c\u5c55\u5f00\u548c\u6307\u6570\u79fb\u52a8\u7a97\u53e3\u7edf\u8ba1\u7684\u6743\u91cd\u3002 \u5176\u4e2d\u5305\u62ec\u603b\u548c\uff0c\u5747\u503c\uff0c\u4e2d\u4f4d\u6570\uff0c\u65b9\u5dee\uff0c\u534f\u65b9\u5dee\uff0c\u76f8\u5173\u6027\u7b49\u3002<\/p>\n
                                                    \u4e0b\u6765\u5b66\u4e60\u5982\u4f55\u5728DataFrame\u5bf9\u8c61\u4e0a\u5e94\u7528\u4e0a\u63d0\u53ca\u7684\u6bcf\u79cd\u65b9\u6cd5\u3002<\/p>\n
                                                    .rolling()\u51fd\u6570<\/h3>\n
                                                    \u8fd9\u4e2a\u51fd\u6570\u53ef\u4ee5\u5e94\u7528\u4e8e\u4e00\u7cfb\u5217\u6570\u636e\u3002\u6307\u5b9awindow=n\u53c2\u6570\u5e76\u5728\u5176\u4e0a\u5e94\u7528\u9002\u5f53\u7684\u7edf\u8ba1\u51fd\u6570\u3002<\/p>\n
                                                    import pandas as pd\r\n\timport numpy as np\r\n\t\r\n\tdf = pd.DataFrame(np.random.randn(10, 4),\r\n\tindex = pd.date_range('1\/1\/2020', periods=10),\r\n\tcolumns = ['A', 'B', 'C', 'D'])\r\n\t\r\n\tprint (df.rolling(window=3).mean())\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2020-01-01       NaN       NaN       NaN       NaN\r\n2020-01-02       NaN       NaN       NaN       NaN\r\n2020-01-03 -0.306293  0.214001 -0.076004 -0.200793\r\n2020-01-04  0.236632 -0.437033  0.046111 -0.252062\r\n2020-01-05  0.761818 -0.181635 -0.546929 -0.738482\r\n2020-01-06  1.306498 -0.411834 -0.680948 -0.070285\r\n2020-01-07  0.956877 -0.749315 -0.503484  0.160620\r\n2020-01-08  0.354319 -1.067165 -1.238036  1.051048\r\n2020-01-09  0.262081 -0.898373 -1.059351  0.342291\r\n2020-01-10  0.326801 -0.350519 -1.064437  0.749869\r\n<\/pre>\n
                                                    \u6ce8 – \u7531\u4e8e\u7a97\u53e3\u5927\u5c0f\u4e3a3(window)\uff0c\u524d\u4e24\u4e2a\u5143\u7d20\u6709\u7a7a\u503c\uff0c\u7b2c\u4e09\u4e2a\u5143\u7d20\u7684\u503c\u5c06\u662fn\uff0cn-1\u548cn-2\u5143\u7d20\u7684\u5e73\u5747\u503c\u3002\u8fd9\u6837\u4e5f\u53ef\u4ee5\u5e94\u7528\u4e0a\u9762\u63d0\u5230\u7684\u5404\u79cd\u51fd\u6570\u4e86\u3002<\/p>\n
                                                    .expanding()\u51fd\u6570<\/h3>\n
                                                    \u8fd9\u4e2a\u51fd\u6570\u53ef\u4ee5\u5e94\u7528\u4e8e\u4e00\u7cfb\u5217\u6570\u636e\u3002 \u6307\u5b9amin_periods = n\u53c2\u6570\u5e76\u5728\u5176\u4e0a\u5e94\u7528\u9002\u5f53\u7684\u7edf\u8ba1\u51fd\u6570\u3002<\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2018', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df.expanding(min_periods=3).mean())\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2018-01-01       NaN       NaN       NaN       NaN\r\n2018-01-02       NaN       NaN       NaN       NaN\r\n2018-01-03 -0.425085 -0.124270 -0.324134 -0.234001\r\n2018-01-04 -0.293824 -0.038188 -0.172855  0.447226\r\n2018-01-05 -0.516146 -0.013441 -0.384935  0.379267\r\n2018-01-06 -0.614905  0.290308 -0.594635  0.414396\r\n2018-01-07 -0.606090  0.121265 -0.604148  0.246296\r\n2018-01-08 -0.597291  0.075374 -0.425182  0.092831\r\n2018-01-09 -0.380505  0.074956 -0.253081  0.146426\r\n2018-01-10 -0.235030  0.018936 -0.259566  0.315200\r\n<\/pre>\n
                                                    .ewm()\u51fd\u6570<\/h3>\n
                                                    ewm()\u53ef\u5e94\u7528\u4e8e\u7cfb\u5217\u6570\u636e\u3002\u6307\u5b9acom\uff0cspan\uff0chalflife\u53c2\u6570\uff0c\u5e76\u5728\u5176\u4e0a\u5e94\u7528\u9002\u5f53\u7684\u7edf\u8ba1\u51fd\u6570\u3002\u5b83\u4ee5\u6307\u6570\u5f62\u5f0f\u5206\u914d\u6743\u91cd\u3002<\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2019', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df.ewm(com=0.5).mean())\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u51fd\u6570\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2019-01-01  1.047165  0.777385 -1.286948 -0.080564\r\n2019-01-02  0.484093 -0.630998 -0.975172 -0.117832\r\n2019-01-03  0.056189  0.830492  0.116325  1.005547\r\n2019-01-04 -0.363824  1.222173  0.497901 -0.235209\r\n2019-01-05 -0.260685  1.066029  0.391480  1.196190\r\n2019-01-06  0.389649  1.458152 -0.231936 -0.481003\r\n2019-01-07  1.071035 -0.016003  0.387420 -0.170811\r\n2019-01-08 -0.573686  1.052081  1.218439  0.829366\r\n2019-01-09  0.222927  0.556430  0.811838 -0.562096\r\n2019-01-10  0.224624 -1.225446  0.204961 -0.800444\r\n<\/pre>\n
                                                    \u7a97\u53e3\u51fd\u6570\u4e3b\u8981\u7528\u4e8e\u901a\u8fc7\u5e73\u6ed1\u66f2\u7ebf\u6765\u4ee5\u56fe\u5f62\u65b9\u5f0f\u67e5\u627e\u6570\u636e\u5185\u7684\u8d8b\u52bf\u3002\u5982\u679c\u65e5\u5e38\u6570\u636e\u4e2d\u6709\u5f88\u591a\u53d8\u5316\uff0c\u5e76\u4e14\u6709\u5f88\u591a\u6570\u636e\u70b9\u53ef\u7528\uff0c\u90a3\u4e48\u91c7\u6837\u548c\u7ed8\u56fe\u5c31\u662f\u4e00\u79cd\u65b9\u6cd5\uff0c\u5e94\u7528\u7a97\u53e3\u8ba1\u7b97\u5e76\u5728\u7ed3\u679c\u4e0a\u7ed8\u5236\u56fe\u5f62\u662f\u53e6\u4e00\u79cd\u65b9\u6cd5\u3002 \u901a\u8fc7\u8fd9\u4e9b\u65b9\u6cd5\uff0c\u53ef\u4ee5\u5e73\u6ed1\u66f2\u7ebf\u6216\u8d8b\u52bf\u3002<\/p>\n
                                                    Python Pandas\u805a\u5408<\/h1>\n
                                                    \u5f53\u6709\u4e86\u6eda\u52a8\uff0c\u6269\u5c55\u548cewm\u5bf9\u8c61\u521b\u5efa\u4e86\u4ee5\u540e\uff0c\u5c31\u6709\u51e0\u79cd\u65b9\u6cd5\u53ef\u4ee5\u5bf9\u6570\u636e\u6267\u884c\u805a\u5408\u3002<\/p>\n
                                                    DataFrame\u5e94\u7528\u805a\u5408<\/h3>\n
                                                    \u8ba9\u6211\u4eec\u521b\u5efa\u4e00\u4e2aDataFrame\u5e76\u5728\u5176\u4e0a\u5e94\u7528\u805a\u5408\u3002<\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2019', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\n\r\nprint (df)\r\nprint(\"=======================================\")\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r)\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2019-01-01 -0.901602 -1.778484  0.728295 -0.758108\r\n2019-01-02 -0.826162  0.994140  0.976164 -0.918249\r\n2019-01-03  0.260841  0.905993  1.505967 -0.124883\r\n2019-01-04 -0.112230 -0.111885  0.702712 -0.871768\r\n2019-01-05 -0.239969  1.435918 -0.160140 -0.547702\r\n2019-01-06 -0.126897 -2.628206 -0.280658  0.167422\r\n2019-01-07  0.367903  0.994337 -0.529830  0.195990\r\n2019-01-08 -0.530872 -0.384915 -0.397150 -0.024074\r\n2019-01-09 -0.418925  0.049046 -0.816616  0.308107\r\n2019-01-10 -0.176857  2.573145  0.010211 -1.427078\r\n=======================================\r\nRolling [window=3,min_periods=1,center=False,axis=0]\r\n<\/pre>\n
                                                    \u53ef\u4ee5\u901a\u8fc7\u5411\u6574\u4e2aDataFrame\u4f20\u9012\u4e00\u4e2a\u51fd\u6570\u6765\u8fdb\u884c\u805a\u5408\uff0c\u6216\u8005\u901a\u8fc7\u6807\u51c6\u7684\u83b7\u53d6\u9879\u76ee\u65b9\u6cd5\u6765\u9009\u62e9\u4e00\u4e2a\u5217\u3002<\/p>\n
                                                    \u5728\u6574\u4e2a\u6570\u636e\u6846\u4e0a\u5e94\u7528\u805a\u5408<\/h3>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2000', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint df\r\n\r\nr = df.rolling(window=3,min_periods=1)\r\nprint r.aggregate(np.sum)\r\n<\/pre>\n
                                                    \u6267\u884c\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2020-01-01  1.069090 -0.802365 -0.323818 -1.994676\r\n2020-01-02  0.190584  0.328272 -0.550378  0.559738\r\n2020-01-03  0.044865  0.478342 -0.976129  0.106530\r\n2020-01-04 -1.349188 -0.391635 -0.292740  1.412755\r\n2020-01-05  0.057659 -1.331901 -0.297858 -0.500705\r\n2020-01-06  2.651680 -1.459706 -0.726023  0.294283\r\n2020-01-07  0.666481  0.679205 -1.511743  2.093833\r\n2020-01-08 -0.284316 -1.079759  1.433632  0.534043\r\n2020-01-09  1.115246 -0.268812  0.190440 -0.712032\r\n2020-01-10 -0.121008  0.136952  1.279354  0.275773\r\n============================================\r\n\t\t\t\t   A         B         C         D\r\n2020-01-01  1.069090 -0.802365 -0.323818 -1.994676\r\n2020-01-02  1.259674 -0.474093 -0.874197 -1.434938\r\n2020-01-03  1.304539  0.004249 -1.850326 -1.328409\r\n2020-01-04 -1.113739  0.414979 -1.819248  2.079023\r\n2020-01-05 -1.246664 -1.245194 -1.566728  1.018580\r\n2020-01-06  1.360151 -3.183242 -1.316621  1.206333\r\n2020-01-07  3.375821 -2.112402 -2.535624  1.887411\r\n2020-01-08  3.033846 -1.860260 -0.804134  2.922160\r\n2020-01-09  1.497411 -0.669366  0.112329  1.915845\r\n2020-01-10  0.709922 -1.211619  2.903427  0.097785\r\n<\/pre>\n
                                                    \u5728\u6570\u636e\u6846\u7684\u5355\u4e2a\u5217\u4e0a\u5e94\u7528\u805a\u5408<\/strong><\/p>\n
                                                    \u793a\u4f8b\u4ee3\u7801<\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2000', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df)\r\nprint(\"====================================\")\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r['A'].aggregate(np.sum))\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2000-01-01 -1.095530 -0.415257 -0.446871 -1.267795\r\n2000-01-02 -0.405793 -0.002723  0.040241 -0.131678\r\n2000-01-03 -0.136526  0.742393 -0.692582 -0.271176\r\n2000-01-04  0.318300 -0.592146 -0.754830  0.239841\r\n2000-01-05 -0.125770  0.849980  0.685083  0.752720\r\n2000-01-06  1.410294  0.054780  0.297992 -0.034028\r\n2000-01-07  0.463223 -1.239204 -0.056420  0.440893\r\n2000-01-08 -2.244446 -0.516937 -2.039601 -0.680606\r\n2000-01-09  0.991139  0.026987 -2.391856  0.585565\r\n2000-01-10  0.112228 -0.701284 -1.139827  1.484032\r\n====================================\r\n2000-01-01   -1.095530\r\n2000-01-02   -1.501323\r\n2000-01-03   -1.637848\r\n2000-01-04   -0.224018\r\n2000-01-05    0.056004\r\n2000-01-06    1.602824\r\n2000-01-07    1.747747\r\n2000-01-08   -0.370928\r\n2000-01-09   -0.790084\r\n2000-01-10   -1.141079\r\nFreq: D, Name: A, dtype: float64\r\n<\/pre>\n
                                                    \u5728DataFrame\u7684\u591a\u5217\u4e0a\u5e94\u7528\u805a\u5408<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('1\/1\/2018', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df)\r\nprint (\"==========================================\")\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r[['A','B']].aggregate(np.sum))\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2018-01-01  0.518897  0.988917  0.435691 -1.005703\r\n2018-01-02  1.793400  0.130314  2.313787  0.870057\r\n2018-01-03 -0.297601  0.504137 -0.951311 -0.146720\r\n2018-01-04  0.282177  0.142360 -0.059013  0.633174\r\n2018-01-05  2.095398 -0.153359  0.431514 -1.185657\r\n2018-01-06  0.134847  0.188138  0.828329 -1.035120\r\n2018-01-07  0.780541  0.138942 -1.001229  0.714896\r\n2018-01-08  0.579742 -0.642858  0.835013 -1.504110\r\n2018-01-09 -1.692986 -0.861327 -1.125359  0.006687\r\n2018-01-10 -0.263689  1.182349 -0.916569  0.617476\r\n==========================================\r\n\t\t\t\t   A         B\r\n2018-01-01  0.518897  0.988917\r\n2018-01-02  2.312297  1.119232\r\n2018-01-03  2.014697  1.623369\r\n2018-01-04  1.777976  0.776811\r\n2018-01-05  2.079975  0.493138\r\n2018-01-06  2.512422  0.177140\r\n2018-01-07  3.010786  0.173722\r\n2018-01-08  1.495130 -0.315777\r\n2018-01-09 -0.332703 -1.365242\r\n2018-01-10 -1.376932 -0.321836\r\n<\/pre>\n
                                                    \u5728DataFrame\u7684\u5355\u4e2a\u5217\u4e0a\u5e94\u7528\u591a\u4e2a\u51fd\u6570<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('2019\/01\/01', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df)\r\n\r\nprint(\"==========================================\")\r\n\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r['A'].aggregate([np.sum,np.mean]))\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                       A         B         C         D\r\n2019-01-01  1.022641 -1.431910  0.780941 -0.029811\r\n2019-01-02 -0.302858  0.009886 -0.359331 -0.417708\r\n2019-01-03 -1.396564  0.944374 -0.238989 -1.873611\r\n2019-01-04  0.396995 -1.152009 -0.560552 -0.144212\r\n2019-01-05 -2.513289 -1.085277 -1.016419 -1.586994\r\n2019-01-06 -0.513179  0.823411  0.670734  1.196546\r\n2019-01-07 -0.363239 -0.991799  0.587564 -1.100096\r\n2019-01-08  1.474317  1.265496 -0.216486 -0.224218\r\n2019-01-09  2.235798 -1.381457 -0.950745 -0.209564\r\n2019-01-10 -0.061891 -0.025342  0.494245 -0.081681\r\n==========================================\r\n\t\t\t\t sum      mean\r\n2019-01-01  1.022641  1.022641\r\n2019-01-02  0.719784  0.359892\r\n2019-01-03 -0.676780 -0.225593\r\n2019-01-04 -1.302427 -0.434142\r\n2019-01-05 -3.512859 -1.170953\r\n2019-01-06 -2.629473 -0.876491\r\n2019-01-07 -3.389707 -1.129902\r\n2019-01-08  0.597899  0.199300\r\n2019-01-09  3.346876  1.115625\r\n2019-01-10  3.648224  1.216075\r\n<\/pre>\n
                                                    \u5728DataFrame\u7684\u591a\u5217\u4e0a\u5e94\u7528\u591a\u4e2a\u51fd\u6570<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10, 4),\r\nindex = pd.date_range('2020\/01\/01', periods=10),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\n\r\nprint (df)\r\nprint(\"==========================================\")\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r[['A','B']].aggregate([np.sum,np.mean]))\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      A         B         C         D\r\n2020-01-01  1.053702  0.355985  0.746638 -0.233968\r\n2020-01-02  0.578520 -1.171843 -1.764249 -0.709913\r\n2020-01-03 -0.491185  0.975212  0.200139 -3.372621\r\n2020-01-04 -1.331328  0.776316  0.216623  0.202313\r\n2020-01-05 -1.023147 -0.913686  1.457512  0.999232\r\n2020-01-06  0.995328 -0.979826 -1.063695  0.057925\r\n2020-01-07  0.576668  1.065767 -0.270744 -0.513707\r\n2020-01-08  0.520258  0.969043 -0.119177 -0.125620\r\n2020-01-09 -0.316480  0.549085  1.862249  1.091265\r\n2020-01-10  0.461321 -0.368662 -0.988323  0.543011\r\n==========================================\r\n\t\t\t\t\t\t\t A                   B\r\n\t\t\t\t sum      mean       sum      mean\r\n2020-01-01  1.053702  1.053702  0.355985  0.355985\r\n2020-01-02  1.632221  0.816111 -0.815858 -0.407929\r\n2020-01-03  1.141037  0.380346  0.159354  0.053118\r\n2020-01-04 -1.243993 -0.414664  0.579686  0.193229\r\n2020-01-05 -2.845659 -0.948553  0.837843  0.279281\r\n2020-01-06 -1.359146 -0.453049 -1.117195 -0.372398\r\n2020-01-07  0.548849  0.182950 -0.827744 -0.275915\r\n2020-01-08  2.092254  0.697418  1.054985  0.351662\r\n2020-01-09  0.780445  0.260148  2.583896  0.861299\r\n2020-01-10  0.665099  0.221700  1.149466  0.383155\r\n<\/pre>\n
                                                    \u5c06\u4e0d\u540c\u7684\u51fd\u6570\u5e94\u7528\u4e8eDataFrame\u7684\u4e0d\u540c\u5217<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(3, 4),\r\nindex = pd.date_range('2020\/01\/01', periods=3),\r\ncolumns = ['A', 'B', 'C', 'D'])\r\nprint (df)\r\nprint(\"==========================================\")\r\nr = df.rolling(window=3,min_periods=1)\r\nprint (r.aggregate({'A' : np.sum,'B' : np.mean}))\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      A         B         C         D\r\n2020-01-01 -0.246302 -0.057202  0.923807 -1.019698\r\n2020-01-02  0.285287  1.467206 -0.368735 -0.397260\r\n2020-01-03 -0.163219 -0.401368  1.254569  0.580188\r\n==========================================\r\n\t\t\t\t   A         B\r\n2020-01-01 -0.246302 -0.057202\r\n2020-01-02  0.038985  0.705002\r\n2020-01-03 -0.124234  0.336212<\/pre>\n
                                                    Python Pandas\u7f3a\u5931\u6570\u636e<\/h1>\n
                                                    \u6570\u636e\u4e22\u5931(\u7f3a\u5931)\u5728\u73b0\u5b9e\u751f\u6d3b\u4e2d\u603b\u662f\u4e00\u4e2a\u95ee\u9898\u3002 \u673a\u5668\u5b66\u4e60\u548c\u6570\u636e\u6316\u6398\u7b49\u9886\u57df\u7531\u4e8e\u6570\u636e\u7f3a\u5931\u5bfc\u81f4\u7684\u6570\u636e\u8d28\u91cf\u5dee\uff0c\u5728\u6a21\u578b\u9884\u6d4b\u7684\u51c6\u786e\u6027\u4e0a\u9762\u4e34\u7740\u4e25\u91cd\u7684\u95ee\u9898\u3002 \u5728\u8fd9\u4e9b\u9886\u57df\uff0c\u7f3a\u5931\u503c\u5904\u7406\u662f\u4f7f\u6a21\u578b\u66f4\u52a0\u51c6\u786e\u548c\u6709\u6548\u7684\u91cd\u70b9\u3002<\/p>\n
                                                    \u4f55\u65f6\u4ee5\u53ca\u4e3a\u4ec0\u4e48\u6570\u636e\u4e22\u5931\uff1f<\/h3>\n
                                                    \u60f3\u8c61\u4e00\u4e0b\u6709\u4e00\u4e2a\u4ea7\u54c1\u7684\u5728\u7ebf\u8c03\u67e5\u3002\u5f88\u591a\u65f6\u5019\uff0c\u4eba\u4eec\u4e0d\u4f1a\u5206\u4eab\u4e0e\u4ed6\u4eec\u6709\u5173\u7684\u6240\u6709\u4fe1\u606f\u3002 \u5f88\u5c11\u6709\u4eba\u5206\u4eab\u4ed6\u4eec\u7684\u7ecf\u9a8c\uff0c\u4f46\u4e0d\u662f\u4ed6\u4eec\u4f7f\u7528\u4ea7\u54c1\u591a\u4e45; \u5f88\u5c11\u6709\u4eba\u5206\u4eab\u4f7f\u7528\u4ea7\u54c1\u7684\u65f6\u95f4\uff0c\u7ecf\u9a8c\uff0c\u4f46\u4e0d\u662f\u4ed6\u4eec\u7684\u4e2a\u4eba\u8054\u7cfb\u4fe1\u606f\u3002 \u56e0\u6b64\uff0c\u4ee5\u67d0\u79cd\u65b9\u5f0f\u6216\u5176\u4ed6\u65b9\u5f0f\uff0c\u603b\u4f1a\u6709\u4e00\u90e8\u5206\u6570\u636e\u603b\u662f\u4f1a\u4e22\u5931\uff0c\u8fd9\u662f\u975e\u5e38\u5e38\u89c1\u7684\u73b0\u8c61\u3002<\/p>\n
                                                    \u73b0\u5728\u6765\u770b\u770b\u5982\u4f55\u5904\u7406\u4f7f\u7528Pandas\u7684\u7f3a\u5931\u503c(\u5982NA\u6216NaN)\u3002<\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\n\r\nprint (df)\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            one       two     three\r\na  0.691764 -0.118095 -0.950871\r\nb       NaN       NaN       NaN\r\nc -0.886898  0.053705 -1.269253\r\nd       NaN       NaN       NaN\r\ne -0.344967 -0.837128  0.730831\r\nf -1.193740  1.767796  0.888104\r\ng       NaN       NaN       NaN\r\nh -0.755934 -1.331638  0.272248\r\n<\/pre>\n
                                                    \u4f7f\u7528\u91cd\u6784\u7d22\u5f15(reindexing)\uff0c\u521b\u5efa\u4e86\u4e00\u4e2a\u7f3a\u5c11\u503c\u7684DataFrame\u3002 \u5728\u8f93\u51fa\u4e2d\uff0cNaN\u8868\u793a\u4e0d\u662f\u6570\u5b57\u7684\u503c\u3002<\/p>\n
                                                    \u68c0\u67e5\u7f3a\u5931\u503c<\/h3>\n
                                                    \u4e3a\u4e86\u66f4\u5bb9\u6613\u5730\u68c0\u6d4b\u7f3a\u5931\u503c(\u4ee5\u53ca\u8de8\u8d8a\u4e0d\u540c\u7684\u6570\u7ec4dtype)\uff0cPandas\u63d0\u4f9b\u4e86isnull()\u548cnotnull()\u51fd\u6570\uff0c\u5b83\u4eec\u4e5f\u662fSeries\u548cDataFrame\u5bf9\u8c61\u7684\u65b9\u6cd5 –<\/p>\n
                                                    \u793a\u4f8b1<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\n\r\nprint (df['one'].isnull())\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                    a    False\r\nb     True\r\nc    False\r\nd     True\r\ne    False\r\nf    False\r\ng     True\r\nh    False\r\nName: one, dtype: bool\r\n<\/pre>\n
                                                    \u793a\u4f8b2<\/strong><\/p>\n
                                                    import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\n\r\nprint (df['one'].notnull())\r\n<\/pre>\n
                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                    a     True\r\nb    False\r\nc     True\r\nd    False\r\ne     True\r\nf     True\r\ng    False\r\nh     True\r\nName: one, dtype: bool\r\n<\/pre>\n
                                                    \u7f3a\u5c11\u6570\u636e\u7684\u8ba1\u7b97<\/h4>\n
                                                      \n
                                                    • \u5728\u6c42\u548c\u6570\u636e\u65f6\uff0cNA\u5c06\u88ab\u89c6\u4e3a0<\/li>\n
                                                    • \u5982\u679c\u6570\u636e\u5168\u90e8\u662fNA\uff0c\u90a3\u4e48\u7ed3\u679c\u5c06\u662fNA<\/li>\n<\/ul>\n
                                                      \u5b9e\u4f8b1<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\n\r\nprint (df['one'].sum())\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                      -2.6163354325445014\r\n<\/pre>\n
                                                      \u793a\u4f8b2<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(index=[0,1,2,3,4,5],columns=['one','two'])\r\nprint (df['one'].sum())\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                      nan\r\n<\/pre>\n
                                                      \u6e05\u7406\/\u586b\u5145\u7f3a\u5c11\u6570\u636e<\/h3>\n
                                                      Pandas\u63d0\u4f9b\u4e86\u5404\u79cd\u65b9\u6cd5\u6765\u6e05\u9664\u7f3a\u5931\u7684\u503c\u3002fillna()\u51fd\u6570\u53ef\u4ee5\u901a\u8fc7\u51e0\u79cd\u65b9\u6cd5\u7528\u975e\u7a7a\u6570\u636e\u201c\u586b\u5145\u201dNA\u503c\uff0c\u5728\u4e0b\u9762\u7684\u7ae0\u8282\u4e2d\u5c06\u5b66\u4e60\u548c\u4f7f\u7528\u3002<\/p>\n
                                                      \u7528\u6807\u91cf\u503c\u66ff\u6362NaN<\/h3>\n
                                                      \u4ee5\u4e0b\u7a0b\u5e8f\u663e\u793a\u5982\u4f55\u75280\u66ff\u6362NaN\u3002<\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(3, 3), index=['a', 'c', 'e'],columns=['one',\r\n'two', 'three'])\r\ndf = df.reindex(['a', 'b', 'c'])\r\nprint (df)\r\nprint (\"NaN replaced with '0':\")\r\nprint (df.fillna(0))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                              one       two     three\r\na -0.479425 -1.711840 -1.453384\r\nb       NaN       NaN       NaN\r\nc -0.733606 -0.813315  0.476788\r\nNaN replaced with '0':\r\none       two     three\r\na -0.479425 -1.711840 -1.453384\r\nb  0.000000  0.000000  0.000000\r\nc -0.733606 -0.813315  0.476788\r\n<\/pre>\n
                                                      \u5728\u8fd9\u91cc\u586b\u5145\u96f6\u503c; \u5f53\u7136\uff0c\u4e5f\u53ef\u4ee5\u586b\u5199\u4efb\u4f55\u5176\u4ed6\u7684\u503c\u3002<\/p>\n
                                                      \u586b\u5199NA\u524d\u8fdb\u548c\u540e\u9000<\/h3>\n
                                                      \u4f7f\u7528\u91cd\u6784\u7d22\u5f15\u7ae0\u8282\u8ba8\u8bba\u7684\u586b\u5145\u6982\u5ff5\uff0c\u6765\u586b\u8865\u7f3a\u5931\u7684\u503c\u3002<\/p>\n\n\n\n\n\n
                                                      \u65b9\u6cd5<\/th>\n\u52a8\u4f5c<\/th>\n<\/tr>\n
                                                      pad\/fill<\/td>\n\u586b\u5145\u65b9\u6cd5\u5411\u524d<\/td>\n<\/tr>\n
                                                      bfill\/backfill<\/td>\n\u586b\u5145\u65b9\u6cd5\u5411\u540e<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                      \u793a\u4f8b1<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\n\r\nprint (df.fillna(method='pad'))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                              one       two     three\r\na  0.614938 -0.452498 -2.113057\r\nb  0.614938 -0.452498 -2.113057\r\nc -0.118390  1.333962 -0.037907\r\nd -0.118390  1.333962 -0.037907\r\ne  0.699733  0.502142 -0.243700\r\nf  0.544225 -0.923116 -1.123218\r\ng  0.544225 -0.923116 -1.123218\r\nh -0.669783  1.187865  1.112835\r\n<\/pre>\n
                                                      \u793a\u4f8b2<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\nprint (df.fillna(method='backfill'))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                              one       two     three\r\na  2.278454  1.550483 -2.103731\r\nb -0.779530  0.408493  1.247796\r\nc -0.779530  0.408493  1.247796\r\nd  0.262713 -1.073215  0.129808\r\ne  0.262713 -1.073215  0.129808\r\nf -0.600729  1.310515 -0.877586\r\ng  0.395212  0.219146 -0.175024\r\nh  0.395212  0.219146 -0.175024\r\n<\/pre>\n
                                                      \u4e22\u5931\u7f3a\u5c11\u7684\u503c<\/h3>\n
                                                      \u5982\u679c\u53ea\u60f3\u6392\u9664\u7f3a\u5c11\u7684\u503c\uff0c\u5219\u4f7f\u7528dropna\u51fd\u6570\u548caxis\u53c2\u6570\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0caxis = 0\uff0c\u5373\u5728\u884c\u4e0a\u5e94\u7528\uff0c\u8fd9\u610f\u5473\u7740\u5982\u679c\u884c\u5185\u7684\u4efb\u4f55\u503c\u662fNA\uff0c\u90a3\u4e48\u6574\u4e2a\u884c\u88ab\u6392\u9664\u3002<\/p>\n
                                                      \u5b9e\u4f8b1<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\nprint (df.dropna())\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                              one       two     three\r\na -0.719623  0.028103 -1.093178\r\nc  0.040312  1.729596  0.451805\r\ne -1.029418  1.920933  1.289485\r\nf  1.217967  1.368064  0.527406\r\nh  0.667855  0.147989 -1.035978\r\n<\/pre>\n
                                                      \u793a\u4f8b2<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(5, 3), index=['a', 'c', 'e', 'f',\r\n'h'],columns=['one', 'two', 'three'])\r\n\r\ndf = df.reindex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h'])\r\nprint (df.dropna(axis=1))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                      Empty DataFrame\r\nColumns: []\r\nIndex: [a, b, c, d, e, f, g, h]\r\n<\/pre>\n
                                                      \u66ff\u6362\u4e22\u5931(\u6216)\u901a\u7528\u503c<\/h3>\n
                                                      \u5f88\u591a\u65f6\u5019\uff0c\u5fc5\u987b\u7528\u4e00\u4e9b\u5177\u4f53\u7684\u503c\u53d6\u4ee3\u4e00\u4e2a\u901a\u7528\u7684\u503c\u3002\u53ef\u4ee5\u901a\u8fc7\u5e94\u7528\u66ff\u6362\u65b9\u6cd5\u6765\u5b9e\u73b0\u8fd9\u4e00\u70b9\u3002<\/p>\n
                                                      \u7528\u6807\u91cf\u503c\u66ff\u6362NA\u662ffillna()\u51fd\u6570\u7684\u7b49\u6548\u884c\u4e3a\u3002<\/p>\n
                                                      \u793a\u4f8b1<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame({'one':[10,20,30,40,50,2000],\r\n'two':[1000,0,30,40,50,60]})\r\nprint (df.replace({1000:10,2000:60}))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                         one  two\r\n0   10   10\r\n1   20    0\r\n2   30   30\r\n3   40   40\r\n4   50   50\r\n5   60   60\r\n<\/pre>\n
                                                      \u793a\u4f8b2<\/strong><\/p>\n
                                                      import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame({'one':[10,20,30,40,50,2000],\r\n'two':[1000,0,30,40,50,60]})\r\nprint (df.replace({1000:10,2000:60}))\r\n<\/pre>\n
                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                         one  two\r\n0   10   10\r\n1   20    0\r\n2   30   30\r\n3   40   40\r\n4   50   50\r\n5   60   60<\/pre>\n
                                                      Python Pandas\u5206\u7ec4<\/h1>\n
                                                      \u4efb\u4f55\u5206\u7ec4(groupby)\u64cd\u4f5c\u90fd\u6d89\u53ca\u539f\u59cb\u5bf9\u8c61\u7684\u4ee5\u4e0b\u64cd\u4f5c\u4e4b\u4e00\u3002\u5b83\u4eec\u662f –<\/p>\n
                                                        \n
                                                      • \u5206\u5272\u5bf9\u8c61<\/li>\n
                                                      • \u5e94\u7528\u4e00\u4e2a\u51fd\u6570<\/li>\n
                                                      • \u7ed3\u5408\u7684\u7ed3\u679c<\/li>\n<\/ul>\n
                                                        \u5728\u8bb8\u591a\u60c5\u51b5\u4e0b\uff0c\u6211\u4eec\u5c06\u6570\u636e\u5206\u6210\u591a\u4e2a\u96c6\u5408\uff0c\u5e76\u5728\u6bcf\u4e2a\u5b50\u96c6\u4e0a\u5e94\u7528\u4e00\u4e9b\u51fd\u6570\u3002\u5728\u5e94\u7528\u51fd\u6570\u4e2d\uff0c\u53ef\u4ee5\u6267\u884c\u4ee5\u4e0b\u64cd\u4f5c –<\/p>\n
                                                          \n
                                                        • \u805a\u5408 – \u8ba1\u7b97\u6c47\u603b\u7edf\u8ba1<\/li>\n
                                                        • \u8f6c\u6362 – \u6267\u884c\u4e00\u4e9b\u7279\u5b9a\u4e8e\u7ec4\u7684\u64cd\u4f5c<\/li>\n
                                                        • \u8fc7\u6ee4 – \u5728\u67d0\u4e9b\u60c5\u51b5\u4e0b\u4e22\u5f03\u6570\u636e<\/li>\n<\/ul>\n
                                                          \u4e0b\u9762\u6765\u770b\u770b\u521b\u5efa\u4e00\u4e2aDataFrame\u5bf9\u8c61\u5e76\u5bf9\u5176\u6267\u884c\u6240\u6709\u64cd\u4f5c –<\/p>\n
                                                          import pandas as pd\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\nprint (df)\r\n<\/pre>\n
                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                              Points  Rank    Team  Year\r\n0      876     1  Riders  2014\r\n1      789     2  Riders  2015\r\n2      863     2  Devils  2014\r\n3      673     3  Devils  2015\r\n4      741     3   Kings  2014\r\n5      812     4   kings  2015\r\n6      756     1   Kings  2016\r\n7      788     1   Kings  2017\r\n8      694     2  Riders  2016\r\n9      701     4  Royals  2014\r\n10     804     1  Royals  2015\r\n11     690     2  Riders  2017\r\n<\/pre>\n
                                                          \u5c06\u6570\u636e\u62c6\u5206\u6210\u7ec4<\/h3>\n
                                                          Pandas\u5bf9\u8c61\u53ef\u4ee5\u5206\u6210\u4efb\u4f55\u5bf9\u8c61\u3002\u6709\u591a\u79cd\u65b9\u5f0f\u6765\u62c6\u5206\u5bf9\u8c61\uff0c\u5982 –<\/p>\n
                                                            \n
                                                          • obj.groupby(\u2018key\u2019)<\/li>\n
                                                          • obj.groupby([\u2018key1\u2019,\u2019key2\u2019])<\/li>\n
                                                          • obj.groupby(key,axis=1)<\/li>\n<\/ul>\n
                                                            \u73b0\u5728\u6765\u770b\u770b\u5982\u4f55\u5c06\u5206\u7ec4\u5bf9\u8c61\u5e94\u7528\u4e8eDataFrame\u5bf9\u8c61<\/p>\n
                                                            \u793a\u4f8b<\/strong><\/p>\n
                                                            import pandas as pd\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\nprint (df.groupby('Team'))\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            <pandas.core.groupby.DataFrameGroupBy object at 0x00000245D60AD518>\r\n<\/pre>\n
                                                            \u67e5\u770b\u5206\u7ec4<\/h3>\n
                                                            import pandas as pd\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],           'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\nprint (df.groupby('Team').groups)\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            {\r\n'Devils': Int64Index([2, 3], dtype='int64'),\r\n'Kings': Int64Index([4, 6, 7], dtype='int64'),\r\n'Riders': Int64Index([0, 1, 8, 11], dtype='int64'),\r\n'Royals': Int64Index([9, 10], dtype='int64'),\r\n'kings': Int64Index([5], dtype='int64')\r\n}\r\n<\/pre>\n
                                                            \u793a\u4f8b<\/strong><\/p>\n
                                                            \u6309\u591a\u5217\u5206\u7ec4 –<\/p>\n
                                                            import pandas as pd\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\nprint (df.groupby(['Team','Year']).groups)\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            {\r\n('Devils', 2014): Int64Index([2], dtype='int64'),\r\n('Devils', 2015): Int64Index([3], dtype='int64'),\r\n('Kings', 2014): Int64Index([4], dtype='int64'),\r\n('Kings', 2016): Int64Index([6], dtype='int64'),\r\n('Kings', 2017): Int64Index([7], dtype='int64'),\r\n('Riders', 2014): Int64Index([0], dtype='int64'),\r\n('Riders', 2015): Int64Index([1], dtype='int64'),\r\n('Riders', 2016): Int64Index([8], dtype='int64'),\r\n('Riders', 2017): Int64Index([11], dtype='int64'),\r\n('Royals', 2014): Int64Index([9], dtype='int64'),\r\n('Royals', 2015): Int64Index([10], dtype='int64'),\r\n('kings', 2015): Int64Index([5], dtype='int64')\r\n}\r\n<\/pre>\n
                                                            \u8fed\u4ee3\u904d\u5386\u5206\u7ec4<\/h3>\n
                                                            \u4f7f\u7528groupby\u5bf9\u8c61\uff0c\u53ef\u4ee5\u904d\u5386\u7c7b\u4f3citertools.obj\u7684\u5bf9\u8c61\u3002<\/p>\n
                                                            import pandas as pd\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\ngrouped = df.groupby('Year')\r\n\r\nfor name,group in grouped:\r\nprint (name)\r\nprint (group)\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            2014\r\nPoints  Rank    Team  Year\r\n0     876     1  Riders  2014\r\n2     863     2  Devils  2014\r\n4     741     3   Kings  2014\r\n9     701     4  Royals  2014\r\n2015\r\nPoints  Rank    Team  Year\r\n1      789     2  Riders  2015\r\n3      673     3  Devils  2015\r\n5      812     4   kings  2015\r\n10     804     1  Royals  2015\r\n2016\r\nPoints  Rank    Team  Year\r\n6     756     1   Kings  2016\r\n8     694     2  Riders  2016\r\n2017\r\nPoints  Rank    Team  Year\r\n7      788     1   Kings  2017\r\n11     690     2  Riders  2017\r\n<\/pre>\n
                                                            \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0cgroupby\u5bf9\u8c61\u5177\u6709\u4e0e\u5206\u7ec4\u540d\u76f8\u540c\u7684\u6807\u7b7e\u540d\u79f0\u3002<\/p>\n
                                                            \u9009\u62e9\u4e00\u4e2a\u5206\u7ec4<\/h3>\n
                                                            \u4f7f\u7528get_group()\u65b9\u6cd5\uff0c\u53ef\u4ee5\u9009\u62e9\u4e00\u4e2a\u7ec4\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                            import pandas as pd\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\ngrouped = df.groupby('Year')\r\nprint (grouped.get_group(2014))\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                               Points  Rank    Team  Year\r\n0     876     1  Riders  2014\r\n2     863     2  Devils  2014\r\n4     741     3   Kings  2014\r\n9     701     4  Royals  2014\r\n<\/pre>\n
                                                            \u805a\u5408<\/h3>\n
                                                            \u805a\u5408\u51fd\u6570\u4e3a\u6bcf\u4e2a\u7ec4\u8fd4\u56de\u5355\u4e2a\u805a\u5408\u503c\u3002\u5f53\u521b\u5efa\u4e86\u5206\u7ec4(group by)\u5bf9\u8c61\uff0c\u5c31\u53ef\u4ee5\u5bf9\u5206\u7ec4\u6570\u636e\u6267\u884c\u591a\u4e2a\u805a\u5408\u64cd\u4f5c\u3002<\/p>\n
                                                            \u4e00\u4e2a\u6bd4\u8f83\u5e38\u7528\u7684\u662f\u901a\u8fc7\u805a\u5408\u6216\u7b49\u6548\u7684agg\u65b9\u6cd5\u805a\u5408 –<\/p>\n
                                                            import pandas as pd\r\nimport numpy as np\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\ngrouped = df.groupby('Year')\r\nprint (grouped['Points'].agg(np.mean))\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            Year\r\n2014    795.25\r\n2015    769.50\r\n2016    725.00\r\n2017    739.00\r\nName: Points, dtype: float64\r\n<\/pre>\n
                                                            \u53e6\u4e00\u79cd\u67e5\u770b\u6bcf\u4e2a\u5206\u7ec4\u7684\u5927\u5c0f\u7684\u65b9\u6cd5\u662f\u5e94\u7528size()\u51fd\u6570 –<\/p>\n
                                                            import pandas as pd\r\nimport numpy as np\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\ngrouped = df.groupby('Team')\r\nprint (grouped.agg(np.size))\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                            Team\r\nDevils       2     2     2\r\nKings        3     3     3\r\nRiders       4     4     4\r\nRoyals       2     2     2\r\nkings        1     1     1\r\n<\/pre>\n
                                                            \u4e00\u6b21\u5e94\u7528\u591a\u4e2a\u805a\u5408\u51fd\u6570<\/h3>\n
                                                            \u901a\u8fc7\u5206\u7ec4\u7cfb\u5217\uff0c\u8fd8\u53ef\u4ee5\u4f20\u9012\u51fd\u6570\u7684\u5217\u8868\u6216\u5b57\u5178\u6765\u8fdb\u884c\u805a\u5408\uff0c\u5e76\u751f\u6210DataFrame\u4f5c\u4e3a\u8f93\u51fa –<\/p>\n
                                                            import pandas as pd\r\nimport numpy as np\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\ngrouped = df.groupby('Team')\r\nagg = grouped['Points'].agg([np.sum, np.mean, np.std])\r\nprint (agg)\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                     sum        mean         std\r\nTeam\r\nDevils  1536  768.000000  134.350288\r\nKings   2285  761.666667   24.006943\r\nRiders  3049  762.250000   88.567771\r\nRoyals  1505  752.500000   72.831998\r\nkings    812  812.000000         NaN\r\n<\/pre>\n
                                                            \u8f6c\u6362<\/h3>\n
                                                            \u5206\u7ec4\u6216\u5217\u4e0a\u7684\u8f6c\u6362\u8fd4\u56de\u7d22\u5f15\u5927\u5c0f\u4e0e\u88ab\u5206\u7ec4\u7684\u7d22\u5f15\u76f8\u540c\u7684\u5bf9\u8c61\u3002\u56e0\u6b64\uff0c\u8f6c\u6362\u5e94\u8be5\u8fd4\u56de\u4e0e\u7ec4\u5757\u5927\u5c0f\u76f8\u540c\u7684\u7ed3\u679c\u3002<\/p>\n
                                                            import pandas as pd\r\nimport numpy as np\r\n\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\n\r\ngrouped = df.groupby('Team')\r\nscore = lambda x: (x - x.mean()) \/ x.std()*10\r\nprint (grouped.transform(score))\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Points       Rank       Year\r\n0   12.843272 -15.000000 -11.618950\r\n1    3.020286   5.000000  -3.872983\r\n2    7.071068  -7.071068  -7.071068\r\n3   -7.071068   7.071068   7.071068\r\n4   -8.608621  11.547005 -10.910895\r\n5         NaN        NaN        NaN\r\n6   -2.360428  -5.773503   2.182179\r\n7   10.969049  -5.773503   8.728716\r\n8   -7.705963   5.000000   3.872983\r\n9   -7.071068   7.071068  -7.071068\r\n10   7.071068  -7.071068   7.071068\r\n11  -8.157595   5.000000  11.618950\r\n<\/pre>\n
                                                            \u8fc7\u6ee4<\/h3>\n
                                                            \u8fc7\u6ee4\u6839\u636e\u5b9a\u4e49\u7684\u6807\u51c6\u8fc7\u6ee4\u6570\u636e\u5e76\u8fd4\u56de\u6570\u636e\u7684\u5b50\u96c6\u3002filter()\u51fd\u6570\u7528\u4e8e\u8fc7\u6ee4\u6570\u636e\u3002<\/p>\n
                                                            import pandas as pd\r\nimport numpy as np\r\nipl_data = {'Team': ['Riders', 'Riders', 'Devils', 'Devils', 'Kings',\r\n'kings', 'Kings', 'Kings', 'Riders', 'Royals', 'Royals', 'Riders'],\r\n'Rank': [1, 2, 2, 3, 3,4 ,1 ,1,2 , 4,1,2],\r\n'Year': [2014,2015,2014,2015,2014,2015,2016,2017,2016,2014,2015,2017],\r\n'Points':[876,789,863,673,741,812,756,788,694,701,804,690]}\r\ndf = pd.DataFrame(ipl_data)\r\nfilter = df.groupby('Team').filter(lambda x: len(x) >= 3)\r\n\r\nprint (filter)\r\n<\/pre>\n
                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                Points  Rank    Team  Year\r\n0      876     1  Riders  2014\r\n1      789     2  Riders  2015\r\n4      741     3   Kings  2014\r\n6      756     1   Kings  2016\r\n7      788     1   Kings  2017\r\n8      694     2  Riders  2016\r\n11     690     2  Riders  2017\r\n<\/pre>\n
                                                            \u5728\u4e0a\u8ff0\u8fc7\u6ee4\u6761\u4ef6\u4e0b\uff0c\u8981\u6c42\u8fd4\u56de\u4e09\u6b21\u4ee5\u4e0a\u53c2\u52a0IPL\u7684\u961f\u4f0d\u3002<\/p>\n
                                                            Python Pandas\u5408\u5e76\/\u8fde\u63a5<\/h1>\n
                                                            Pandas\u5177\u6709\u529f\u80fd\u5168\u9762\u7684\u9ad8\u6027\u80fd\u5185\u5b58\u4e2d\u8fde\u63a5\u64cd\u4f5c\uff0c\u4e0eSQL\u7b49\u5173\u7cfb\u6570\u636e\u5e93\u975e\u5e38\u76f8\u4f3c\u3002
                                                            \nPandas\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5355\u72ec\u7684merge()\u51fd\u6570\uff0c\u4f5c\u4e3aDataFrame\u5bf9\u8c61\u4e4b\u95f4\u6240\u6709\u6807\u51c6\u6570\u636e\u5e93\u8fde\u63a5\u64cd\u4f5c\u7684\u5165\u53e3 –<\/p>\n
                                                            pd.merge(left, right, how='inner', on=None, left_on=None, right_on=None,\r\nleft_index=False, right_index=False, sort=True)\r\n<\/pre>\n
                                                            \u5728\u8fd9\u91cc\uff0c\u6709\u4ee5\u4e0b\u51e0\u4e2a\u53c2\u6570\u53ef\u4ee5\u4f7f\u7528 –<\/p>\n
                                                              \n
                                                            • left – \u4e00\u4e2aDataFrame\u5bf9\u8c61\u3002<\/li>\n
                                                            • right – \u53e6\u4e00\u4e2aDataFrame\u5bf9\u8c61\u3002<\/li>\n
                                                            • on – \u5217(\u540d\u79f0)\u8fde\u63a5\uff0c\u5fc5\u987b\u5728\u5de6\u548c\u53f3DataFrame\u5bf9\u8c61\u4e2d\u5b58\u5728(\u627e\u5230)\u3002<\/li>\n
                                                            • left_on – \u5de6\u4fa7DataFrame\u4e2d\u7684\u5217\u7528\u4f5c\u952e\uff0c\u53ef\u4ee5\u662f\u5217\u540d\u6216\u957f\u5ea6\u7b49\u4e8eDataFrame\u957f\u5ea6\u7684\u6570\u7ec4\u3002<\/li>\n
                                                            • right_on – \u6765\u81ea\u53f3\u7684DataFrame\u7684\u5217\u4f5c\u4e3a\u952e\uff0c\u53ef\u4ee5\u662f\u5217\u540d\u6216\u957f\u5ea6\u7b49\u4e8eDataFrame\u957f\u5ea6\u7684\u6570\u7ec4\u3002<\/li>\n
                                                            • left_index – \u5982\u679c\u4e3aTrue\uff0c\u5219\u4f7f\u7528\u5de6\u4fa7DataFrame\u4e2d\u7684\u7d22\u5f15(\u884c\u6807\u7b7e)\u4f5c\u4e3a\u5176\u8fde\u63a5\u952e\u3002 \u5728\u5177\u6709MultiIndex(\u5206\u5c42)\u7684DataFrame\u7684\u60c5\u51b5\u4e0b\uff0c\u7ea7\u522b\u7684\u6570\u91cf\u5fc5\u987b\u4e0e\u6765\u81ea\u53f3DataFrame\u7684\u8fde\u63a5\u952e\u7684\u6570\u91cf\u76f8\u5339\u914d\u3002<\/li>\n
                                                            • right_index – \u4e0e\u53f3DataFrame\u7684left_index\u5177\u6709\u76f8\u540c\u7684\u7528\u6cd5\u3002<\/li>\n
                                                            • how – \u5b83\u662fleft, right, outer\u4ee5\u53cainner\u4e4b\u4e2d\u7684\u4e00\u4e2a\uff0c\u9ed8\u8ba4\u4e3a\u5185inner\u3002 \u4e0b\u9762\u5c06\u4ecb\u7ecd\u6bcf\u79cd\u65b9\u6cd5\u7684\u7528\u6cd5\u3002<\/li>\n
                                                            • sort – \u6309\u7167\u5b57\u5178\u987a\u5e8f\u901a\u8fc7\u8fde\u63a5\u952e\u5bf9\u7ed3\u679cDataFrame\u8fdb\u884c\u6392\u5e8f\u3002\u9ed8\u8ba4\u4e3aTrue\uff0c\u8bbe\u7f6e\u4e3aFalse\u65f6\uff0c\u5728\u5f88\u591a\u60c5\u51b5\u4e0b\u5927\u5927\u63d0\u9ad8\u6027\u80fd\u3002<\/li>\n<\/ul>\n
                                                              \u73b0\u5728\u521b\u5efa\u4e24\u4e2a\u4e0d\u540c\u7684DataFrame\u5e76\u5bf9\u5176\u6267\u884c\u5408\u5e76\u64cd\u4f5c\u3002<\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nprint (left)\r\nprint(\"========================================\")\r\nprint (right)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Name  id subject_id\r\n0    Alex   1       sub1\r\n1     Amy   2       sub2\r\n2   Allen   3       sub4\r\n3   Alice   4       sub6\r\n4  Ayoung   5       sub5\r\n========================================\r\nName  id subject_id\r\n0  Billy   1       sub2\r\n1  Brian   2       sub4\r\n2   Bran   3       sub3\r\n3  Bryce   4       sub6\r\n4  Betty   5       sub5\r\n<\/pre>\n
                                                              \u5728\u4e00\u4e2a\u952e\u4e0a\u5408\u5e76\u4e24\u4e2a\u6570\u636e\u5e27<\/strong><\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left,right,on='id')\r\nprint(rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id subject_id_x Name_y subject_id_y\r\n0    Alex   1         sub1  Billy         sub2\r\n1     Amy   2         sub2  Brian         sub4\r\n2   Allen   3         sub4   Bran         sub3\r\n3   Alice   4         sub6  Bryce         sub6\r\n4  Ayoung   5         sub5  Betty         sub5\r\n<\/pre>\n
                                                              \u5408\u5e76\u591a\u4e2a\u952e\u4e0a\u7684\u4e24\u4e2a\u6570\u636e\u6846<\/strong><\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left,right,on=['id','subject_id'])\r\nprint(rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id subject_id Name_y\r\n0   Alice   4       sub6  Bryce\r\n1  Ayoung   5       sub5  Betty\r\n<\/pre>\n
                                                              \u5408\u5e76\u4f7f\u7528\u201chow\u201d\u7684\u53c2\u6570<\/h3>\n
                                                              \u5982\u4f55\u5408\u5e76\u53c2\u6570\u6307\u5b9a\u5982\u4f55\u786e\u5b9a\u54ea\u4e9b\u952e\u5c06\u88ab\u5305\u542b\u5728\u7ed3\u679c\u8868\u4e2d\u3002\u5982\u679c\u7ec4\u5408\u952e\u6ca1\u6709\u51fa\u73b0\u5728\u5de6\u4fa7\u6216\u53f3\u4fa7\u8868\u4e2d\uff0c\u5219\u8fde\u63a5\u8868\u4e2d\u7684\u503c\u5c06\u4e3aNA\u3002<\/p>\n
                                                              \u8fd9\u91cc\u662fhow\u9009\u9879\u548cSQL\u7b49\u6548\u540d\u79f0\u7684\u603b\u7ed3 –<\/p>\n\n\n\n\n\n\n\n
                                                              \u5408\u5e76\u65b9\u6cd5<\/th>\nSQL\u7b49\u6548<\/th>\n\u63cf\u8ff0<\/th>\n<\/tr>\n
                                                              left<\/td>\nLEFT OUTER JOIN<\/td>\n\u4f7f\u7528\u5de6\u4fa7\u5bf9\u8c61\u7684\u952e<\/td>\n<\/tr>\n
                                                              right<\/td>\nRIGHT OUTER JOIN<\/td>\n\u4f7f\u7528\u53f3\u4fa7\u5bf9\u8c61\u7684\u952e<\/td>\n<\/tr>\n
                                                              outer<\/td>\nFULL OUTER JOIN<\/td>\n\u4f7f\u7528\u952e\u7684\u8054\u5408<\/td>\n<\/tr>\n
                                                              inner<\/td>\nINNER JOIN<\/td>\n\u4f7f\u7528\u952e\u7684\u4ea4\u96c6<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                              Left Join\u793a\u4f8b<\/strong><\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left, right, on='subject_id', how='left')\r\nprint (rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id_x subject_id Name_y  id_y\r\n0    Alex     1       sub1    NaN   NaN\r\n1     Amy     2       sub2  Billy   1.0\r\n2   Allen     3       sub4  Brian   2.0\r\n3   Alice     4       sub6  Bryce   4.0\r\n4  Ayoung     5       sub5  Betty   5.0\r\n<\/pre>\n
                                                              Right Join\u793a\u4f8b<\/strong><\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left, right, on='subject_id', how='right')\r\nprint (rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id_x subject_id Name_y  id_y\r\n0     Amy   2.0       sub2  Billy     1\r\n1   Allen   3.0       sub4  Brian     2\r\n2   Alice   4.0       sub6  Bryce     4\r\n3  Ayoung   5.0       sub5  Betty     5\r\n4     NaN   NaN       sub3   Bran     3\r\n<\/pre>\n
                                                              Outer Join\u793a\u4f8b<\/strong><\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left, right, how='outer', on='subject_id')\r\nprint (rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id_x subject_id Name_y  id_y\r\n0    Alex   1.0       sub1    NaN   NaN\r\n1     Amy   2.0       sub2  Billy   1.0\r\n2   Allen   3.0       sub4  Brian   2.0\r\n3   Alice   4.0       sub6  Bryce   4.0\r\n4  Ayoung   5.0       sub5  Betty   5.0\r\n5     NaN   NaN       sub3   Bran   3.0\r\n<\/pre>\n
                                                              Inner Join\u793a\u4f8b<\/strong><\/p>\n
                                                              \u8fde\u63a5\u5c06\u5728\u7d22\u5f15\u4e0a\u8fdb\u884c\u3002\u8fde\u63a5(Join)\u64cd\u4f5c\u5c06\u6388\u4e88\u5b83\u6240\u8c03\u7528\u7684\u5bf9\u8c61\u3002\u6240\u4ee5\uff0ca.join(b)\u4e0d\u7b49\u4e8eb.join(a)\u3002<\/p>\n
                                                              import pandas as pd\r\nleft = pd.DataFrame({\r\n'id':[1,2,3,4,5],\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5']})\r\nright = pd.DataFrame(\r\n{'id':[1,2,3,4,5],\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5']})\r\nrs = pd.merge(left, right, on='subject_id', how='inner')\r\nprint (rs)\r\n<\/pre>\n
                                                              \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                 Name_x  id_x subject_id Name_y  id_y\r\n0     Amy     2       sub2  Billy     1\r\n1   Allen     3       sub4  Brian     2\r\n2   Alice     4       sub6  Bryce     4\r\n3  Ayoung     5       sub5  Betty     5<\/pre>\n
                                                              Python Pandas\u7ea7\u8054<\/h1>\n
                                                              Pandas<\/strong>\u63d0\u4f9b\u4e86\u5404\u79cd\u5de5\u5177(\u529f\u80fd)\uff0c\u53ef\u4ee5\u8f7b\u677e\u5730\u5c06Series\uff0cDataFrame\u548cPanel\u5bf9\u8c61\u7ec4\u5408\u5728\u4e00\u8d77\u3002<\/p>\n
                                                              pd.concat(objs,axis=0,join='outer',join_axes=None,\r\nignore_index=False)\r\n<\/pre>\n
                                                              \u5176\u4e2d\uff0c<\/p>\n
                                                                \n
                                                              • objs – \u8fd9\u662fSeries\uff0cDataFrame\u6216Panel\u5bf9\u8c61\u7684\u5e8f\u5217\u6216\u6620\u5c04\u3002<\/li>\n
                                                              • axis – {0\uff0c1\uff0c…}\uff0c\u9ed8\u8ba4\u4e3a0\uff0c\u8fd9\u662f\u8fde\u63a5\u7684\u8f74\u3002<\/li>\n
                                                              • join – {‘inner’, ‘outer’}\uff0c\u9ed8\u8ba4inner\u3002\u5982\u4f55\u5904\u7406\u5176\u4ed6\u8f74\u4e0a\u7684\u7d22\u5f15\u3002\u8054\u5408\u7684\u5916\u90e8\u548c\u4ea4\u53c9\u7684\u5185\u90e8\u3002<\/li>\n
                                                              • ignore_index \u2212 \u5e03\u5c14\u503c\uff0c\u9ed8\u8ba4\u4e3aFalse\u3002\u5982\u679c\u6307\u5b9a\u4e3aTrue\uff0c\u5219\u4e0d\u8981\u4f7f\u7528\u8fde\u63a5\u8f74\u4e0a\u7684\u7d22\u5f15\u503c\u3002\u7ed3\u679c\u8f74\u5c06\u88ab\u6807\u8bb0\u4e3a\uff1a0\uff0c…\uff0cn-1\u3002<\/li>\n
                                                              • join_axes – \u8fd9\u662fIndex\u5bf9\u8c61\u7684\u5217\u8868\u3002\u7528\u4e8e\u5176\u4ed6(n-1)\u8f74\u7684\u7279\u5b9a\u7d22\u5f15\uff0c\u800c\u4e0d\u662f\u6267\u884c\u5185\u90e8\/\u5916\u90e8\u96c6\u903b\u8f91\u3002<\/li>\n<\/ul>\n
                                                                \u8fde\u63a5\u5bf9\u8c61<\/h3>\n
                                                                concat()\u51fd\u6570\u5b8c\u6210\u4e86\u6cbf\u8f74\u6267\u884c\u7ea7\u8054\u64cd\u4f5c\u7684\u6240\u6709\u91cd\u8981\u5de5\u4f5c\u3002\u4e0b\u9762\u4ee3\u7801\u4e2d\uff0c\u521b\u5efa\u4e0d\u540c\u7684\u5bf9\u8c61\u5e76\u8fdb\u884c\u8fde\u63a5\u3002<\/p>\n
                                                                import pandas as pd\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = pd.concat([one,two])\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Marks_scored    Name subject_id\r\n1            98    Alex       sub1\r\n2            90     Amy       sub2\r\n3            87   Allen       sub4\r\n4            69   Alice       sub6\r\n5            78  Ayoung       sub5\r\n1            89   Billy       sub2\r\n2            80   Brian       sub4\r\n3            79    Bran       sub3\r\n4            97   Bryce       sub6\r\n5            88   Betty       sub5\r\n<\/pre>\n
                                                                \u5047\u8bbe\u60f3\u628a\u7279\u5b9a\u7684\u952e\u4e0e\u6bcf\u4e2a\u788e\u7247\u7684DataFrame\u5173\u8054\u8d77\u6765\u3002\u53ef\u4ee5\u901a\u8fc7\u4f7f\u7528\u952e\u53c2\u6570\u6765\u5b9e\u73b0\u8fd9\u4e00\u70b9 –<\/p>\n
                                                                import pandas as pd\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = pd.concat([one,two],keys=['x','y'])\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                     Marks_scored    Name subject_id\r\nx 1            98    Alex       sub1\r\n  2            90     Amy       sub2\r\n  3            87   Allen       sub4\r\n  4            69   Alice       sub6\r\n  5            78  Ayoung       sub5\r\ny 1            89   Billy       sub2\r\n  2            80   Brian       sub4\r\n  3            79    Bran       sub3\r\n  4            97   Bryce       sub6\r\n  5            88   Betty       sub5\r\n<\/pre>\n
                                                                \u7ed3\u679c\u7684\u7d22\u5f15\u662f\u91cd\u590d\u7684; \u6bcf\u4e2a\u7d22\u5f15\u91cd\u590d\u3002\u5982\u679c\u60f3\u8981\u751f\u6210\u7684\u5bf9\u8c61\u5fc5\u987b\u9075\u5faa\u81ea\u5df1\u7684\u7d22\u5f15\uff0c\u8bf7\u5c06ignore_index\u8bbe\u7f6e\u4e3aTrue\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                import pandas as pd\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = pd.concat([one,two],keys=['x','y'],ignore_index=True)\r\n\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Marks_scored    Name subject_id\r\n0            98    Alex       sub1\r\n1            90     Amy       sub2\r\n2            87   Allen       sub4\r\n3            69   Alice       sub6\r\n4            78  Ayoung       sub5\r\n5            89   Billy       sub2\r\n6            80   Brian       sub4\r\n7            79    Bran       sub3\r\n8            97   Bryce       sub6\r\n9            88   Betty       sub5\r\n<\/pre>\n
                                                                \u89c2\u5bdf\uff0c\u7d22\u5f15\u5b8c\u5168\u6539\u53d8\uff0c\u952e\u4e5f\u88ab\u8986\u76d6\u3002\u5982\u679c\u9700\u8981\u6cbfaxis=1\u6dfb\u52a0\u4e24\u4e2a\u5bf9\u8c61\uff0c\u5219\u4f1a\u6dfb\u52a0\u65b0\u5217\u3002<\/p>\n
                                                                import pandas as pd\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = pd.concat([one,two],axis=1)\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Marks_scored    Name subject_id  Marks_scored   Name subject_id\r\n1            98    Alex       sub1            89  Billy       sub2\r\n2            90     Amy       sub2            80  Brian       sub4\r\n3            87   Allen       sub4            79   Bran       sub3\r\n4            69   Alice       sub6            97  Bryce       sub6\r\n5            78  Ayoung       sub5            88  Betty       sub5\r\n<\/pre>\n
                                                                \u4f7f\u7528\u9644\u52a0\u8fde\u63a5<\/h4>\n
                                                                \u8fde\u63a5\u7684\u4e00\u4e2a\u6709\u7528\u7684\u5feb\u6377\u65b9\u5f0f\u662f\u5728Series\u548cDataFrame\u5b9e\u4f8b\u7684append\u65b9\u6cd5\u3002\u8fd9\u4e9b\u65b9\u6cd5\u5b9e\u9645\u4e0a\u65e9\u4e8econcat()\u65b9\u6cd5\u3002 \u5b83\u4eec\u6cbfaxis=0\u8fde\u63a5\uff0c\u5373\u7d22\u5f15 –<\/p>\n
                                                                import pandas as pd\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = one.append(two)\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Marks_scored    Name subject_id\r\n1            98    Alex       sub1\r\n2            90     Amy       sub2\r\n3            87   Allen       sub4\r\n4            69   Alice       sub6\r\n5            78  Ayoung       sub5\r\n1            89   Billy       sub2\r\n2            80   Brian       sub4\r\n3            79    Bran       sub3\r\n4            97   Bryce       sub6\r\n5            88   Betty       sub5\r\n<\/pre>\n
                                                                append()\u51fd\u6570\u4e5f\u53ef\u4ee5\u5e26\u591a\u4e2a\u5bf9\u8c61 –<\/p>\n
                                                                import pandas as pd\r\n\r\none = pd.DataFrame({\r\n'Name': ['Alex', 'Amy', 'Allen', 'Alice', 'Ayoung'],\r\n'subject_id':['sub1','sub2','sub4','sub6','sub5'],\r\n'Marks_scored':[98,90,87,69,78]},\r\nindex=[1,2,3,4,5])\r\n\r\ntwo = pd.DataFrame({\r\n'Name': ['Billy', 'Brian', 'Bran', 'Bryce', 'Betty'],\r\n'subject_id':['sub2','sub4','sub3','sub6','sub5'],\r\n'Marks_scored':[89,80,79,97,88]},\r\nindex=[1,2,3,4,5])\r\nrs = one.append([two,one,two])\r\nprint(rs)\r\n<\/pre>\n
                                                                \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                   Marks_scored    Name subject_id\r\n1            98    Alex       sub1\r\n2            90     Amy       sub2\r\n3            87   Allen       sub4\r\n4            69   Alice       sub6\r\n5            78  Ayoung       sub5\r\n1            89   Billy       sub2\r\n2            80   Brian       sub4\r\n3            79    Bran       sub3\r\n4            97   Bryce       sub6\r\n5            88   Betty       sub5\r\n1            98    Alex       sub1\r\n2            90     Amy       sub2\r\n3            87   Allen       sub4\r\n4            69   Alice       sub6\r\n5            78  Ayoung       sub5\r\n1            89   Billy       sub2\r\n2            80   Brian       sub4\r\n3            79    Bran       sub3\r\n4            97   Bryce       sub6\r\n5            88   Betty       sub5\r\n<\/pre>\n
                                                                \u65f6\u95f4\u5e8f\u5217<\/h3>\n
                                                                Pandas\u4e3a\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u7684\u5de5\u4f5c\u65f6\u95f4\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5f3a\u5927\u7684\u5de5\u5177\uff0c\u5c24\u5176\u662f\u5728\u91d1\u878d\u9886\u57df\u3002\u5728\u5904\u7406\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u65f6\uff0c\u6211\u4eec\u7ecf\u5e38\u9047\u5230\u4ee5\u4e0b\u60c5\u51b5 –<\/p>\n
                                                                  \n
                                                                • \u751f\u6210\u65f6\u95f4\u5e8f\u5217<\/li>\n
                                                                • \u5c06\u65f6\u95f4\u5e8f\u5217\u8f6c\u6362\u4e3a\u4e0d\u540c\u7684\u9891\u7387<\/li>\n<\/ul>\n
                                                                  Pandas\u63d0\u4f9b\u4e86\u4e00\u4e2a\u76f8\u5bf9\u7d27\u51d1\u548c\u81ea\u5305\u542b\u7684\u5de5\u5177\u6765\u6267\u884c\u4e0a\u8ff0\u4efb\u52a1\u3002<\/p>\n
                                                                  \u83b7\u53d6\u5f53\u524d\u65f6\u95f4<\/h4>\n
                                                                  datetime.now()\u7528\u4e8e\u83b7\u53d6\u5f53\u524d\u7684\u65e5\u671f\u548c\u65f6\u95f4\u3002<\/p>\n
                                                                  import pandas as pd\r\nprint pd.datetime.now()\r\n<\/pre>\n
                                                                  \u4e0a\u8ff0\u4ee3\u7801\u6267\u884c\u7ed3\u679c\u5982\u4e0b –<\/p>\n
                                                                  2017-11-03 02:17:45.997992\r\n<\/pre>\n
                                                                  \u521b\u5efa\u4e00\u4e2a\u65f6\u95f4\u6233<\/h4>\n
                                                                  \u65f6\u95f4\u6233\u6570\u636e\u662f\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u7684\u6700\u57fa\u672c\u7c7b\u578b\uff0c\u5b83\u5c06\u6570\u503c\u4e0e\u65f6\u95f4\u70b9\u76f8\u5173\u8054\u3002 \u5bf9\u4e8ePandas\u5bf9\u8c61\u6765\u8bf4\uff0c\u610f\u5473\u7740\u4f7f\u7528\u65f6\u95f4\u70b9\u3002\u4e3e\u4e2a\u4f8b\u5b50 –<\/p>\n
                                                                  import pandas as pd\r\ntime = pd.Timestamp('2018-11-01')\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  2018-11-01 00:00:00\r\n<\/pre>\n
                                                                  \u4e5f\u53ef\u4ee5\u8f6c\u6362\u6574\u6570\u6216\u6d6e\u52a8\u65f6\u671f\u3002\u8fd9\u4e9b\u7684\u9ed8\u8ba4\u5355\u4f4d\u662f\u7eb3\u79d2(\u56e0\u4e3a\u8fd9\u4e9b\u662f\u5982\u4f55\u5b58\u50a8\u65f6\u95f4\u6233\u7684)\u3002 \u7136\u800c\uff0c\u65f6\u4ee3\u5f80\u5f80\u5b58\u50a8\u5728\u53e6\u4e00\u4e2a\u53ef\u4ee5\u6307\u5b9a\u7684\u5355\u5143\u4e2d\u3002 \u518d\u4e3e\u4e00\u4e2a\u4f8b\u5b50 –<\/p>\n
                                                                  import pandas as pd\r\ntime = pd.Timestamp(1588686880,unit='s')\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  2020-05-05 13:54:40\r\n<\/pre>\n
                                                                  \u521b\u5efa\u4e00\u4e2a\u65f6\u95f4\u8303\u56f4<\/h4>\n
                                                                  import pandas as pd\r\n\r\ntime = pd.date_range(\"12:00\", \"23:59\", freq=\"30min\").time\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  [datetime.time(12, 0) datetime.time(12, 30) datetime.time(13, 0)\r\ndatetime.time(13, 30) datetime.time(14, 0) datetime.time(14, 30)\r\ndatetime.time(15, 0) datetime.time(15, 30) datetime.time(16, 0)\r\ndatetime.time(16, 30) datetime.time(17, 0) datetime.time(17, 30)\r\ndatetime.time(18, 0) datetime.time(18, 30) datetime.time(19, 0)\r\ndatetime.time(19, 30) datetime.time(20, 0) datetime.time(20, 30)\r\ndatetime.time(21, 0) datetime.time(21, 30) datetime.time(22, 0)\r\ndatetime.time(22, 30) datetime.time(23, 0) datetime.time(23, 30)]\r\n<\/pre>\n
                                                                  \u6539\u53d8\u65f6\u95f4\u7684\u9891\u7387<\/h4>\n
                                                                  import pandas as pd\r\n\r\ntime = pd.date_range(\"12:00\", \"23:59\", freq=\"H\").time\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  [datetime.time(12, 0) datetime.time(13, 0) datetime.time(14, 0)\r\ndatetime.time(15, 0) datetime.time(16, 0) datetime.time(17, 0)\r\ndatetime.time(18, 0) datetime.time(19, 0) datetime.time(20, 0)\r\ndatetime.time(21, 0) datetime.time(22, 0) datetime.time(23, 0)]\r\n<\/pre>\n
                                                                  \u8f6c\u6362\u4e3a\u65f6\u95f4\u6233<\/h4>\n
                                                                  \u8981\u8f6c\u6362\u7c7b\u4f3c\u65e5\u671f\u7684\u5bf9\u8c61(\u4f8b\u5982\u5b57\u7b26\u4e32\uff0c\u65f6\u4ee3\u6216\u6df7\u5408)\u7684\u5e8f\u5217\u6216\u7c7b\u4f3c\u5217\u8868\u7684\u5bf9\u8c61\uff0c\u53ef\u4ee5\u4f7f\u7528to_datetime\u51fd\u6570\u3002\u5f53\u4f20\u9012\u65f6\u5c06\u8fd4\u56de\u4e00\u4e2aSeries(\u5177\u6709\u76f8\u540c\u7684\u7d22\u5f15)\uff0c\u800c\u7c7b\u4f3c\u5217\u8868\u88ab\u8f6c\u6362\u4e3aDatetimeIndex\u3002 \u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50 –<\/p>\n
                                                                  import pandas as pd\r\n\r\ntime = pd.to_datetime(pd.Series(['Jul 31, 2009','2019-10-10', None]))\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  0   2009-07-31\r\n1   2019-10-10\r\n2          NaT\r\ndtype: datetime64[ns]\r\n<\/pre>\n
                                                                  NaT\u8868\u793a\u4e0d\u662f\u4e00\u4e2a\u65f6\u95f4\u7684\u503c(\u76f8\u5f53\u4e8eNaN)<\/p>\n
                                                                  \u4e3e\u4e00\u4e2a\u4f8b\u5b50\uff0c<\/p>\n
                                                                  import pandas as pd\r\nimport pandas as pd\r\ntime = pd.to_datetime(['2009\/11\/23', '2019.12.31', None])\r\nprint(time)\r\n<\/pre>\n
                                                                  \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                  DatetimeIndex(['2009-11-23', '2019-12-31', 'NaT'], dtype='datetime64[ns]', freq=None)<\/pre>\n
                                                                  Python Pandas\u65e5\u671f\u529f\u80fd<\/h1>\n
                                                                  \u65e5\u671f\u529f\u80fd\u6269\u5c55\u4e86\u65f6\u95f4\u5e8f\u5217\uff0c\u5728\u8d22\u52a1\u6570\u636e\u5206\u6790\u4e2d\u8d77\u4e3b\u8981\u4f5c\u7528\u3002\u5728\u5904\u7406\u65e5\u671f\u6570\u636e\u7684\u540c\u65f6\uff0c\u6211\u4eec\u7ecf\u5e38\u4f1a\u9047\u5230\u4ee5\u4e0b\u60c5\u51b5 –<\/p>\n
                                                                    \n
                                                                  • \u751f\u6210\u65e5\u671f\u5e8f\u5217<\/li>\n
                                                                  • \u5c06\u65e5\u671f\u5e8f\u5217\u8f6c\u6362\u4e3a\u4e0d\u540c\u7684\u9891\u7387<\/li>\n<\/ul>\n
                                                                    \u521b\u5efa\u4e00\u4e2a\u65e5\u671f\u8303\u56f4<\/h3>\n
                                                                    \u901a\u8fc7\u6307\u5b9a\u5468\u671f\u548c\u9891\u7387\uff0c\u4f7f\u7528date.range()\u51fd\u6570\u5c31\u53ef\u4ee5\u521b\u5efa\u65e5\u671f\u5e8f\u5217\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u8303\u56f4\u7684\u9891\u7387\u662f\u5929\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                    import pandas as pd\r\ndatelist = pd.date_range('2020\/11\/21', periods=5)\r\nprint(datelist)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    DatetimeIndex(['2020-11-21', '2020-11-22', '2020-11-23', '2020-11-24',\r\n'2020-11-25'],\r\ndtype='datetime64[ns]', freq='D')\r\n<\/pre>\n
                                                                    \u66f4\u6539\u65e5\u671f\u9891\u7387<\/h4>\n
                                                                    import pandas as pd\r\ndatelist = pd.date_range('2020\/11\/21', periods=5,freq='M')\r\nprint(datelist)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    DatetimeIndex(['2020-11-30', '2020-12-31', '2021-01-31', '2021-02-28',\r\n'2021-03-31'],\r\ndtype='datetime64[ns]', freq='M')\r\n<\/pre>\n
                                                                    bdate_range()\u51fd\u6570<\/h3>\n
                                                                    bdate_range()\u7528\u6765\u8868\u793a\u5546\u4e1a\u65e5\u671f\u8303\u56f4\uff0c\u4e0d\u540c\u4e8edate_range()\uff0c\u5b83\u4e0d\u5305\u62ec\u661f\u671f\u516d\u548c\u661f\u671f\u5929\u3002<\/p>\n
                                                                    import pandas as pd\r\ndatelist = pd.date_range('2011\/11\/03', periods=5)\r\nprint(datelist)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    DatetimeIndex(['2017-11-03', '2017-11-06', '2017-11-07', '2017-11-08',\r\n'2017-11-09'],\r\ndtype='datetime64[ns]', freq='B')\r\n<\/pre>\n
                                                                    \u89c2\u5bdf\u523011\u67083\u65e5\u4ee5\u540e\uff0c\u65e5\u671f\u8df3\u81f311\u67086\u65e5\uff0c\u4e0d\u5305\u62ec4\u65e5\u548c5\u65e5(\u56e0\u4e3a\u5b83\u4eec\u662f\u5468\u516d\u548c\u5468\u65e5)\u3002<\/p>\n
                                                                    \u50cfdate_range\u548cbdate_range\u8fd9\u6837\u7684\u4fbf\u5229\u51fd\u6570\u5229\u7528\u4e86\u5404\u79cd\u9891\u7387\u522b\u540d\u3002date_range\u7684\u9ed8\u8ba4\u9891\u7387\u662f\u65e5\u5386\u4e2d\u7684\u81ea\u7136\u65e5\uff0c\u800cbdate_range\u7684\u9ed8\u8ba4\u9891\u7387\u662f\u5de5\u4f5c\u65e5\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                    import pandas as pd\r\nstart = pd.datetime(2017, 11, 1)\r\nend = pd.datetime(2017, 11, 5)\r\ndates = pd.date_range(start, end)\r\nprint(dates)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    DatetimeIndex(['2017-11-01', '2017-11-02', '2017-11-03', '2017-11-04',\r\n'2017-11-05'],\r\ndtype='datetime64[ns]', freq='D')\r\n<\/pre>\n
                                                                    \u504f\u79fb\u522b\u540d<\/h3>\n
                                                                    \u5927\u91cf\u7684\u5b57\u7b26\u4e32\u522b\u540d\u88ab\u8d4b\u4e88\u5e38\u7528\u7684\u65f6\u95f4\u5e8f\u5217\u9891\u7387\u3002\u6211\u4eec\u628a\u8fd9\u4e9b\u522b\u540d\u79f0\u4e3a\u504f\u79fb\u522b\u540d\u3002<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
                                                                    \u522b\u540d<\/th>\n\u63cf\u8ff0\u8bf4\u660e<\/th>\n<\/tr>\n
                                                                    B<\/td>\n\u5de5\u4f5c\u65e5\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BQS<\/td>\n\u5546\u52a1\u5b63\u5ea6\u5f00\u59cb\u9891\u7387<\/td>\n<\/tr>\n
                                                                    D<\/td>\n\u65e5\u5386\/\u81ea\u7136\u65e5\u9891\u7387<\/td>\n<\/tr>\n
                                                                    A<\/td>\n\u5e74\u5ea6(\u5e74)\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    W<\/td>\n\u6bcf\u5468\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BA<\/td>\n\u5546\u52a1\u5e74\u5e95\u7ed3\u675f<\/td>\n<\/tr>\n
                                                                    M<\/td>\n\u6708\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BAS<\/td>\n\u5546\u52a1\u5e74\u5ea6\u5f00\u59cb\u9891\u7387<\/td>\n<\/tr>\n
                                                                    SM<\/td>\n\u534a\u6708\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BH<\/td>\n\u5546\u52a1\u65f6\u95f4\u9891\u7387<\/td>\n<\/tr>\n
                                                                    SM<\/td>\n\u534a\u6708\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BH<\/td>\n\u5546\u52a1\u65f6\u95f4\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BM<\/td>\n\u5546\u52a1\u6708\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    H<\/td>\n\u5c0f\u65f6\u9891\u7387<\/td>\n<\/tr>\n
                                                                    MS<\/td>\n\u6708\u8d77\u59cb\u9891\u7387<\/td>\n<\/tr>\n
                                                                    T, min<\/td>\n\u5206\u949f\u7684\u9891\u7387<\/td>\n<\/tr>\n
                                                                    SMS<\/td>\nSMS\u534a\u5f00\u59cb\u9891\u7387<\/td>\n<\/tr>\n
                                                                    S<\/td>\n\u79d2\u9891\u7387<\/td>\n<\/tr>\n
                                                                    BMS<\/td>\n\u5546\u52a1\u6708\u5f00\u59cb\u9891\u7387<\/td>\n<\/tr>\n
                                                                    L, ms<\/td>\n\u6beb\u79d2<\/td>\n<\/tr>\n
                                                                    Q<\/td>\n\u5b63\u5ea6\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    U, us<\/td>\n\u5fae\u79d2<\/td>\n<\/tr>\n
                                                                    BQ<\/td>\n\u5546\u52a1\u5b63\u5ea6\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    N<\/td>\n\u7eb3\u79d2<\/td>\n<\/tr>\n
                                                                    BQ<\/td>\n\u5546\u52a1\u5b63\u5ea6\u7ed3\u675f\u9891\u7387<\/td>\n<\/tr>\n
                                                                    QS<\/td>\n\u5b63\u5ea6\u5f00\u59cb\u9891\u7387<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                                    Python Pandas\u65f6\u95f4\u5dee<\/h1>\n
                                                                    \u65f6\u95f4\u5dee(Timedelta)\u662f\u65f6\u95f4\u4e0a\u7684\u5dee\u5f02\uff0c\u4ee5\u4e0d\u540c\u7684\u5355\u4f4d\u6765\u8868\u793a\u3002\u4f8b\u5982\uff1a\u65e5\uff0c\u5c0f\u65f6\uff0c\u5206\u949f\uff0c\u79d2\u3002\u5b83\u4eec\u53ef\u4ee5\u662f\u6b63\u503c\uff0c\u4e5f\u53ef\u4ee5\u662f\u8d1f\u503c\u3002
                                                                    \n\u53ef\u4ee5\u4f7f\u7528\u5404\u79cd\u53c2\u6570\u521b\u5efaTimedelta\u5bf9\u8c61\uff0c\u5982\u4e0b\u6240\u793a –<\/p>\n
                                                                    \u5b57\u7b26\u4e32<\/h3>\n
                                                                    \u901a\u8fc7\u4f20\u9012\u5b57\u7b26\u4e32\uff0c\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2atimedelta\u5bf9\u8c61\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                    import pandas as pd\r\n\r\ntimediff = pd.Timedelta('2 days 2 hours 15 minutes 30 seconds')\r\nprint(timediff)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u6551\u547d\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    2 days 02:15:30\r\n<\/pre>\n
                                                                    \u6574\u6570<\/h3>\n
                                                                    \u901a\u8fc7\u4f20\u9012\u4e00\u4e2a\u6574\u6570\u503c\u4e0e\u6307\u5b9a\u5355\u4f4d\uff0c\u8fd9\u6837\u7684\u4e00\u4e2a\u53c2\u6570\u4e5f\u53ef\u4ee5\u7528\u6765\u521b\u5efaTimedelta\u5bf9\u8c61\u3002<\/p>\n
                                                                    import pandas as pd\r\n\r\ntimediff = pd.Timedelta(6,unit='h')\r\nprint(timediff)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u6551\u547d\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    0 days 06:00:00\r\n<\/pre>\n
                                                                    \u6570\u636e\u504f\u79fb<\/h3>\n
                                                                    \u4f8b\u5982 – \u5468\uff0c\u5929\uff0c\u5c0f\u65f6\uff0c\u5206\u949f\uff0c\u79d2\uff0c\u6beb\u79d2\uff0c\u5fae\u79d2\uff0c\u7eb3\u79d2\u7684\u6570\u636e\u504f\u79fb\u4e5f\u53ef\u7528\u4e8e\u6784\u5efa\u3002<\/p>\n
                                                                    import pandas as pd\r\n\r\ntimediff = pd.Timedelta(days=2)\r\nprint(timediff)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u6551\u547d\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                    2 days 00:00:00\r\n<\/pre>\n
                                                                    \u8fd0\u7b97\u64cd\u4f5c<\/h3>\n
                                                                    \u53ef\u4ee5\u5728Series\/DataFrames\u4e0a\u6267\u884c\u8fd0\u7b97\u64cd\u4f5c\uff0c\u5e76\u901a\u8fc7\u5728datetime64 [ns]\u7cfb\u5217\u6216\u5728\u65f6\u95f4\u6233\u4e0a\u51cf\u6cd5\u64cd\u4f5c\u6765\u6784\u9020timedelta64 [ns]\u7cfb\u5217\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                    import pandas as pd\r\n\r\ns = pd.Series(pd.date_range('2012-1-1', periods=3, freq='D'))\r\ntd = pd.Series([ pd.Timedelta(days=i) for i in range(3) ])\r\ndf = pd.DataFrame(dict(A = s, B = td))\r\nprint(df)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                               A      B\r\n0 2012-01-01 0 days\r\n1 2012-01-02 1 days\r\n2 2012-01-03 2 days\r\n<\/pre>\n
                                                                    \u76f8\u52a0\u64cd\u4f5c<\/h4>\n
                                                                    import pandas as pd\r\n\r\ns = pd.Series(pd.date_range('2018-1-1', periods=3, freq='D'))\r\ntd = pd.Series([ pd.Timedelta(days=i) for i in range(3) ])\r\ndf = pd.DataFrame(dict(A = s, B = td))\r\ndf['C']=df['A']+df['B']\r\nprint(df)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                               A      B          C\r\n0 2018-01-01 0 days 2018-01-01\r\n1 2018-01-02 1 days 2018-01-03\r\n2 2018-01-03 2 days 2018-01-05\r\n<\/pre>\n
                                                                    \u76f8\u51cf\u64cd\u4f5c<\/h4>\n
                                                                    import pandas as pd\r\n\r\ns = pd.Series(pd.date_range('2012-1-1', periods=3, freq='D'))\r\ntd = pd.Series([ pd.Timedelta(days=i) for i in range(3) ])\r\ndf = pd.DataFrame(dict(A = s, B = td))\r\ndf['C']=df['A']+df['B']\r\ndf['D']=df['C']-df['B']\r\nprint(df)\r\n<\/pre>\n
                                                                    \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                               A      B          C          D\r\n0 2018-01-01 0 days 2018-01-01 2018-01-01\r\n1 2018-01-02 1 days 2018-01-03 2018-01-02\r\n2 2018-01-03 2 days 2018-01-05 2018-01-03<\/pre>\n
                                                                    Python Pandas\u5206\u7c7b\u6570\u636e<\/h1>\n
                                                                    \u901a\u5e38\u5b9e\u65f6\u7684\u6570\u636e\u5305\u62ec\u91cd\u590d\u7684\u6587\u672c\u5217\u3002\u4f8b\u5982\uff1a\u6027\u522b\uff0c\u56fd\u5bb6\u548c\u4ee3\u7801\u7b49\u7279\u5f81\u603b\u662f\u91cd\u590d\u7684\u3002\u8fd9\u4e9b\u662f\u5206\u7c7b\u6570\u636e\u7684\u4f8b\u5b50\u3002<\/p>\n
                                                                    \u5206\u7c7b\u53d8\u91cf\u53ea\u80fd\u91c7\u7528\u6709\u9650\u7684\u6570\u91cf\uff0c\u800c\u4e14\u901a\u5e38\u662f\u56fa\u5b9a\u7684\u6570\u91cf\u3002\u9664\u4e86\u56fa\u5b9a\u957f\u5ea6\uff0c\u5206\u7c7b\u6570\u636e\u53ef\u80fd\u6709\u987a\u5e8f\uff0c\u4f46\u4e0d\u80fd\u6267\u884c\u6570\u5b57\u64cd\u4f5c\u3002 \u5206\u7c7b\u662fPandas<\/em>\u6570\u636e\u7c7b\u578b\u3002<\/p>\n
                                                                    \u5206\u7c7b\u6570\u636e\u7c7b\u578b\u5728\u4ee5\u4e0b\u60c5\u51b5\u4e0b\u975e\u5e38\u6709\u7528 –<\/p>\n
                                                                      \n
                                                                    • \u4e00\u4e2a\u5b57\u7b26\u4e32\u53d8\u91cf\uff0c\u53ea\u5305\u542b\u51e0\u4e2a\u4e0d\u540c\u7684\u503c\u3002\u5c06\u8fd9\u6837\u7684\u5b57\u7b26\u4e32\u53d8\u91cf\u8f6c\u6362\u4e3a\u5206\u7c7b\u53d8\u91cf\u5c06\u4f1a\u8282\u7701\u4e00\u4e9b\u5185\u5b58\u3002<\/li>\n
                                                                    • \u53d8\u91cf\u7684\u8bcd\u6c47\u987a\u5e8f\u4e0e\u903b\u8f91\u987a\u5e8f(“one”\uff0c”two”\uff0c”three”)\u4e0d\u540c\u3002 \u901a\u8fc7\u8f6c\u6362\u4e3a\u5206\u7c7b\u5e76\u6307\u5b9a\u7c7b\u522b\u4e0a\u7684\u987a\u5e8f\uff0c\u6392\u5e8f\u548c\u6700\u5c0f\/\u6700\u5927\u5c06\u4f7f\u7528\u903b\u8f91\u987a\u5e8f\uff0c\u800c\u4e0d\u662f\u8bcd\u6cd5\u987a\u5e8f\u3002<\/li>\n
                                                                    • \u4f5c\u4e3a\u5176\u4ed6python\u5e93\u7684\u4e00\u4e2a\u4fe1\u53f7\uff0c\u8fd9\u4e2a\u5217\u5e94\u8be5\u88ab\u5f53\u4f5c\u4e00\u4e2a\u5206\u7c7b\u53d8\u91cf(\u4f8b\u5982\uff0c\u4f7f\u7528\u5408\u9002\u7684\u7edf\u8ba1\u65b9\u6cd5\u6216plot\u7c7b\u578b)\u3002<\/li>\n<\/ul>\n
                                                                      \u5bf9\u8c61\u521b\u5efa<\/h3>\n
                                                                      \u5206\u7c7b\u5bf9\u8c61\u53ef\u4ee5\u901a\u8fc7\u591a\u79cd\u65b9\u5f0f\u521b\u5efa\u3002\u4e0b\u9762\u4ecb\u7ecd\u4e86\u4e0d\u540c\u7684\u65b9\u6cd5 –<\/p>\n
                                                                      \u7c7b\u522b\/\u5206\u7c7b<\/strong><\/p>\n
                                                                      \u901a\u8fc7\u5728pandas\u5bf9\u8c61\u521b\u5efa\u4e2d\u5c06dtype\u6307\u5b9a\u4e3a\u201ccategory\u201d\u3002<\/p>\n
                                                                      import pandas as pd\r\ns = pd.Series([\"a\",\"b\",\"c\",\"a\"], dtype=\"category\")\r\nprint (s)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      0    a\r\n1    b\r\n2    c\r\n3    a\r\ndtype: category\r\nCategories (3, object): [a, b, c]\r\n<\/pre>\n
                                                                      \u4f20\u9012\u7ed9\u7cfb\u5217\u5bf9\u8c61\u7684\u5143\u7d20\u6570\u91cf\u662f\u56db\u4e2a\uff0c\u4f46\u7c7b\u522b\u53ea\u6709\u4e09\u4e2a\u3002\u89c2\u5bdf\u76f8\u540c\u7684\u8f93\u51fa\u7c7b\u522b\u3002<\/p>\n
                                                                      pd.Categorical<\/strong><\/p>\n
                                                                      \u4f7f\u7528\u6807\u51c6Pandas<\/em>\u5206\u7c7b\u6784\u9020\u51fd\u6570\uff0c\u6211\u4eec\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a\u7c7b\u522b\u5bf9\u8c61\u3002\u8bed\u6cd5\u5982\u4e0b –<\/p>\n
                                                                      pandas.Categorical(values, categories, ordered)\r\n<\/pre>\n
                                                                      \u4e3e\u4e2a\u4f8b\u5b50 –<\/p>\n
                                                                      import pandas as pd\r\ncat = pd.Categorical(['a', 'b', 'c', 'a', 'b', 'c'])\r\nprint (cat)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      [a, b, c, a, b, c]\r\nCategories (3, object): [a, b, c]\r\n<\/pre>\n
                                                                      \u518d\u4e3e\u4e00\u4e2a\u4f8b\u5b50 –<\/p>\n
                                                                      import pandas as pd\r\ncat = cat=pd.Categorical(['a','b','c','a','b','c','d'], ['c', 'b', 'a'])\r\nprint (cat)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      [a, b, c, a, b, c, NaN]\r\nCategories (3, object): [c, b, a]\r\n<\/pre>\n
                                                                      \u8fd9\u91cc\uff0c\u7b2c\u4e8c\u4e2a\u53c2\u6570\u8868\u793a\u7c7b\u522b\u3002\u56e0\u6b64\uff0c\u5728\u7c7b\u522b\u4e2d\u4e0d\u5b58\u5728\u7684\u4efb\u4f55\u503c\u5c06\u88ab\u89c6\u4e3aNaN\u3002<\/p>\n
                                                                      \u73b0\u5728\uff0c\u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50 –<\/p>\n
                                                                      import pandas as pd\r\ncat = cat=pd.Categorical(['a','b','c','a','b','c','d'], ['c', 'b', 'a'],ordered=True)\r\nprint (cat)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u7ed3\u679c –<\/p>\n
                                                                      [a, b, c, a, b, c, NaN]\r\nCategories (3, object): [c < b < a]\r\n<\/pre>\n
                                                                      \u4ece\u903b\u8f91\u4e0a\u8bb2\uff0c\u6392\u5e8f(ordered<\/em>)\u610f\u5473\u7740\uff0ca\u5927\u4e8eb\uff0cb\u5927\u4e8ec\u3002<\/p>\n
                                                                      \u63cf\u8ff0<\/strong><\/p>\n
                                                                      \u4f7f\u7528\u5206\u7c7b\u6570\u636e\u4e0a\u7684.describe()\u547d\u4ee4\uff0c\u53ef\u4ee5\u5f97\u5230\u4e0e\u7c7b\u578b\u5b57\u7b26\u4e32\u7684Series\u6216DataFrame\u7c7b\u4f3c\u7684\u8f93\u51fa\u3002<\/p>\n
                                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ncat = pd.Categorical([\"a\", \"c\", \"c\", np.nan], categories=[\"b\", \"a\", \"c\"])\r\ndf = pd.DataFrame({\"cat\":cat, \"s\":[\"a\", \"c\", \"c\", np.nan]})\r\nprint (df.describe())\r\nprint (\"=============================\")\r\nprint (df[\"cat\"].describe())\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                             cat  s\r\ncount    3  3\r\nunique   2  2\r\ntop      c  c\r\nfreq     2  2\r\n=============================\r\ncount     3\r\nunique    2\r\ntop       c\r\nfreq      2\r\nName: cat, dtype: object\r\n<\/pre>\n
                                                                      \u83b7\u53d6\u7c7b\u522b\u7684\u5c5e\u6027<\/strong><\/p>\n
                                                                      obj.cat.categories\u547d\u4ee4\u7528\u4e8e\u83b7\u53d6\u5bf9\u8c61\u7684\u7c7b\u522b\u3002<\/p>\n
                                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Categorical([\"a\", \"c\", \"c\", np.nan], categories=[\"b\", \"a\", \"c\"])\r\nprint (s.categories)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      Index(['b', 'a', 'c'], dtype='object')\r\n<\/pre>\n
                                                                      obj.ordered<\/strong>\u547d\u4ee4\u7528\u4e8e\u83b7\u53d6\u5bf9\u8c61\u7684\u987a\u5e8f\u3002<\/p>\n
                                                                      import pandas as pd\r\nimport numpy as np\r\n\r\ncat = pd.Categorical([\"a\", \"c\", \"c\", np.nan], categories=[\"b\", \"a\", \"c\"])\r\nprint (cat.ordered)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      False\r\n<\/pre>\n
                                                                      \u8be5\u51fd\u6570\u8fd4\u56de\u7ed3\u679c\u4e3a\uff1aFalse<\/em>\uff0c\u56e0\u4e3a\u8fd9\u91cc\u6ca1\u6709\u6307\u5b9a\u4efb\u4f55\u987a\u5e8f\u3002<\/p>\n
                                                                      \u91cd\u547d\u540d\u7c7b\u522b<\/strong><\/p>\n
                                                                      \u91cd\u547d\u540d\u7c7b\u522b\u662f\u901a\u8fc7\u5c06\u65b0\u503c\u5206\u914d\u7ed9series.cat.categories\u5c5e\u6027\u6765\u5b8c\u6210\u7684\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                      import pandas as pd\r\n\r\ns = pd.Series([\"a\",\"b\",\"c\",\"a\"], dtype=\"category\")\r\ns.cat.categories = [\"Group %s\" % g for g in s.cat.categories]\r\n\r\nprint (s.cat.categories)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      Index(['Group a', 'Group b', 'Group c'], dtype='object')\r\n<\/pre>\n
                                                                      \u521d\u59cb\u7c7b\u522b[a\uff0cb\uff0cc]\u7531\u5bf9\u8c61\u7684s.cat.categories\u5c5e\u6027\u66f4\u65b0\u3002<\/p>\n
                                                                      \u9644\u52a0\u65b0\u7c7b\u522b<\/strong>
                                                                      \n\u4f7f\u7528Categorical.add.categories()\u65b9\u6cd5\uff0c\u53ef\u4ee5\u8ffd\u52a0\u65b0\u7684\u7c7b\u522b\u3002<\/p>\n
                                                                      import pandas as pd\r\n\r\ns = pd.Series([\"a\",\"b\",\"c\",\"a\"], dtype=\"category\")\r\ns = s.cat.add_categories([4])\r\nprint (s.cat.categories)\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      Index(['a', 'b', 'c', 4], dtype='object')\r\n<\/pre>\n
                                                                      \u5220\u9664\u7c7b\u522b<\/strong>
                                                                      \n\u4f7f\u7528Categorical.remove_categories()\u65b9\u6cd5\uff0c\u53ef\u4ee5\u5220\u9664\u4e0d\u9700\u8981\u7684\u7c7b\u522b\u3002<\/p>\n
                                                                      import pandas as pd\r\n\r\ns = pd.Series([\"a\",\"b\",\"c\",\"a\"], dtype=\"category\")\r\nprint (\"Original object:\")\r\nprint (s)\r\nprint(\"=====================================\")\r\nprint (\"After removal:\")\r\nprint (s.cat.remove_categories(\"a\"))\r\n<\/pre>\n
                                                                      \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                      Original object:\r\n0    a\r\n1    b\r\n2    c\r\n3    a\r\ndtype: category\r\nCategories (3, object): [a, b, c]\r\n=====================================\r\nAfter removal:\r\n0    NaN\r\n1      b\r\n2      c\r\n3    NaN\r\ndtype: category\r\nCategories (2, object): [b, c]\r\n<\/pre>\n
                                                                      \u5206\u7c7b\u6570\u636e\u7684\u6bd4\u8f83<\/strong><\/p>\n
                                                                      \u5728\u4e09\u79cd\u60c5\u51b5\u4e0b\u53ef\u4ee5\u5c06\u5206\u7c7b\u6570\u636e\u4e0e\u5176\u4ed6\u5bf9\u8c61\u8fdb\u884c\u6bd4\u8f83 –<\/p>\n
                                                                        \n
                                                                      • \u5c06\u7b49\u53f7(==\u548c!=)\u4e0e\u7c7b\u522b\u6570\u636e\u76f8\u540c\u957f\u5ea6\u7684\u7c7b\u4f3c\u5217\u8868\u7684\u5bf9\u8c61(\u5217\u8868\uff0c\u7cfb\u5217\uff0c\u6570\u7ec4\u2026)\u8fdb\u884c\u6bd4\u8f83\u3002<\/li>\n
                                                                      • \u5f53ordered==True\u548c\u7c7b\u522b\u662f\u76f8\u540c\u65f6\uff0c\u6240\u6709\u6bd4\u8f83(==\uff0c!=\uff0c>\uff0c>=\uff0c<\uff0c\u548c<=)\u5206\u7c7b\u6570\u636e\u5230\u53e6\u4e00\u4e2a\u5206\u7c7b\u7cfb\u5217\u3002<\/li>\n
                                                                      • \u5c06\u5206\u7c7b\u6570\u636e\u4e0e\u6807\u91cf\u8fdb\u884c\u6bd4\u8f83\u3002<\/li>\n<\/ul>\n
                                                                        \u770b\u770b\u4e0b\u9762\u7684\u4f8b\u5b50 –<\/p>\n
                                                                        import pandas as pd\r\n\r\ncat = pd.Series([1,2,3]).astype(\"category\", categories=[1,2,3], ordered=True)\r\ncat1 = pd.Series([2,2,2]).astype(\"category\", categories=[1,2,3], ordered=True)\r\n\r\nprint (cat>cat1)\r\n<\/pre>\n
                                                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                        0    False\r\n1    False\r\n2     True\r\ndtype: bool<\/pre>\n
                                                                        Python Pandas\u53ef\u89c6\u5316<\/h1>\n
                                                                        \u57fa\u672c\u7ed8\u56fe\uff1a\u7ed8\u56fe<\/strong><\/p>\n
                                                                        Series\u548cDataFrame\u4e0a\u7684\u8fd9\u4e2a\u529f\u80fd\u53ea\u662f\u4f7f\u7528matplotlib\u5e93\u7684plot()\u65b9\u6cd5\u7684\u7b80\u5355\u5305\u88c5\u5b9e\u73b0\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                        import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10,4),index=pd.date_range('2018\/12\/18',\r\nperiods=10), columns=list('ABCD'))\r\n\r\ndf.plot()\r\n<\/pre>\n
                                                                        \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                        \"\"<\/div><\/p>\n
                                                                        \u5982\u679c\u7d22\u5f15\u7531\u65e5\u671f\u7ec4\u6210\uff0c\u5219\u8c03\u7528gct().autofmt_xdate()\u6765\u683c\u5f0f\u5316x\u8f74\uff0c\u5982\u4e0a\u56fe\u6240\u793a\u3002<\/p>\n
                                                                        \u6211\u4eec\u53ef\u4ee5\u4f7f\u7528x\u548cy\u5173\u952e\u5b57\u7ed8\u5236\u4e00\u5217\u4e0e\u53e6\u4e00\u5217\u3002<\/p>\n
                                                                        \u7ed8\u56fe\u65b9\u6cd5\u5141\u8bb8\u9664\u9ed8\u8ba4\u7ebf\u56fe\u4e4b\u5916\u7684\u5c11\u6570\u7ed8\u56fe\u6837\u5f0f\u3002 \u8fd9\u4e9b\u65b9\u6cd5\u53ef\u4ee5\u4f5c\u4e3aplot()\u7684kind\u5173\u952e\u5b57\u53c2\u6570\u63d0\u4f9b\u3002\u8fd9\u4e9b\u5305\u62ec –<\/p>\n
                                                                          \n
                                                                        • bar\u6216barh\u4e3a\u6761\u5f62<\/li>\n
                                                                        • hist\u4e3a\u76f4\u65b9\u56fe<\/li>\n
                                                                        • boxplot\u4e3a\u76d2\u578b\u56fe<\/li>\n
                                                                        • area\u4e3a\u201c\u9762\u79ef\u201d<\/li>\n
                                                                        • scatter\u4e3a\u6563\u70b9\u56fe<\/li>\n<\/ul>\n
                                                                          \u6761\u5f62\u56fe<\/h3>\n
                                                                          \u73b0\u5728\u901a\u8fc7\u521b\u5efa\u4e00\u4e2a\u6761\u5f62\u56fe\u6765\u770b\u770b\u6761\u5f62\u56fe\u662f\u4ec0\u4e48\u3002\u6761\u5f62\u56fe\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u65b9\u5f0f\u6765\u521b\u5efa –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.rand(10,4),columns=['a','b','c','d'])\r\ndf.plot.bar()\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u8981\u751f\u6210\u4e00\u4e2a\u5806\u79ef\u6761\u5f62\u56fe\uff0c\u901a\u8fc7\u6307\u5b9a\uff1apass stacked=True<\/b> –<\/p>\n
                                                                          import pandas as pd\r\ndf = pd.DataFrame(np.random.rand(10,4),columns=['a','b','c','d'])\r\ndf.plot.bar(stacked=True)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u8981\u83b7\u5f97\u6c34\u5e73\u6761\u5f62\u56fe\uff0c\u4f7f\u7528barh()\u65b9\u6cd5 –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\t\t \r\ndf = pd.DataFrame(np.random.rand(10,4),columns=['a','b','c','d'])\r\n\r\ndf.plot.barh(stacked=True)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u76f4\u65b9\u56fe<\/h3>\n
                                                                          \u53ef\u4ee5\u4f7f\u7528plot.hist()\u65b9\u6cd5\u7ed8\u5236\u76f4\u65b9\u56fe\u3002\u6211\u4eec\u53ef\u4ee5\u6307\u5b9abins\u7684\u6570\u91cf\u503c\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\t\r\ndf = pd.DataFrame({'a':np.random.randn(1000)+1,'b':np.random.randn(1000),'c':\r\n\t\t   np.random.randn(1000) - 1}, columns=['a', 'b', 'c'])\r\n\r\ndf.plot.hist(bins=20)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u8981\u4e3a\u6bcf\u5217\u7ed8\u5236\u4e0d\u540c\u7684\u76f4\u65b9\u56fe\uff0c\u8bf7\u4f7f\u7528\u4ee5\u4e0b\u4ee3\u7801 –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf=pd.DataFrame({'a':np.random.randn(1000)+1,'b':np.random.randn(1000),'c':\r\nnp.random.randn(1000) - 1}, columns=['a', 'b', 'c'])\r\n\r\ndf.hist(bins=20)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u7bb1\u5f62\u56fe<\/h3>\n
                                                                          Boxplot\u53ef\u4ee5\u7ed8\u5236\u8c03\u7528Series.box.plot()\u548cDataFrame.box.plot()\u6216DataFrame.boxplot()\u6765\u53ef\u89c6\u5316\u6bcf\u5217\u4e2d\u503c\u7684\u5206\u5e03\u3002<\/p>\n
                                                                          \u4f8b\u5982\uff0c\u8fd9\u91cc\u662f\u4e00\u4e2a\u7bb1\u5f62\u56fe\uff0c\u8868\u793a\u5bf9[0,1)\u4e0a\u7684\u7edf\u4e00\u968f\u673a\u53d8\u91cf\u768410\u6b21\u89c2\u5bdf\u7684\u4e94\u6b21\u8bd5\u9a8c\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.rand(10, 5), columns=['A', 'B', 'C', 'D', 'E'])\r\ndf.plot.box()\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u533a\u57df\u5757\u56fe\u5f62<\/h3>\n
                                                                          \u53ef\u4ee5\u4f7f\u7528Series.plot.area()\u6216DataFrame.plot.area()\u65b9\u6cd5\u521b\u5efa\u533a\u57df\u56fe\u5f62\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.rand(10, 4), columns=['a', 'b', 'c', 'd'])\r\ndf.plot.area()\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u6563\u70b9\u56fe\u5f62<\/h3>\n
                                                                          \u53ef\u4ee5\u4f7f\u7528DataFrame.plot.scatter()\u65b9\u6cd5\u521b\u5efa\u6563\u70b9\u56fe\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.rand(50, 4), columns=['a', 'b', 'c', 'd'])\r\ndf.plot.scatter(x='a', y='b')\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          \u997c\u72b6\u56fe<\/h3>\n
                                                                          \u997c\u72b6\u56fe\u53ef\u4ee5\u4f7f\u7528DataFrame.plot.pie()\u65b9\u6cd5\u521b\u5efa\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(3 * np.random.rand(4), index=['a', 'b', 'c', 'd'], columns=['x'])\r\ndf.plot.pie(subplots=True)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          \"\"<\/div><\/p>\n
                                                                          Python Pandas IO\u5de5\u5177<\/h1>\n
                                                                          Pandas I\/O API\u662f\u4e00\u5957\u50cfpd.read_csv()\u4e00\u6837\u8fd4\u56dePandas\u5bf9\u8c61\u7684\u9876\u7ea7\u8bfb\u53d6\u5668\u51fd\u6570\u3002<\/p>\n
                                                                          \u8bfb\u53d6\u6587\u672c\u6587\u4ef6(\u6216\u5e73\u9762\u6587\u4ef6)\u7684\u4e24\u4e2a\u4e3b\u8981\u529f\u80fd\u662fread_csv()\u548cread_table()\u3002\u5b83\u4eec\u90fd\u4f7f\u7528\u76f8\u540c\u7684\u89e3\u6790\u4ee3\u7801\u6765\u667a\u80fd\u5730\u5c06\u8868\u683c\u6570\u636e\u8f6c\u6362\u4e3aDataFrame\u5bf9\u8c61 –<\/p>\n
                                                                          pandas.read_csv(filepath_or_buffer, sep=',', delimiter=None, header='infer',\r\n\tnames=None, index_col=None, usecols=None)\r\n<\/pre>\n
                                                                          pandas.read_csv(filepath_or_buffer, sep='\\t', delimiter=None, header='infer',\r\n\tnames=None, index_col=None, usecols=None)\r\n<\/pre>\n
                                                                          \u4ee5\u4e0b\u662fcsv\u6587\u4ef6\u6570\u636e\u7684\u5185\u5bb9 –<\/p>\n
                                                                          S.No,Name,Age,City,Salary\r\n1,Tom,28,Toronto,20000\r\n2,Lee,32,HongKong,3000\r\n3,Steven,43,Bay Area,8300\r\n4,Ram,38,Hyderabad,3900\r\n<\/pre>\n
                                                                          \u5c06\u8fd9\u4e9b\u6570\u636e\u4fdd\u5b58\u4e3atemp.csv\u5e76\u5bf9\u5176\u8fdb\u884c\u64cd\u4f5c\u3002<\/p>\n
                                                                          S.No,Name,Age,City,Salary\r\n1,Tom,28,Toronto,20000\r\n2,Lee,32,HongKong,3000\r\n3,Steven,43,Bay Area,8300\r\n4,Ram,38,Hyderabad,3900\r\n<\/pre>\n
                                                                          read.csv<\/h3>\n
                                                                          read.csv\u4ececsv\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6570\u636e\u5e76\u521b\u5efa\u4e00\u4e2aDataFrame\u5bf9\u8c61\u3002<\/p>\n
                                                                          import pandas as pd\r\ndf=pd.read_csv(\"temp.csv\")\r\nprint (df)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                             S.No    Name  Age       City  Salary\r\n0     1     Tom   28    Toronto   20000\r\n1     2     Lee   32   HongKong    3000\r\n2     3  Steven   43   Bay Area    8300\r\n3     4     Ram   38  Hyderabad    3900\r\n<\/pre>\n
                                                                          \u81ea\u5b9a\u4e49\u7d22\u5f15<\/strong><\/p>\n
                                                                          \u53ef\u4ee5\u6307\u5b9acsv\u6587\u4ef6\u4e2d\u7684\u4e00\u5217\u6765\u4f7f\u7528index_col\u5b9a\u5236\u7d22\u5f15\u3002<\/p>\n
                                                                          import pandas as pd\r\n\r\ndf=pd.read_csv(\"temp.csv\",index_col=['S.No'])\r\nprint (df)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                                  Name  Age       City  Salary\r\nS.No\r\n1        Tom   28    Toronto   20000\r\n2        Lee   32   HongKong    3000\r\n3     Steven   43   Bay Area    8300\r\n4        Ram   38  Hyderabad    3900\r\n<\/pre>\n
                                                                          \u8f6c\u6362\u5668<\/strong>
                                                                          \ndtype\u7684\u5217\u53ef\u4ee5\u4f5c\u4e3a\u5b57\u5178\u4f20\u9012\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\ndf = pd.read_csv(\"temp.csv\", dtype={'Salary': np.float64})\r\nprint (df.dtypes)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          S.No        int64\r\nName       object\r\nAge         int64\r\nCity       object\r\nSalary    float64\r\ndtype: object\r\n<\/pre>\n
                                                                          \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0cSalary\u5217\u7684dtype\u662fint\uff0c\u4f46\u7ed3\u679c\u663e\u793a\u4e3afloat\uff0c\u56e0\u4e3a\u6211\u4eec\u660e\u786e\u5730\u8f6c\u6362\u4e86\u7c7b\u578b\u3002<\/p>\n
                                                                          \u56e0\u6b64\uff0c\u6570\u636e\u770b\u8d77\u6765\u50cf\u6d6e\u70b9\u6570 –<\/p>\n
                                                                           S.No   Name   Age      City    Salary\r\n0   1     Tom   28    Toronto   20000.0\r\n1   2     Lee   32   HongKong    3000.0\r\n2   3  Steven   43   Bay Area    8300.0\r\n3   4     Ram   38  Hyderabad    3900.0\r\n<\/pre>\n
                                                                          header_names<\/strong>
                                                                          \n\u4f7f\u7528names\u53c2\u6570\u6307\u5b9a\u6807\u9898\u7684\u540d\u79f0\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf=pd.read_csv(\"temp.csv\", names=['a', 'b', 'c','d','e'])\r\nprint (df)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                                a       b    c          d       e\r\n0  S.No    Name  Age       City  Salary\r\n1     1     Tom   28    Toronto   20000\r\n2     2     Lee   32   HongKong    3000\r\n3     3  Steven   43   Bay Area    8300\r\n4     4     Ram   38  Hyderabad    3900\r\n<\/pre>\n
                                                                          \u89c2\u5bdf\u53ef\u4ee5\u770b\u5230\uff0c\u6807\u9898\u540d\u79f0\u9644\u52a0\u4e86\u81ea\u5b9a\u4e49\u540d\u79f0\uff0c\u4f46\u6587\u4ef6\u4e2d\u7684\u6807\u9898\u8fd8\u6ca1\u6709\u88ab\u6d88\u9664\u3002 \u73b0\u5728\uff0c\u4f7f\u7528header\u53c2\u6570\u6765\u5220\u9664\u5b83\u3002<\/p>\n
                                                                          \u5982\u679c\u6807\u9898\u4e0d\u662f\u7b2c\u4e00\u884c\uff0c\u5219\u5c06\u884c\u53f7\u4f20\u9012\u7ed9\u6807\u9898\u3002\u8fd9\u5c06\u8df3\u8fc7\u524d\u9762\u7684\u884c\u3002<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf=pd.read_csv(\"temp.csv\",names=['a','b','c','d','e'],header=0)\r\nprint (df)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                             a       b   c          d      e\r\n0  1     Tom  28    Toronto  20000\r\n1  2     Lee  32   HongKong   3000\r\n2  3  Steven  43   Bay Area   8300\r\n3  4     Ram  38  Hyderabad   3900\r\n<\/pre>\n
                                                                          skiprows<\/strong><\/p>\n
                                                                          skiprows\u8df3\u8fc7\u6307\u5b9a\u7684\u884c\u6570\u3002\u53c2\u8003\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801 –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf=pd.read_csv(\"temp.csv\", skiprows=2)\r\nprint (df)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          2     Lee  32   HongKong  3000\r\n0  3  Steven  43   Bay Area  8300\r\n1  4     Ram  38  Hyderabad  3900<\/pre>\n
                                                                          Python Pandas \u7a00\u758f\u6570\u636e<\/h1>\n
                                                                          \u5f53\u4efb\u4f55\u5339\u914d\u7279\u5b9a\u503c\u7684\u6570\u636e(NaN\/\u7f3a\u5931\u503c\uff0c\u5c3d\u7ba1\u53ef\u4ee5\u9009\u62e9\u4efb\u4f55\u503c)\u88ab\u7701\u7565\u65f6\uff0c\u7a00\u758f\u5bf9\u8c61\u88ab\u201c\u538b\u7f29\u201d\u3002 \u4e00\u4e2a\u7279\u6b8a\u7684SparseIndex\u5bf9\u8c61\u8ddf\u8e2a\u6570\u636e\u88ab\u201c\u7a00\u758f\u201d\u7684\u5730\u65b9\u3002 \u8fd9\u5c06\u5728\u4e00\u4e2a\u4f8b\u5b50\u4e2d\u66f4\u6709\u610f\u4e49\u3002 \u6240\u6709\u7684\u6807\u51c6Pandas\u6570\u636e\u7ed3\u6784\u90fd\u5e94\u7528\u4e86to_sparse<\/b>\u65b9\u6cd5 –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\nts = pd.Series(np.random.randn(10))\r\nts[2:-2] = np.nan\r\nsts = ts.to_sparse()\r\nprint (sts)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          0   -0.391926\r\n1   -1.774880\r\n2         NaN\r\n3         NaN\r\n4         NaN\r\n5         NaN\r\n6         NaN\r\n7         NaN\r\n8    0.642988\r\n9   -0.373698\r\ndtype: float64\r\nBlockIndex\r\nBlock locations: array([0, 8])\r\nBlock lengths: array([2, 2])\r\n<\/pre>\n
                                                                          \u4e3a\u4e86\u5185\u5b58\u6548\u7387\u7684\u539f\u56e0\uff0c\u6240\u4ee5\u9700\u8981\u7a00\u758f\u5bf9\u8c61\u7684\u5b58\u5728\u3002<\/p>\n
                                                                          \u73b0\u5728\u5047\u8bbe\u6709\u4e00\u4e2a\u5927\u7684NA DataFrame\u5e76\u6267\u884c\u4e0b\u9762\u7684\u4ee3\u7801 –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(10000, 4))\r\ndf.ix[:9998] = np.nan\r\nsdf = df.to_sparse()\r\n\r\nprint (sdf.density)\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          0.0001\r\n<\/pre>\n
                                                                          \u901a\u8fc7\u8c03\u7528to_dense<\/b>\u53ef\u4ee5\u5c06\u4efb\u4f55\u7a00\u758f\u5bf9\u8c61\u8f6c\u6362\u56de\u6807\u51c6\u5bc6\u96c6\u5f62\u5f0f –<\/p>\n
                                                                          import pandas as pd\r\nimport numpy as np\r\nts = pd.Series(np.random.randn(10))\r\nts[2:-2] = np.nan\r\nsts = ts.to_sparse()\r\nprint (sts.to_dense())\r\n<\/pre>\n
                                                                          \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                          0   -0.275846\r\n1    1.172722\r\n2         NaN\r\n3         NaN\r\n4         NaN\r\n5         NaN\r\n6         NaN\r\n7         NaN\r\n8   -0.612009\r\n9   -1.413996\r\ndtype: float64\r\n<\/pre>\n
                                                                          \u7a00\u758fDtypes<\/h3>\n
                                                                          \u7a00\u758f\u6570\u636e\u5e94\u8be5\u5177\u6709\u4e0e\u5176\u5bc6\u96c6\u8868\u793a\u76f8\u540c\u7684dtype\u3002 \u76ee\u524d\uff0c\u652f\u6301float64<\/b>\uff0cint64<\/b>\u548cbooldtypes<\/b>\u3002 \u53d6\u51b3\u4e8e\u539f\u59cb\u7684dtype<\/b>\uff0cfill_value<\/b>\u9ed8\u8ba4\u503c\u7684\u66f4\u6539 –<\/p>\n
                                                                            \n
                                                                          • float64<\/b> \u2212 np.nan<\/li>\n
                                                                          • int64<\/b> \u2212 0<\/li>\n
                                                                          • bool<\/b> \u2212 False<\/li>\n<\/ul>\n
                                                                            \u6267\u884c\u4e0b\u9762\u7684\u4ee3\u7801\u6765\u7406\u89e3\u76f8\u540c\u7684\u5185\u5bb9 –<\/p>\n
                                                                            import pandas as pd\r\nimport numpy as np\r\n\r\ns = pd.Series([1, np.nan, np.nan])\r\nprint (s)\r\nprint (\"=============================\")\r\ns.to_sparse()\r\nprint (s)\r\n<\/pre>\n
                                                                            \u6267\u884c\u4e0a\u9762\u793a\u4f8b\u4ee3\u7801\uff0c\u5f97\u5230\u4ee5\u4e0b\u7ed3\u679c –<\/p>\n
                                                                            0    1.0\r\n1    NaN\r\n2    NaN\r\ndtype: float64\r\n=============================\r\n0    1.0\r\n1    NaN\r\n2    NaN\r\ndtype: float64<\/pre>\n
                                                                            Python Pandas \u8b66\u544a\u548c\u6280\u5de7<\/h1>\n
                                                                            \u8b66\u544a\u548c\u7591\u96be\u610f\u5473\u7740\u4e00\u4e2a\u770b\u4e0d\u89c1\u7684\u95ee\u9898\u3002\u5728\u4f7f\u7528Pandas\u8fc7\u7a0b\u4e2d\uff0c\u9700\u8981\u7279\u522b\u6ce8\u610f\u7684\u5730\u65b9\u3002<\/p>\n
                                                                            \u4f7f\u7528If\/Truth\u8bed\u53e5<\/h3>\n
                                                                            \u5f53\u5c1d\u8bd5\u5c06\u67d0\u4e9b\u4e1c\u897f\u8f6c\u6362\u6210\u5e03\u5c14\u503c\u65f6\uff0cPandas\u9075\u5faa\u4e86\u4e00\u4e2a\u9519\u8bef\u7684\u60ef\u4f8b\u3002 \u8fd9\u79cd\u60c5\u51b5\u53d1\u751f\u5728\u4f7f\u7528\u5e03\u5c14\u8fd0\u7b97\u7684\u3002 \u76ee\u524d\u8fd8\u4e0d\u6e05\u695a\u7ed3\u679c\u662f\u4ec0\u4e48\u3002 \u5982\u679c\u5b83\u662f\u771f\u7684\uff0c\u56e0\u4e3a\u5b83\u4e0d\u662fzerolength\uff1f \u9519\u8bef\uff0c\u56e0\u4e3a\u6709\u9519\u8bef\u7684\u503c\uff1f \u76ee\u524d\u8fd8\u4e0d\u6e05\u695a\uff0cPandas\u63d0\u51fa\u4e86\u4e00\u4e2aValueError –<\/p>\n
                                                                            import pandas as pd\r\n\r\nif pd.Series([False, True, False]):\r\nprint 'I am True'\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool() a.item(), a.any() or a.all().<\/pre>\n
                                                                            \u5728if\u6761\u4ef6\u7684\u60c5\u51b5\u4e0b\uff0c\u4e0d\u6e05\u695a\u8be5\u5982\u4f55\u5904\u7406\u3002 \u9519\u8bef\u63d0\u793a\u662f\u5426\u4f7f\u7528None\u6216\u4efb\u4f55\u8fd9\u4e9b\u3002<\/p>\n
                                                                            import pandas as pd\r\nif pd.Series([False, True, False]).any():\r\n\tprint(\"I am any\")\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            I am any\r\n<\/pre>\n
                                                                            \u8981\u5728\u5e03\u5c14\u4e0a\u4e0b\u6587\u4e2d\u8bc4\u4f30\u5355\u5143\u7d20Pandas\u5bf9\u8c61\uff0c\u8bf7\u4f7f\u7528\u65b9\u6cd5.bool() –<\/p>\n
                                                                            import pandas as pd\r\nprint pd.Series([True]).bool()\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            True\r\n<\/pre>\n
                                                                            \u6309\u4f4d\u5e03\u5c14\u503c<\/h3>\n
                                                                            \u6309\u4f4d\u5e03\u5c14\u8fd0\u7b97\u7b26\uff08\u5982==\u548c\uff01=\uff09\u5c06\u8fd4\u56de\u4e00\u4e2a\u5e03\u5c14\u7cfb\u5217\u3002<\/p>\n
                                                                            import pandas as pd\r\n\r\ns = pd.Series(range(5))\r\nprint s==4\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            0 False\r\n1 False\r\n2 False\r\n3 False\r\n4 True\r\ndtype: bool\r\n<\/pre>\n
                                                                            isin\u64cd\u4f5c\u7b26<\/h3>\n
                                                                            \u8fd9\u5c06\u8fd4\u56de\u4e00\u4e2a\u5e03\u5c14\u7cfb\u5217\uff0c\u663e\u793a\u7cfb\u5217\u4e2d\u7684\u6bcf\u4e2a\u5143\u7d20\u662f\u5426\u5b8c\u5168\u5305\u542b\u5728\u4f20\u9012\u7684\u503c\u5e8f\u5217\u4e2d\u3002<\/p>\n
                                                                            import pandas as pd\r\n\r\ns = pd.Series(list('abc'))\r\ns = s.isin(['a', 'c', 'e'])\r\nprint s\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            0 True\r\n1 False\r\n2 True\r\ndtype: bool\r\n<\/pre>\n
                                                                            \u91cd\u6784\u7d22\u5f15\u4e0eix\u9677\u9631<\/h3>\n
                                                                            \u8bb8\u591a\u7528\u6237\u4f1a\u53d1\u73b0\u81ea\u5df1\u4f7f\u7528ix\u7d22\u5f15\u529f\u80fd\u4f5c\u4e3a\u4ecePandas\u5bf9\u8c61\u4e2d\u9009\u62e9\u6570\u636e\u7684\u7b80\u6d01\u65b9\u6cd5 –<\/p>\n
                                                                            import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(6, 4), columns=['one', 'two', 'three',\r\n'four'],index=list('abcdef'))\r\n\r\nprint df\r\nprint df.ix[['b', 'c', 'e']]\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            one        two      three       four\r\na   -1.582025   1.335773   0.961417  -1.272084\r\nb    1.461512   0.111372  -0.072225   0.553058\r\nc   -1.240671   0.762185   1.511936  -0.630920\r\nd   -2.380648  -0.029981   0.196489   0.531714\r\ne    1.846746   0.148149   0.275398  -0.244559\r\nf   -1.842662  -0.933195   2.303949   0.677641\r\n\r\none        two      three       four\r\nb    1.461512   0.111372  -0.072225   0.553058\r\nc   -1.240671   0.762185   1.511936  -0.630920\r\ne    1.846746   0.148149   0.275398  -0.244559\r\n<\/pre>\n
                                                                            \u8fd9\u5f53\u7136\u5728\u8fd9\u79cd\u60c5\u51b5\u4e0b\u5b8c\u5168\u7b49\u540c\u4e8e\u4f7f\u7528reindex\u65b9\u6cd5 –<\/p>\n
                                                                            import pandas as pd\r\nimport numpy as np\r\ndf = pd.DataFrame(np.random.randn(6, 4), columns=['one', 'two', 'three',\r\n'four'],index=list('abcdef'))\r\nprint df\r\nprint df.reindex(['b', 'c', 'e'])\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            one        two      three       four\r\na    1.639081   1.369838   0.261287  -1.662003\r\nb   -0.173359   0.242447  -0.494384   0.346882\r\nc   -0.106411   0.623568   0.282401  -0.916361\r\nd   -1.078791  -0.612607  -0.897289  -1.146893\r\ne    0.465215   1.552873  -1.841959   0.329404\r\nf    0.966022  -0.190077   1.324247   0.678064\r\n\r\none        two      three       four\r\nb   -0.173359   0.242447  -0.494384   0.346882\r\nc   -0.106411   0.623568   0.282401  -0.916361\r\ne    0.465215   1.552873  -1.841959   0.329404\r\n<\/pre>\n
                                                                            \u6709\u4eba\u53ef\u80fd\u4f1a\u5f97\u51fa\u8fd9\u6837\u7684\u7ed3\u8bba\uff0cix\u548creindex\u662f\u57fa\u4e8e\u8fd9\u4e2a100\uff05\u7684\u7b49\u4ef7\u7269\u3002 \u9664\u4e86\u6574\u6570\u7d22\u5f15\u7684\u60c5\u51b5\uff0c\u5b83\u662ftrue\u3002\u4f8b\u5982\uff0c\u4e0a\u8ff0\u64cd\u4f5c\u53ef\u9009\u5730\u8868\u793a\u4e3a –<\/p>\n
                                                                            import pandas as pd\r\nimport numpy as np\r\n\r\ndf = pd.DataFrame(np.random.randn(6, 4), columns=['one', 'two', 'three',\r\n'four'],index=list('abcdef'))\r\n\r\nprint df\r\nprint df.ix[[1, 2, 4]]\r\nprint df.reindex([1, 2, 4])\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                                one        two      three       four\r\na   -1.015695  -0.553847   1.106235  -0.784460\r\nb   -0.527398  -0.518198  -0.710546  -0.512036\r\nc   -0.842803  -1.050374   0.787146   0.205147\r\nd   -1.238016  -0.749554  -0.547470  -0.029045\r\ne   -0.056788   1.063999  -0.767220   0.212476\r\nf    1.139714   0.036159   0.201912   0.710119\r\n\t\r\n\tone        two      three       four\r\nb   -0.527398  -0.518198  -0.710546  -0.512036\r\nc   -0.842803  -1.050374   0.787146   0.205147\r\ne   -0.056788   1.063999  -0.767220   0.212476\r\n\t\r\n\tone  two  three  four\r\n1   NaN  NaN    NaN   NaN\r\n2   NaN  NaN    NaN   NaN\r\n4   NaN  NaN    NaN   NaN\r\n<\/pre>\n
                                                                            \u91cd\u8981\u7684\u662f\u8981\u8bb0\u4f4f\uff0creindex\u53ea\u662f\u4e25\u683c\u7684\u6807\u7b7e\u7d22\u5f15\u3002\u8fd9\u53ef\u80fd\u4f1a\u5bfc\u81f4\u4e00\u4e9b\u6f5c\u5728\u7684\u4ee4\u4eba\u60ca\u8bb6\u7684\u7ed3\u679c\uff0c\u4f8b\u5982\u7d22\u5f15\u5305\u542b\u6574\u6570\u548c\u5b57\u7b26\u4e32\u7684\u75c5\u6001\u60c5\u51b5\u3002<\/p>\n
                                                                            Python Pandas \u4e0eSQL\u6bd4\u8f83<\/h1>\n
                                                                            \u7531\u4e8e\u8bb8\u591a\u6f5c\u5728\u7684Pandas\u7528\u6237\u5bf9SQL\u6709\u4e00\u5b9a\u7684\u4e86\u89e3\uff0c\u56e0\u6b64\u672c\u9875\u65e8\u5728\u63d0\u4f9b\u4e00\u4e9b\u4f7f\u7528pandas\u6267\u884c\u5404\u79cdSQL\u64cd\u4f5c\u7684\u793a\u4f8b\u3002<\/p>\n
                                                                            import pandas as pd\r\nurl = 'https:\/\/raw.github.com\/pandasdev\/\r\npandas\/master\/pandas\/tests\/data\/tips.csv'\r\ntips=pd.read_csv(url)\r\nprint tips.head()\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                                total_bill   tip      sex  smoker  day     time  size\r\n0        16.99  1.01   Female      No  Sun  Dinner      2\r\n1        10.34  1.66     Male      No  Sun  Dinner      3\r\n2        21.01  3.50     Male      No  Sun  Dinner      3\r\n3        23.68  3.31     Male      No  Sun  Dinner      2\r\n4        24.59  3.61   Female      No  Sun  Dinner      4\r\n<\/pre>\n
                                                                            SELECT<\/h3>\n
                                                                            \u5728SQL\u4e2d\uff0c\u9009\u62e9\u662f\u4f7f\u7528\u60a8\u9009\u62e9\u7684\u5217\uff08\u6216*\u6765\u9009\u62e9\u6240\u6709\u5217\uff09\u7684\u9017\u53f7\u5206\u9694\u5217\u8868\u5b8c\u6210\u7684 –<\/p>\n
                                                                            SELECT total_bill, tip, smoker, time\r\nFROM tips\r\nLIMIT 5;\r\n<\/pre>\n
                                                                            \u5bf9\u4e8ePandas\u6765\u8bf4\uff0c\u5217\u9009\u62e9\u662f\u901a\u8fc7\u5c06\u5217\u540d\u5217\u8868\u4f20\u9012\u7ed9DataFrame\u6765\u5b8c\u6210\u7684 –<\/p>\n
                                                                            tips[['total_bill', 'tip', 'smoker', 'time']].head(5)\r\n<\/pre>\n
                                                                            \u8ba9\u6211\u4eec\u6765\u770b\u770b\u5b8c\u6574\u7684\u7a0b\u5e8f –<\/p>\n
                                                                            import pandas as pd\r\n\r\nurl = 'https:\/\/raw.github.com\/pandasdev\/\r\npandas\/master\/pandas\/tests\/data\/tips.csv'\r\n\r\ntips=pd.read_csv(url)\r\nprint tips[['total_bill', 'tip', 'smoker', 'time']].head(5)\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                                total_bill   tip  smoker     time\r\n0       16.99  1.01      No   Dinner\r\n1       10.34  1.66      No   Dinner\r\n2       21.01  3.50      No   Dinner\r\n3       23.68  3.31      No   Dinner\r\n4       24.59  3.61      No   Dinner\r\n<\/pre>\n
                                                                            \u8c03\u7528\u6ca1\u6709\u5217\u540d\u79f0\u5217\u8868\u7684DataFrame\u5c06\u663e\u793a\u6240\u6709\u5217\uff08\u7c7b\u4f3c\u4e8eSQL\u7684*\uff09\u3002<\/p>\n
                                                                            WHERE<\/h3>\n
                                                                            SQL\u4e2d\u7684\u8fc7\u6ee4\u901a\u8fc7WHERE\u5b50\u53e5\u5b8c\u6210\u3002<\/p>\n
                                                                            SELECT * FROM tips WHERE time = 'Dinner' LIMIT 5;\r\n<\/pre>\n
                                                                            DataFrame\u53ef\u4ee5\u901a\u8fc7\u591a\u79cd\u65b9\u5f0f\u8fdb\u884c\u8fc7\u6ee4; \u5176\u4e2d\u6700\u76f4\u89c2\u7684\u662f\u4f7f\u7528\u5e03\u5c14\u7d22\u5f15\u3002<\/p>\n
                                                                            tips[tips['time'] == 'Dinner'].head(5)\r\n<\/pre>\n
                                                                            \u8ba9\u6211\u4eec\u6765\u770b\u770b\u5b8c\u6574\u7684\u4f8b\u5b50 –<\/p>\n
                                                                            import pandas as pd\r\n\r\nurl = 'https:\/\/raw.github.com\/pandasdev\/\r\npandas\/master\/pandas\/tests\/data\/tips.csv'\r\n\r\ntips=pd.read_csv(url)\r\nprint tips[tips['time'] == 'Dinner'].head(5)\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                                total_bill   tip      sex  smoker  day    time  size\r\n0       16.99  1.01   Female     No   Sun  Dinner    2\r\n1       10.34  1.66     Male     No   Sun  Dinner    3\r\n2       21.01  3.50     Male     No   Sun  Dinner    3\r\n3       23.68  3.31     Male     No   Sun  Dinner    2\r\n4       24.59  3.61   Female     No   Sun  Dinner    4\r\n<\/pre>\n
                                                                            \u4e0a\u8ff0\u8bed\u53e5\u5c06Series True \/ False\u5bf9\u8c61\u4f20\u9012\u7ed9DataFrame\uff0c\u5e76\u5c06\u6240\u6709\u884c\u8fd4\u56deTrue\u3002<\/p>\n
                                                                            GroupBy<\/h3>\n
                                                                            \u6b64\u64cd\u4f5c\u5c06\u83b7\u53d6\u6574\u4e2a\u6570\u636e\u96c6\u4e2d\u6bcf\u4e2a\u7ec4\u4e2d\u7684\u8bb0\u5f55\u6570\u3002 \u4f8b\u5982\uff0c\u4e00\u4e2a\u67e5\u8be2\u4e0d\u540c\u6027\u522b\u7684\u603b\u6570 –<\/p>\n
                                                                            SELECT sex, count(*)\r\nFROM tips\r\nGROUP BY sex;\r\n<\/pre>\n
                                                                            Pandas\u7684\u8bed\u53e5\u662f<\/p>\n
                                                                            tips.groupby('sex').size()\r\n<\/pre>\n
                                                                            \u8ba9\u6211\u4eec\u6765\u770b\u770b\u5b8c\u6574\u7684\u4f8b\u5b50 –<\/p>\n
                                                                            import pandas as pd\r\n\r\nurl = 'https:\/\/raw.github.com\/pandasdev\/\r\npandas\/master\/pandas\/tests\/data\/tips.csv'\r\n\r\ntips=pd.read_csv(url)\r\nprint tips.groupby('sex').size()\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                            sex\r\nFemale   87\r\nMale    157\r\ndtype: int64\r\n<\/pre>\n
                                                                            Top N rows<\/h3>\n
                                                                            SQL\u4f7f\u7528LIMIT\u8fd4\u56de\u524dn\u884c –<\/p>\n
                                                                            SELECT * FROM tips\r\nLIMIT 5 ;\r\n<\/pre>\n
                                                                            Pandas\u7684\u8bed\u53e5\u662f<\/p>\n
                                                                            tips.head(5)\r\n<\/pre>\n
                                                                            \u8ba9\u6211\u4eec\u6765\u770b\u770b\u5b8c\u6574\u7684\u4f8b\u5b50 –<\/p>\n
                                                                            import pandas as pd\r\nurl = 'https:\/\/raw.github.com\/pandas-dev\/pandas\/master\/pandas\/tests\/data\/tips.csv'\r\n\r\ntips=pd.read_csv(url)\r\ntips = tips[['smoker', 'day', 'time']].head(5)\r\nprint tips\r\n<\/pre>\n
                                                                            \u5176\u8f93\u51fa\u5982\u4e0b –<\/p>\n
                                                                                smoker   day     time\r\n0      No   Sun   Dinner\r\n1      No   Sun   Dinner\r\n2      No   Sun   Dinner\r\n3      No   Sun   Dinner\r\n4      No   Sun   Dinner\r\n<\/pre>\n
                                                                            \u8fd9\u4e9b\u662f\u6211\u4eec\u6bd4\u8f83\u7684\u51e0\u4e2a\u57fa\u672c\u64cd\u4f5c\uff0c\u6211\u4eec\u5728\u524d\u9762\u7684Pandas\u5e93\u7684\u7ae0\u8282\u4e2d\u4e86\u89e3\u5230\u3002<\/p>\n
                                                                             <\/p>\n\n
                                                                            <\/div>","protected":false},"excerpt":{"rendered":"
                                                                            Python Pandas \u4ecb\u7ecd Pandas\u662f\u4e00\u4e2a\u5f00\u6e90\u7684Python\u5e93\uff0c\u4f7f\u7528\u5176\u5f3a\u5927\u7684\u6570\u636e\u7ed3\u6784\u63d0\u4f9b\u9ad8\u6027\u80fd\u7684\u6570\u636e\u5904\u7406\u548c\u5206\u6790\u5de5\u5177\u3002 Pandas\u8fd9\u4e2a\u540d\u5b57\u6e90\u81ea\u9762\u677f\u6570\u636e – \u6765\u81ea\u591a\u7ef4\u6570\u636e\u7684\u8ba1\u91cf\u7ecf\u6d4e\u5b66\u3002 2008\u5e74\uff0c\u5f00\u53d1\u4eba\u5458Wes McKinney\u5728\u9700\u8981\u9ad8\u6027\u80fd\uff0c\u7075\u6d3b\u7684\u6570\u636e\u5206\u6790\u5de5\u5177\u65f6\u5f00\u59cb\u5f00\u53d1Pandas …<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_bbp_topic_count":0,"_bbp_reply_count":0,"_bbp_total_topic_count":0,"_bbp_total_reply_count":0,"_bbp_voice_count":0,"_bbp_anonymous_reply_count":0,"_bbp_topic_count_hidden":0,"_bbp_reply_count_hidden":0,"_bbp_forum_subforum_count":0},"categories":[12],"tags":[],"_links":{"self":[{"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/posts\/432"}],"collection":[{"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/comments?post=432"}],"version-history":[{"count":0,"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/posts\/432\/revisions"}],"wp:attachment":[{"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/media?parent=432"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/categories?post=432"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/zhang.mba\/index.php\/wp-json\/wp\/v2\/tags?post=432"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}