{"id":115,"date":"2026-08-06T00:31:09","date_gmt":"2026-08-05T16:31:09","guid":{"rendered":"https:\/\/ltrainyard.icu\/?p=115"},"modified":"2026-08-06T00:45:59","modified_gmt":"2026-08-05T16:45:59","slug":"day-24-%e6%95%b0%e6%8d%ae%e6%b8%85%e6%b4%97%ef%bc%9a%e7%bc%ba%e5%a4%b1%e5%80%bc%e3%80%81%e9%87%8d%e5%a4%8d%e5%80%bc%e3%80%81%e7%b1%bb%e5%9e%8b%e8%bd%ac%e6%8d%a2%e4%b8%8e%e5%bc%82%e5%b8%b8%e5%80%bc","status":"publish","type":"post","link":"https:\/\/ltrainyard.icu\/index.php\/2026\/08\/06\/day-24-%e6%95%b0%e6%8d%ae%e6%b8%85%e6%b4%97%ef%bc%9a%e7%bc%ba%e5%a4%b1%e5%80%bc%e3%80%81%e9%87%8d%e5%a4%8d%e5%80%bc%e3%80%81%e7%b1%bb%e5%9e%8b%e8%bd%ac%e6%8d%a2%e4%b8%8e%e5%bc%82%e5%b8%b8%e5%80%bc\/","title":{"rendered":"| Day 24 | \u6570\u636e\u6e05\u6d17\uff1a\u7f3a\u5931\u503c\u3001\u91cd\u590d\u503c\u3001\u7c7b\u578b\u8f6c\u6362\u4e0e\u5f02\u5e38\u503c |"},"content":{"rendered":"\n

1 \u7f3a\u5931\u503c \u2014\u2014 \u6570\u636e\u91cc\u7684\u300c\u7a7a\u4f4d\u300d<\/h3>\n\n\n\n

\u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f<\/strong><\/p>\n\n\n\n

\u7f3a\u5931\u503c<\/strong>\u5c31\u662f\u8868\u683c\u91cc\u300c\u6ca1\u586b\u300d\u7684\u683c\u5b50\uff0cPandas \u91cc\u663e\u793a\u4e3a NaN<\/code>\uff08Not a Number\uff09\u3002\u771f\u5b9e\u6570\u636e\u51e0\u4e4e\u5fc5\u6709\u7f3a\u5931\uff1a\u7528\u6237\u6ca1\u586b\u7684\u9009\u9879\u3001\u4f20\u611f\u5668\u6ca1\u91c7\u5230\u7684\u70b9\u3001\u5408\u5e76\u8868\u683c\u65f6\u5bf9\u4e0d\u4e0a\u7684\u884c\u2026\u2026\u5982\u679c\u4e0d\u5904\u7406\uff0c\u7edf\u8ba1\u7ed3\u679c\u4f1a\u504f\uff08\u6bd4\u5982\u7b97\u5e73\u5747\u5206\u65f6 NaN \u53c2\u4e0e\u8ba1\u7b97\u76f4\u63a5\u53d8 NaN\uff09\uff0c\u753b\u56fe\u4f1a\u65ad\u7ebf\u3002\u6240\u4ee5\u62ff\u5230\u6570\u636e\u7b2c\u4e00\u6b65\u5c31\u662f\u300c\u67e5\u7f3a\u300d\u3002\u771f\u5b9e\u4e16\u754c\u7684\u6570\u636e\u6c38\u8fdc\u662f\u810f\u7684\uff0c\u6570\u636e\u6e05\u6d17<\/strong>\u662f\u6570\u636e\u5206\u6790\u4e2d\u6700\u82b1\u65f6\u95f4\u7684\u90e8\u5206\uff0c\u4e5f\u662f\u51b3\u5b9a\u5206\u6790\u7ed3\u679c\u51c6\u4e0d\u51c6\u7684\u5173\u952e\u3002<\/p>\n\n\n\n

\u5e95\u5c42\u539f\u7406<\/strong><\/p>\n\n\n\n

NaN<\/code> \u662f Pandas \u7528\u6765\u8868\u793a\u300c\u7f3a\u5931\u300d\u7684\u7279\u6b8a\u6d6e\u70b9\u503c<\/strong>\uff1a\u4efb\u4f55\u8fd0\u7b97\u78b0\u5230 NaN \u90fd\u4f1a\u4f20\u67d3\u6210 NaN\uff0890 + NaN = NaN<\/code>\uff09\uff0c\u6240\u4ee5\u4e0d\u80fd\u9760\u300c\u7b97\u300d\u6765\u7ed5\u8fc7\u5b83\uff0c\u5fc5\u987b\u663e\u5f0f\u5904\u7406<\/strong>\u3002\u4e09\u4e2a\u5de5\u5177\u5404\u7ba1\u4e00\u4ef6\u4e8b\uff1aisnull()<\/code> \u628a\u6bcf\u4e2a\u683c\u5b50\u53d8\u6210 True\/False\uff08True = \u7f3a\u5931\uff09\uff0c\u914d\u5408 sum()<\/code> \u76f4\u63a5\u6570\u51fa\u6bcf\u5217\u7f3a\u4e86\u51e0\u4e2a\uff1bdropna()<\/code> \u5220\u6389\u542b\u7f3a\u5931\u7684\u884c\uff1bfillna(\u503c)<\/code> \u7528\u6307\u5b9a\u503c\u586b\u7a7a\u3002\u600e\u4e48\u586b\u6709\u8bb2\u7a76\uff1a\u6570\u503c\u5217\u7528\u5e73\u5747\u503c \/ \u4e2d\u4f4d\u6570\uff08\u4e0d\u53d7\u6781\u7aef\u503c\u5f71\u54cd\uff09\uff0c\u6587\u672c\u5217\u7528\u56fa\u5b9a\u503c\uff08\u5982\u300c\u672a\u77e5\u300d\uff09\u3002<\/p>\n\n\n\n

\u751f\u6d3b\u7c7b\u6bd4<\/strong><\/p>\n\n\n\n

\u50cf\u4e00\u5f20\u6ca1\u586b\u5b8c\u7684\u62a5\u540d\u8868<\/strong>\uff1aisnull<\/code> \u662f\u300c\u9010\u683c\u68c0\u67e5\u6709\u6ca1\u6709\u7a7a\u7740\u300d\uff0cdropna<\/code> \u662f\u300c\u8fd9\u5f20\u8868\u7a7a\u5f97\u592a\u591a\uff0c\u4f5c\u5e9f\u91cd\u586b\u300d\uff0cfillna<\/code> \u662f\u300c\u5c31\u7a7a\u4e00\u683c\uff1f\u6211\u5e2e\u4f60\u8865\u4e2a’\u65e0’\u6216\u586b\u4e2a\u5e73\u5747\u53c2\u8003\u503c\u300d\u3002<\/p>\n\n\n\n

\u6ce8\u610f\u4e8b\u9879\u603b\u7ed3<\/strong><\/p>\n\n\n\n

1 \u53d1\u73b0\u7f3a\u5931 \u2014\u2014 isnull \u9010\u683c\u68c0\u67e5 + \u6570\u51fa\u7f3a\u5931\u4e2a\u6570<\/strong>\uff1a<\/p>\n\n\n\n

import pandas as pd\nimport numpy as np\n\ndf = pd.DataFrame({\n    \"\u59d3\u540d\": [\"\u5f20\u4e09\", \"\u674e\u56db\", \"\u738b\u4e94\"],\n    \"\u5206\u6570\": [90, np.nan, 95],\n    \"\u57ce\u5e02\": [\"\u5317\u4eac\", np.nan, \"\u5e7f\u5dde\"],\n})\nprint(df)\nprint(df.isnull())           # \u6bcf\u4e2a\u683c\u5b50\u662f\u4e0d\u662f\u7f3a\u5931\nprint(\"\u5206\u6570\u7f3a\u5931\u4e2a\u6570:\", df[\"\u5206\u6570\"].isnull().sum())\n<\/code><\/pre>\n\n\n\n
\u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
   \u59d3\u540d    \u5206\u6570   \u57ce\u5e02\n0  \u5f20\u4e09  90.0   \u5317\u4eac\n1  \u674e\u56db   NaN  NaN\n2  \u738b\u4e94  95.0   \u5e7f\u5dde\n---\n      \u59d3\u540d     \u5206\u6570     \u57ce\u5e02\n0  False  False  False\n1  False   True   True\n2  False  False  False\n---\n\u5206\u6570\u7f3a\u5931\u4e2a\u6570: 1\n<\/code><\/pre>\n\n\n\n
np.nan<\/code> \u5c31\u662f\u300c\u9020\u4e00\u4e2a\u7f3a\u5931\u503c\u300d\uff1b\u6ce8\u610f\u674e\u56db\u7684\u5206\u6570\u5217\u53d8\u6210 90.0<\/code> \/ NaN<\/code> \/ 95.0<\/code>\u2014\u2014\u6574\u5217\u56e0\u4e3a\u6df7\u5165 NaN \u81ea\u52a8\u53d8\u6210\u6d6e\u70b9<\/strong>\uff1bdf[\"\u5206\u6570\"].isnull().sum()<\/code> \u662f\u7ecf\u5178\u7ec4\u5408\uff1a\u5148\u53d8\u5e03\u5c14\u518d\u6c42\u548c\uff0c\u6570\u51fa\u8fd9\u4e00\u5217\u7f3a\u4e86\u51e0\u4e2a\u3002<\/p>\n\n\n\n
2 \u5904\u7406\u7f3a\u5931 \u2014\u2014 dropna \u5220\u9664 \/ fillna \u586b\u5145<\/strong>\uff1a<\/p>\n\n\n\n
print(\"=== dropna \u5220\u9664\u6240\u6709\u542b\u7f3a\u5931\u7684\u884c ===\")\nprint(df.dropna())\n\ndf2 = pd.DataFrame({\"\u59d3\u540d\": [\"\u5f20\u4e09\", \"\u674e\u56db\", \"\u738b\u4e94\"], \"\u5206\u6570\": [90, np.nan, 95]})\ndf2[\"\u5206\u6570\"] = df2[\"\u5206\u6570\"].fillna(df2[\"\u5206\u6570\"].mean())   # \u7528\u5e73\u5747\u5206\u586b\nprint(\"\\n=== fillna \u7528\u5e73\u5747\u503c\u586b\u5145 ===\")\nprint(df2)\n<\/code><\/pre>\n\n\n\n
\u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
=== dropna \u5220\u9664\u6240\u6709\u542b\u7f3a\u5931\u7684\u884c ===\n   \u59d3\u540d    \u5206\u6570  \u57ce\u5e02\n0  \u5f20\u4e09  90.0  \u5317\u4eac\n2  \u738b\u4e94  95.0  \u5e7f\u5dde\n\n=== fillna \u7528\u5e73\u5747\u503c\u586b\u5145 ===\n   \u59d3\u540d    \u5206\u6570\n0  \u5f20\u4e09  90.0\n1  \u674e\u56db  92.5\n2  \u738b\u4e94  95.0\n<\/code><\/pre>\n\n\n\n
df.dropna()<\/code> \u9ed8\u8ba4\u5220\u6389\u4efb\u4f55\u542b\u7f3a\u5931\u7684\u884c<\/strong>\uff1bdf2[\"\u5206\u6570\"].fillna(df2[\"\u5206\u6570\"].mean())<\/code> \u5148\u7b97\u5e73\u5747\u503c\uff0890 \u548c 95 \u5e73\u5747\u662f 92.5\uff09\uff0c\u518d\u586b\u8fdb\u7f3a\u5931\u683c\u2014\u2014\u6ce8\u610f fillna \u4e0d\u4f1a\u6539\u539f\u8868\uff0c\u5fc5\u987b\u91cd\u65b0\u8d4b\u503c<\/strong>\u3002\u60f3\u53ea\u6309\u67d0\u4e00\u5217\u5224\u65ad\u5220\u9664\uff0c\u7528 dropna(subset=[\"\u5206\u6570\"])<\/code>\u3002<\/p>\n\n\n\n
\u6613\u9519\u70b9<\/strong><\/p>\n\n\n\n
    \n
  • \u9519\u8bef\u5199\u6cd5\uff1adf.fillna(0)<\/code> \u540e\u4ee5\u4e3a\u539f\u8868\u53d8\u4e86 \u2192 \u95ee\u9898\uff1afillna \/ dropna \u9ed8\u8ba4\u8fd4\u56de\u65b0\u8868<\/strong>\uff0c\u4e0d\u6539\u539f\u8868 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u91cd\u65b0\u8d4b\u503c\uff1adf = df.fillna(0)<\/code>\uff0c\u6216\u52a0\u53c2\u6570 inplace=True<\/code>\u3002<\/li>\n\n\n\n
  • \u9519\u8bef\u5199\u6cd5\uff1a\u6240\u6709\u7f3a\u5931\u90fd\u586b 0 \u2192 \u95ee\u9898\uff1a\u6587\u672c\u5217\u586b 0 \u4e0d\u4f26\u4e0d\u7c7b\uff0c\u6570\u503c\u5217\u586b 0 \u4f1a\u62c9\u4f4e\u5e73\u5747\u503c \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u6570\u503c\u5217\u586b\u5e73\u5747\u503c\/\u4e2d\u4f4d\u6570\uff0c\u6587\u672c\u5217\u586b\u300c\u672a\u77e5\u300d\u7b49\u56fa\u5b9a\u503c\u3002<\/li>\n\n\n\n
  • \u9519\u8bef\u5199\u6cd5\uff1adf.isnull()<\/code> \u540e\u76f4\u63a5\u770b \u2192 \u95ee\u9898\uff1a\u8fd4\u56de\u7684\u662f True\/False \u8868\uff0c\u4e0d\u662f\u7f3a\u5931\u6570\u636e\u672c\u8eab \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u60f3\u7b5b\u51fa\u7f3a\u5931\u884c\uff1adf[df[\"\u5206\u6570\"].isnull()]<\/code>\u3002<\/li>\n\n\n\n
  • \u9519\u8bef\u5199\u6cd5\uff1a\u7f3a\u5931\u8d85\u8fc7\u4e00\u534a\u8fd8 fillna \u2192 \u95ee\u9898\uff1a\u586b\u5145\u7b49\u4e8e\u300c\u7f16\u6570\u636e\u300d\uff0c\u7edf\u8ba1\u7ed3\u679c\u4e25\u91cd\u5931\u771f \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u7f3a\u5931\u4e25\u91cd\uff08>50%\uff09\u76f4\u63a5 dropna<\/code> \u5220\u9664\u8be5\u5217\u6216\u8be5\u884c\u3002<\/li>\n<\/ul>\n\n\n\n
    \u8bb0\u5fc6\u53e3\u8bc0<\/strong>\uff1a\u5148 isnull \u67e5\u7f3a\uff0c\u518d dropna \u5220\u3001fillna \u586b\uff1b\u6570\u503c\u586b\u5747\u503c\uff0c\u6587\u672c\u586b\u672a\u77e5\uff1b\u7f3a\u5f97\u591a\u5c31\u5220\u3002<\/p>\n\n\n\n
    2 \u91cd\u590d\u503c \u2014\u2014 \u6570\u636e\u91cc\u7684\u300c\u590d\u5370\u673a\u300d<\/h3>\n\n\n\n
    \u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f<\/strong><\/p>\n\n\n\n
    \u91cd\u590d\u884c<\/strong>\u5c31\u662f\u4e00\u6a21\u4e00\u6837\u7684\u8bb0\u5f55\u51fa\u73b0\u591a\u6b21\uff1a\u7528\u6237\u91cd\u590d\u63d0\u4ea4\u7684\u8ba2\u5355\u3001\u4f20\u611f\u5668\u91cd\u590d\u4e0a\u62a5\u3001\u5bfc\u51fa\u8868\u683c\u81ea\u5e26\u91cd\u590d\u884c\u3002\u91cd\u590d\u884c\u4f1a\u8ba9\u7edf\u8ba1\u300c\u865a\u9ad8\u300d\u2014\u2014\u6570\u4eba\u6570\u3001\u7b97\u9500\u91cf\u5168\u90fd\u4f1a\u591a\u7b97\u3002drop_duplicates()<\/code> \u4e00\u884c\u53bb\u91cd\uff0c\u662f\u5bfc\u5165\u6570\u636e\u540e\u7684\u6807\u51c6\u7b2c\u4e00\u6b65\u3002<\/p>\n\n\n\n
    \u5e95\u5c42\u539f\u7406<\/strong><\/p>\n\n\n\n
    drop_duplicates()<\/code> \u9010\u884c\u6bd4\u8f83\u6574\u884c\u5185\u5bb9\uff0c\u5b8c\u5168\u4e00\u6837<\/strong>\u7684\u884c\u53ea\u4fdd\u7559\u7b2c\u4e00\u6761\uff08\u9ed8\u8ba4 keep=\"first\"<\/code>\uff09\uff0c\u5176\u4f59\u5220\u9664\uff1b\u52a0 subset=[\"\u52171\", \"\u52172\"]<\/code> \u5c31\u53ea\u6309\u8fd9\u51e0\u5217\u5224\u65ad\u91cd\u590d\uff08\u6bd4\u5982\u300c\u540c\u4e00\u59d3\u540d\u53ea\u7559\u4e00\u6761\u300d\uff09\uff0c\u5176\u4f59\u5217\u4e0d\u540c\u4e5f\u7b97\u91cd\u590d\uff1bkeep=\"last\"<\/code> \u4fdd\u7559\u6700\u540e\u4e00\u6761\u3002\u53bb\u91cd\u9ed8\u8ba4\u8fd4\u56de\u65b0\u8868<\/strong>\uff0c\u540c\u6837\u8981\u91cd\u65b0\u8d4b\u503c\u3002<\/p>\n\n\n\n
    \u751f\u6d3b\u7c7b\u6bd4<\/strong><\/p>\n\n\n\n
    \u50cf\u6e05\u70b9\u5feb\u9012\u67dc\u91cc\u7684\u91cd\u590d\u5305\u88f9<\/strong>\uff1adrop_duplicates()<\/code> \u662f\u300c\u4e00\u6a21\u4e00\u6837\u7684\u4e24\u4ef6\u53ea\u7559\u4e00\u4ef6\u300d\uff1bsubset<\/code> \u662f\u300c\u53ea\u770b\u6536\u4ef6\u4eba\uff0c\u540c\u4e00\u4eba\u5c31\u7b97\u91cd\u590d\u300d\uff08\u54ea\u6015\u5305\u88f9\u5927\u5c0f\u4e0d\u540c\uff09\uff1bkeep<\/code> \u662f\u300c\u7559\u7b2c\u4e00\u4ef6\u8fd8\u662f\u7559\u6700\u540e\u4e00\u4ef6\u300d\u3002<\/p>\n\n\n\n
    \u6ce8\u610f\u4e8b\u9879\u603b\u7ed3<\/strong><\/p>\n\n\n\n
    1 \u6574\u884c\u53bb\u91cd + \u6309\u5217\u53bb\u91cd<\/strong>\uff1a<\/p>\n\n\n\n
    import pandas as pd\n\ndf = pd.DataFrame({\n    \"\u59d3\u540d\": [\"\u5f20\u4e09\", \"\u674e\u56db\", \"\u5f20\u4e09\", \"\u738b\u4e94\", \"\u674e\u56db\"],\n    \"\u65e5\u671f\": [\"2026-08-01\", \"2026-08-01\", \"2026-08-02\", \"2026-08-02\", \"2026-08-01\"],\n})\nprint(\"=== \u5220\u9664\u6574\u884c\u91cd\u590d ===\")\nprint(df.drop_duplicates())\nprint(\"\\n=== \u53ea\u6309\u59d3\u540d\u5224\u65ad\u91cd\u590d ===\")\nprint(df.drop_duplicates(subset=[\"\u59d3\u540d\"]))\n<\/code><\/pre>\n\n\n\n
    \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
    === \u5220\u9664\u6574\u884c\u91cd\u590d ===\n   \u59d3\u540d          \u65e5\u671f\n0  \u5f20\u4e09  2026-08-01\n1  \u674e\u56db  2026-08-01\n2  \u5f20\u4e09  2026-08-02\n3  \u738b\u4e94  2026-08-02\n\n=== \u53ea\u6309\u59d3\u540d\u5224\u65ad\u91cd\u590d ===\n   \u59d3\u540d          \u65e5\u671f\n0  \u5f20\u4e09  2026-08-01\n1  \u674e\u56db  2026-08-01\n3  \u738b\u4e94  2026-08-02\n<\/code><\/pre>\n\n\n\n
    \u7b2c 0 \u884c\u548c\u7b2c 4 \u884c\u6574\u884c\u5b8c\u5168\u4e00\u6837<\/strong>\uff0c\u9ed8\u8ba4\u53bb\u91cd\u540e\u53ea\u5269\u7b2c\u4e00\u6761\uff1bsubset=[\"\u59d3\u540d\"]<\/code> \u628a\u5224\u65ad\u6807\u51c6\u6362\u6210\u300c\u53ea\u770b\u59d3\u540d\u300d\u2014\u2014\u5f20\u4e09\u51fa\u73b0\u4e24\u6b21\u53ea\u7559\u7b2c\u4e00\u6761\uff0808-01\uff09\u2014\u2014\u6ce8\u610f\u5f20\u4e09\u7684 08-02 \u8bb0\u5f55\u4e5f\u88ab\u5220\u4e86<\/strong>\uff0c\u56e0\u4e3a\u6309\u59d3\u540d\u5224\u65ad\u65f6\u300c\u5f20\u4e09\u300d\u5df2\u7ecf\u51fa\u73b0\u8fc7\u3002\u60f3\u4fdd\u7559 08-02 \u90a3\u6761\u5c31\u52a0 keep=\"last\"<\/code>\u3002<\/p>\n\n\n\n
    \u6613\u9519\u70b9<\/strong><\/p>\n\n\n\n
      \n
    • \u9519\u8bef\u5199\u6cd5\uff1adf.drop_duplicates()<\/code> \u540e\u539f\u8868\u8fd8\u6709\u91cd\u590d \u2192 \u95ee\u9898\uff1a\u8fd4\u56de\u65b0\u8868\uff0c\u4e0d\u6539\u539f\u8868 \u2192 \u6b63\u786e\u5199\u6cd5\uff1adf = df.drop_duplicates()<\/code>\u3002<\/li>\n\n\n\n
    • \u9519\u8bef\u5199\u6cd5\uff1a\u53bb\u91cd\u540e\u60f3\u4fdd\u7559\u300c\u6700\u540e\u4e00\u6761\u300d \u2192 \u95ee\u9898\uff1a\u9ed8\u8ba4 keep=\"first\"<\/code> \u4fdd\u7559\u7b2c\u4e00\u6761 \u2192 \u6b63\u786e\u5199\u6cd5\uff1adf.drop_duplicates(keep=\"last\")<\/code>\u3002<\/li>\n\n\n\n
    • \u9519\u8bef\u5199\u6cd5\uff1a\u60f3\u6309\u4e24\u5217\u5224\u65ad\u5374\u5199 subset=\"\u59d3\u540d\"<\/code> \u2192 \u95ee\u9898\uff1asubset \u53ea\u8ba4\u4e00\u5217 \u2192 \u6b63\u786e\u5199\u6cd5\uff1asubset=[\"\u59d3\u540d\", \"\u65e5\u671f\"]<\/code>\uff08\u5217\u8868\uff09\u3002<\/li>\n\n\n\n
    • \u9519\u8bef\u5199\u6cd5\uff1a\u4ee5\u4e3a\u53bb\u91cd\u4f1a\u53bb\u300c\u5217\u91cd\u590d\u300d \u2192 \u95ee\u9898\uff1adrop_duplicates \u53bb\u7684\u662f\u884c<\/strong>\u91cd\u590d \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u5217\u540d\u91cd\u590d\u662f\u53e6\u4e00\u56de\u4e8b\uff0c\u7528 df.T.drop_duplicates().T<\/code> \u6216\u5148\u770b df.columns<\/code>\u3002<\/li>\n<\/ul>\n\n\n\n
      \u8bb0\u5fc6\u53e3\u8bc0<\/strong>\uff1a\u4e00\u6a21\u4e00\u6837\u7b97\u91cd\u590d\uff0ckeep \u7559\u9996\u6216\u7559\u5c3e\uff1b\u53ea\u770b\u67d0\u51e0\u5217\uff0csubset \u6765\u6307\u5b9a\u3002<\/p>\n\n\n\n
      3 \u7c7b\u578b\u8f6c\u6362 \u2014\u2014 \u7ed9\u6570\u636e\u300c\u6362\u8eab\u4efd\u8bc1\u300d<\/h3>\n\n\n\n
      \u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f<\/strong><\/p>\n\n\n\n
      \u4ece\u6587\u4ef6\u8bfb\u8fdb\u6765\u7684\u6570\u636e\u7ecf\u5e38\u662f\u6587\u672c<\/strong>\uff1a\"12.5\"<\/code> \u662f\u5b57\u7b26\u4e32\u4e0d\u662f\u6570\u5b57\uff0c\"2026-08-01\"<\/code> \u662f\u5b57\u7b26\u4e32\u4e0d\u662f\u65e5\u671f\u3002\u6587\u672c\u4e0d\u80fd\u505a\u52a0\u51cf\u4e58\u9664\u3001\u4e0d\u80fd\u6309\u65e5\u671f\u6392\u5e8f\u7b5b\u9009\u2014\u2014\u7b97\u4e4b\u524d\u5fc5\u987b\u5148\u8f6c\u7c7b\u578b<\/strong>\u3002astype<\/code> \u8f6c\u6570\u503c\u3001pd.to_datetime<\/code> \u8f6c\u65e5\u671f\uff0c\u662f\u6e05\u6d17\u6d41\u7a0b\u7684\u300c\u4e2d\u8f6c\u7ad9\u300d\u3002<\/p>\n\n\n\n
      \u5e95\u5c42\u539f\u7406<\/strong><\/p>\n\n\n\n
      Pandas \u91cc\u6bcf\u4e2a\u5217\u90fd\u6709 dtype\uff08\u6570\u636e\u7c7b\u578b\uff09\uff1a\u6587\u672c\u5217\u5728 pandas 3.0 \u91cc\u9ed8\u8ba4\u662f str<\/code>\uff08\u65e7\u7248\u672c\u53eb object<\/code>\uff0c\u542b\u4e49\u4e00\u6837\uff09\uff0c\u6570\u503c\u5217\u662f int64<\/code> \/ float64<\/code>\uff0c\u65e5\u671f\u5217\u662f datetime64[us]<\/code>\uff08\u5fae\u79d2\u7cbe\u5ea6\uff0cpandas 3.0 \u9ed8\u8ba4\u5355\u4f4d\uff0c\u65e7\u7248\u662f ns<\/code> \u7eb3\u79d2\uff09\u3002**\u4e3a\u4ec0\u4e48\u6587\u672c \"10\"<\/code> \u52a0 \"9\"<\/code> \u662f \"109\"<\/code>**\uff1a\u5b57\u7b26\u4e32\u7684 +<\/code> \u662f\u62fc\u63a5\u4e0d\u662f\u76f8\u52a0\u2014\u2014\u7c7b\u578b\u9519\u4e86\uff0c\u8fd0\u7b97\u8bed\u4e49\u5168\u53d8\u3002astype(float)<\/code> \u628a\u6574\u5217\u8f6c\u6210\u6d6e\u70b9\uff1bpd.to_datetime(\u5217)<\/code> \u628a\u5b57\u7b26\u4e32\u89e3\u6790\u6210\u771f\u6b63\u7684\u65e5\u671f\u5bf9\u8c61\uff0c\u8f6c\u5b8c\u624d\u80fd\u7528 .dt.day<\/code> \u53d6\u300c\u65e5\u300d\u3001\u624d\u80fd\u6bd4\u8f83\u5927\u5c0f\u3002<\/p>\n\n\n\n
      \u751f\u6d3b\u7c7b\u6bd4<\/strong><\/p>\n\n\n\n
      \u50cf\u7ed9\u5916\u56fd\u4eba\u529e\u7b7e\u8bc1<\/strong>\uff1a\u62a4\u7167\uff08\u5b57\u7b26\u4e32\uff09\u4e0a\u5199\u7684\u300c12.5\u300d\u53ea\u662f\u6587\u5b57\uff0c\u53ea\u6709\u6362\u6210\u300c\u6570\u5b57\u7b7e\u8bc1\u300d\uff08float\uff09\u624d\u80fd\u505a\u7b97\u672f\uff1b\u300c2026-08-01\u300d\u53ea\u6709\u6362\u6210\u300c\u65e5\u671f\u7b7e\u8bc1\u300d\uff08datetime\uff09\u624d\u80fd\u6bd4\u5148\u540e\u3002\u7c7b\u578b\u5c31\u662f\u6570\u636e\u5728 Pandas \u4e16\u754c\u91cc\u7684\u300c\u8eab\u4efd\u300d\uff0c\u8eab\u4efd\u4e0d\u5bf9\uff0c\u4ec0\u4e48\u64cd\u4f5c\u90fd\u522b\u626d\u3002<\/p>\n\n\n\n
      \u6ce8\u610f\u4e8b\u9879\u603b\u7ed3<\/strong><\/p>\n\n\n\n
      1 \u6587\u672c\u8f6c\u6570\u5b57 \u2014\u2014 astype<\/strong>\uff1a<\/p>\n\n\n\n
      import pandas as pd\n\ndf = pd.DataFrame({\"\u4ef7\u683c\": [\"12.5\", \"9.9\", \"20.0\"]})\nprint(\"\u8f6c\u6362\u524d\u7c7b\u578b:\", df[\"\u4ef7\u683c\"].dtype)      # str \u662f\u6587\u672c\n\ndf[\"\u4ef7\u683c\"] = df[\"\u4ef7\u683c\"].astype(float)       # \u6587\u672c\u8f6c\u5c0f\u6570\nprint(\"\u8f6c\u6362\u540e\u7c7b\u578b:\", df[\"\u4ef7\u683c\"].dtype)\nprint(\"\u4ef7\u683c\u603b\u548c:\", df[\"\u4ef7\u683c\"].sum())        # \u8f6c\u5b8c\u624d\u80fd\u8ba1\u7b97\n<\/code><\/pre>\n\n\n\n
      \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
      \u8f6c\u6362\u524d\u7c7b\u578b: str\n\u8f6c\u6362\u540e\u7c7b\u578b: float64\n\u4ef7\u683c\u603b\u548c: 42.4\n<\/code><\/pre>\n\n\n\n
      astype(float)<\/code> \u6574\u5217\u8f6c\u6d6e\u70b9\uff0cdtype \u53d8 float64<\/code>\uff1b\u8f6c\u5b8c\u540e sum()<\/code> \u624d\u662f\u771f\u52a0\u6cd5\u2014\u2014\u4e0d\u8f6c\u7684\u8bdd \"12.5\" + \"9.9\"<\/code> \u4f1a\u53d8\u6210 \"12.59.9\"<\/code>\uff08\u5b57\u7b26\u4e32\u62fc\u63a5\uff09<\/strong>\u3002<\/p>\n\n\n\n
      2 \u6587\u672c\u8f6c\u65e5\u671f \u2014\u2014 pd.to_datetime<\/strong>\uff1a<\/p>\n\n\n\n
      df = pd.DataFrame({\"\u65e5\u671f\": [\"2026-08-01\", \"2026-08-02\", \"2026-08-03\"]})\ndf[\"\u65e5\u671f\"] = pd.to_datetime(df[\"\u65e5\u671f\"])\nprint(\"\u8f6c\u6362\u540e\u7c7b\u578b:\", df[\"\u65e5\u671f\"].dtype)\nprint(\"\u6bcf\u5929\u7684'\u65e5':\", list(df[\"\u65e5\u671f\"].dt.day))\nprint(\"\u665a\u4e8e 8 \u6708 1 \u65e5\u7684\u8bb0\u5f55:\")\nprint(df[df[\"\u65e5\u671f\"] > \"2026-08-01\"])\n<\/code><\/pre>\n\n\n\n
      \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
      \u8f6c\u6362\u540e\u7c7b\u578b: datetime64[us]\n\u6bcf\u5929\u7684'\u65e5': [1, 2, 3]\n\u665a\u4e8e 8 \u6708 1 \u65e5\u7684\u8bb0\u5f55:\n          \u65e5\u671f\n1 2026-08-02\n2 2026-08-03\n<\/code><\/pre>\n\n\n\n
      pd.to_datetime<\/code> \u628a\u5b57\u7b26\u4e32\u89e3\u6790\u6210\u65e5\u671f\u7c7b\u578b\uff1b.dt.day<\/code> \u662f\u65e5\u671f\u5217\u7684\u4e13\u5c5e\u8bbf\u95ee\u5668\uff0c\u53d6\u300c\u65e5\u300d\u5b57\u6bb5\uff1b\u8f6c\u5b8c\u540e\u65e5\u671f\u5b57\u7b26\u4e32\u4e5f\u80fd\u76f4\u63a5\u6bd4\u8f83\u5927\u5c0f\uff0c\u7b5b\u9009\u51fa\u665a\u4e8e 8 \u6708 1 \u65e5\u7684\u8bb0\u5f55\u3002<\/p>\n\n\n\n
      \u6613\u9519\u70b9<\/strong><\/p>\n\n\n\n
        \n
      • \u9519\u8bef\u5199\u6cd5\uff1adf[\"\u4ef7\u683c\"].astype(int)<\/code> \u8f6c \"12.5\"<\/code> \u2192 \u95ee\u9898\uff1a\u5e26\u5c0f\u6570\u7684\u6587\u672c\u8f6c int \u62a5 ValueError<\/code> \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u5148 astype(float)<\/code>\uff0c\u9700\u8981\u6574\u6570\u518d .astype(int)<\/code>\u3002<\/li>\n\n\n\n
      • \u9519\u8bef\u5199\u6cd5\uff1a\u6587\u672c\u5217\u91cc\u6df7\u7740 \"\u4e0d\u8be6\"<\/code> \u76f4\u63a5 astype \u2192 \u95ee\u9898\uff1a\u8f6c\u4e0d\u8fc7\u53bb\u7684\u503c\u62a5\u9519 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u5148 fillna<\/code> \u6216\u6e05\u6d17\u6389\u975e\u6570\u5b57\uff0c\u518d\u8f6c\u6362\uff1b\u6216 pd.to_numeric(\u5217, errors=\"coerce\")<\/code> \u628a\u8f6c\u4e0d\u4e86\u7684\u53d8 NaN\u3002<\/li>\n\n\n\n
      • \u9519\u8bef\u5199\u6cd5\uff1a\u8f6c\u65e5\u671f\u540e\u8bbf\u95ee .dt.day<\/code> \u62a5\u9519 \u2192 \u95ee\u9898\uff1a.dt<\/code> \u53ea\u80fd\u7528\u4e8e\u65e5\u671f\u7c7b\u578b\u5217 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u5148\u786e\u8ba4 pd.to_datetime<\/code> \u8f6c\u6210\u529f\u3001dtype \u662f datetime64<\/code>\u3002<\/li>\n\n\n\n
      • \u9519\u8bef\u5199\u6cd5\uff1a\u8f6c\u5b8c\u65e5\u671f\u76f4\u63a5 df[\"\u65e5\u671f\"]<\/code> \u53d6\u6570 \u2192 \u95ee\u9898\uff1a\u8f6c\u7c7b\u578b\u8fd4\u56de\u65b0\u5217\/\u65b0\u8868\uff0c\u9700\u8d4b\u503c \u2192 \u6b63\u786e\u5199\u6cd5\uff1adf[\"\u65e5\u671f\"] = pd.to_datetime(df[\"\u65e5\u671f\"])<\/code>\uff08\u91cd\u65b0\u8d4b\u503c\uff09\u3002<\/li>\n<\/ul>\n\n\n\n
        \u8bb0\u5fc6\u53e3\u8bc0<\/strong>\uff1a\u6587\u672c\u4e0d\u80fd\u7b97\uff0c\u65e5\u671f\u4e0d\u80fd\u6bd4\uff1bastype \u8f6c\u6570\u503c\uff0cto_datetime \u8f6c\u65e5\u671f\uff1b\u8f6c\u5b8c\u8bb0\u5f97\u91cd\u65b0\u8d4b\u503c\u3002<\/p>\n\n\n\n
        4 apply \u4e0e map \u2014\u2014 \u6279\u91cf\u52a0\u5de5\u6bcf\u4e2a\u503c<\/h3>\n\n\n\n
        \u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f<\/strong><\/p>\n\n\n\n
        \u6e05\u6d17\u7ecf\u5e38\u8981\u300c\u6309\u89c4\u5219\u6539\u4e00\u6279\u503c\u300d\uff1a\u53bb\u7a7a\u683c\u3001\u7edf\u4e00\u5927\u5c0f\u5199\u3001\u5355\u4f4d\u6362\u7b97\u3001\u7ed9\u6bcf\u4e2a\u683c\u5b50\u6253\u6807\u8bb0\u3002\u624b\u5199\u5faa\u73af\u592a\u5570\u55e6\uff0cPandas \u7684 apply<\/code>\uff08\u6309\u884c\/\u6309\u5217\u6279\u5904\u7406\uff09<\/strong> \u548c map<\/code>\uff08\u6309\u683c\u5b50\u6279\u5904\u7406\uff09<\/strong> \u4e00\u884c\u641e\u5b9a\u3002\u8fd9\u662f\u300c\u6e05\u6d17\u89c4\u5219\u300d\u843d\u5730\u7684\u5730\u65b9\u3002<\/p>\n\n\n\n
        \u5e95\u5c42\u539f\u7406<\/strong><\/p>\n\n\n\n
          \n
        • Series.apply(\u51fd\u6570)<\/code>\uff1a\u5bf9\u4e00\u5217\u7684\u6bcf\u4e2a\u5143\u7d20<\/strong>\u6267\u884c\u51fd\u6570\uff1bDataFrame.apply(\u51fd\u6570, axis=1)<\/code>\uff1a\u5bf9\u6bcf\u4e00\u884c<\/strong>\u6267\u884c\u51fd\u6570\uff08\u884c\u4f1a\u53d8\u6210\u4e00\u4e2a\u5e26\u5217\u540d\u7684 Series\uff0c\u51fd\u6570\u91cc\u7528 row[\"\u5217\u540d\"]<\/code> \u53d6\u503c\uff09<\/li>\n\n\n\n
        • df.map(\u51fd\u6570)<\/code>\uff1a\u5bf9\u6bcf\u4e2a\u683c\u5b50<\/strong>\u6267\u884c\u51fd\u6570\uff08pandas 2.1 \u8d77\u63a8\u8350\uff0c\u65e7\u7248\u53eb applymap<\/code>\uff0c\u65b0\u7248\u5df2\u79fb\u9664\uff09<\/li>\n\n\n\n
        • \u51fd\u6570\u53ef\u4ee5\u662f lambda<\/code> \u533f\u540d\u51fd\u6570\uff0c\u4e5f\u53ef\u4ee5\u662f\u6709\u540d\u5b57\u7684\u51fd\u6570\u2014\u2014\u89c4\u5219\u590d\u6742\u5c31\u5b9a\u4e49\u51fd\u6570\uff0c\u89c4\u5219\u7b80\u5355\u5c31\u7528 lambda<\/strong><\/li>\n<\/ul>\n\n\n\n
          \u751f\u6d3b\u7c7b\u6bd4<\/strong><\/p>\n\n\n\n
          \u50cf\u6d41\u6c34\u7ebf\u4e0a\u7684\u8d28\u68c0\u5458<\/strong>\uff1aapply<\/code> \u662f\u300c\u6309\u5217\uff08\u6216\u6309\u884c\uff09\u6328\u4e2a\u8fc7\u4e00\u904d\uff0c\u6309\u4f60\u7684\u89c4\u5219\u5904\u7406\u300d\uff1bmap<\/code> \u662f\u300c\u6bcf\u4e2a\u96f6\u4ef6\uff08\u683c\u5b50\uff09\u90fd\u8d34\u4e00\u4e2a\u6807\u7b7e\u300d\u3002\u4f60\u53ea\u8d1f\u8d23\u5199\u300c\u89c4\u5219\u300d\uff0cPandas \u8d1f\u8d23\u300c\u6328\u4e2a\u6267\u884c\u300d\u3002<\/p>\n\n\n\n
          \u6ce8\u610f\u4e8b\u9879\u603b\u7ed3<\/strong><\/p>\n\n\n\n
          1 apply \u6309\u884c\u7b97\u5e73\u5747 + map \u6309\u683c\u5b50\u653e\u5927<\/strong>\uff1a<\/p>\n\n\n\n
          import pandas as pd\n\ndf = pd.DataFrame({\n    \"\u8bed\u6587\": [88, 75, 90],\n    \"\u6570\u5b66\": [92, 80, 88],\n})\ndf[\"\u5e73\u5747\u5206\"] = df.apply(\n    lambda row: (row[\"\u8bed\u6587\"] + row[\"\u6570\u5b66\"]) \/ 2, axis=1\n)\nprint(df)\nprint(\"---\")\ndf2 = pd.DataFrame({\"a\": [1, 2], \"b\": [3, 4]})\nprint(df2.map(lambda x: x * 10))\n<\/code><\/pre>\n\n\n\n
          \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
             \u8bed\u6587  \u6570\u5b66   \u5e73\u5747\u5206\n0  88  92  90.0\n1  75  80  77.5\n2  90  88  89.0\n---\n    a   b\n0  10  30\n1  20  40\n<\/code><\/pre>\n\n\n\n
          df.apply(\u51fd\u6570, axis=1)<\/code> \u91cc axis=1<\/code> \u8868\u793a\u6309\u884c<\/strong>\u5904\u7406\u2014\u2014\u6bcf\u4e00\u884c\u662f\u4e00\u4e2a\u300c\u7ad6\u7740\u7684 Series\u300d\uff0crow[\"\u8bed\u6587\"]<\/code> \u53d6\u8be5\u884c\u7684\u8bed\u6587\u5206\uff1bdf2.map(lambda x: x * 10)<\/code> \u5bf9\u6bcf\u4e2a\u683c\u5b50<\/strong>\u4e58 10\u2014\u2014\u300c\u6309\u683c\u5b50\u300d\u548c\u300c\u6309\u884c\/\u5217\u300d\u662f\u4e24\u79cd\u7c92\u5ea6\uff0c\u522b\u6df7\u3002<\/p>\n\n\n\n
          2 \u6587\u672c\u6e05\u6d17 \u2014\u2014 apply + \u5b57\u7b26\u4e32\u65b9\u6cd5<\/strong>\uff1a<\/p>\n\n\n\n
          df = pd.DataFrame({\"\u57ce\u5e02\": [\" \u5317\u4eac \", \"shanghai\", \"\u5e7f\u5dde\"]})\ndf[\"\u57ce\u5e02\"] = df[\"\u57ce\u5e02\"].apply(lambda s: s.strip().title())\nprint(df)\n<\/code><\/pre>\n\n\n\n
          \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
                   \u57ce\u5e02\n0        \u5317\u4eac\n1  Shanghai\n2        \u5e7f\u5dde\n<\/code><\/pre>\n\n\n\n
          .strip()<\/code> \u53bb\u6389\u9996\u5c3e\u7a7a\u683c\uff0c.title()<\/code> \u9996\u5b57\u6bcd\u5927\u5199\uff1bapply<\/code> \u628a\u8fd9\u4e24\u4e2a\u65b9\u6cd5\u5957\u5230\u8fd9\u4e00\u5217\u7684\u6bcf\u4e2a\u5143\u7d20\u4e0a<\/strong>\u2014\u2014\u8fd9\u662f\u6e05\u6d17\u6587\u672c\u5217\u7684\u6807\u51c6\u59ff\u52bf\u3002\u4e2d\u6587\u4e0d\u53d7 title \u5f71\u54cd\uff0c\u6240\u4ee5\u5317\u4eac\u3001\u5e7f\u5dde\u539f\u6837\u4fdd\u7559\u3002<\/p>\n\n\n\n
          \u6613\u9519\u70b9<\/strong><\/p>\n\n\n\n
            \n
          • \u9519\u8bef\u5199\u6cd5\uff1adf.apply(\u51fd\u6570)<\/code> \u5fd8\u5199 axis=1<\/code> \u2192 \u95ee\u9898\uff1a\u9ed8\u8ba4 axis=0 \u6309\u5217\u5904\u7406\uff0c\u51fd\u6570\u91cc row[\"\u8bed\u6587\"]<\/code> \u76f4\u63a5\u62a5\u9519 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u6309\u884c\u5904\u7406\u5fc5\u987b axis=1<\/code>\u3002<\/li>\n\n\n\n
          • \u9519\u8bef\u5199\u6cd5\uff1a\u7528 df.applymap(...)<\/code> \u2192 \u95ee\u9898\uff1apandas \u65b0\u7248\u5df2\u79fb\u9664 applymap \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u7528 df.map(...)<\/code>\uff082.1+\uff09\u3002<\/li>\n\n\n\n
          • \u9519\u8bef\u5199\u6cd5\uff1alambda \u91cc\u5199\u591a\u884c\u903b\u8f91 \u2192 \u95ee\u9898\uff1alambda \u53ea\u80fd\u5199\u4e00\u884c\u8868\u8fbe\u5f0f \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u903b\u8f91\u590d\u6742\u5c31 def \u51fd\u6570(): ...<\/code> \u518d apply(\u51fd\u6570)<\/code>\u3002<\/li>\n\n\n\n
          • \u9519\u8bef\u5199\u6cd5\uff1aapply<\/code> \u91cc\u76f4\u63a5 return<\/code> \u4e2d\u6587\u6807\u70b9\u6216 None \u2192 \u95ee\u9898\uff1a\u8fd4\u56de\u503c\u4f1a\u539f\u6837\u5199\u8fdb\u5217\uff0c\u53ef\u80fd\u5f15\u5165\u65b0\u810f\u6570\u636e \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u6e05\u6d17\u89c4\u5219\u91cc\u660e\u786e\u7a7a\u503c\u8fd4\u56de\u4ec0\u4e48\uff08\u5982 \"\u672a\u77e5\"<\/code>\uff09\u3002<\/li>\n<\/ul>\n\n\n\n
            \u8bb0\u5fc6\u53e3\u8bc0<\/strong>\uff1aapply \u7ba1\u884c\u5217\uff0cmap \u7ba1\u683c\u5b50\uff1b\u4e00\u884c lambda\uff0c\u590d\u6742\u5c31 def\uff1baxis=1 \u6309\u884c\u8d70\u3002<\/p>\n\n\n\n
            5 \u5f02\u5e38\u503c \u2014\u2014 \u6570\u636e\u91cc\u7684\u300c\u79bb\u8c31\u503c\u300d<\/h3>\n\n\n\n
            \u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f<\/strong><\/p>\n\n\n\n
            \u5f02\u5e38\u503c<\/strong>\u662f\u660e\u663e\u4e0d\u5408\u7406\u7684\u6570\u503c\uff1a\u5206\u6570 150\u3001\u5e74\u9f84 -5\u3001\u5de5\u8d44 99999\u3002\u5b83\u4eec\u53ef\u80fd\u662f\u5f55\u5165\u9519\u8bef\u3001\u5355\u4f4d\u9519\u8bef\uff0c\u4e5f\u53ef\u80fd\u662f\u771f\u5b9e\u4f46\u6781\u7aef<\/strong>\u7684\u6570\u636e\uff08\u5982\u8001\u677f\u7684\u5de5\u8d44\uff09\u3002\u5f02\u5e38\u503c\u4f1a\u628a\u5e73\u5747\u503c\u62c9\u504f\u3001\u8ba9\u56fe\u8868\u5931\u771f\u2014\u2014\u53d1\u73b0\u5b83\u3001\u5224\u65ad\u5b83\u3001\u5904\u7406\u5b83<\/strong>\u662f\u6e05\u6d17\u7684\u6700\u540e\u4e00\u5173\u3002<\/p>\n\n\n\n
            \u5e95\u5c42\u539f\u7406<\/strong><\/p>\n\n\n\n
            \u53d1\u73b0\u5f02\u5e38\u9760\u4e24\u62db\uff1a**describe()<\/code> \u770b\u7edf\u8ba1\u6458\u8981\uff08max=150<\/code>\u3001min=-5<\/code> \u4e00\u773c\u53ef\u7591\uff09\uff1b<\/strong>\u5e03\u5c14\u7b5b\u9009\u63ea\u51fa\u6765\uff08(df[\"\u5206\u6570\"] > 100) | (df[\"\u5206\u6570\"] < 0)<\/code> \u76f4\u63a5\u5217\u51fa\u5f02\u5e38\u8bb0\u5f55\uff09\u3002\u5904\u7406\u4e5f\u5206\u4e24\u62db\uff1a<\/strong>\u5220\u9664\uff08\u786e\u8ba4\u662f\u5f55\u5165\u9519\u8bef\uff0c\u7528 df[~mask]<\/code> \u53d6\u300c\u4e0d\u662f\u5f02\u5e38\u300d\u7684\u884c\uff09\uff1b<\/strong>\u66ff\u6362\uff08\u7528\u4e2d\u4f4d\u6570\u7b49\u5408\u7406\u503c\u9876\u66ff\uff0cdf.loc[mask, \"\u5206\u6570\"] = \u4e2d\u4f4d\u6570<\/code>\uff09\u3002\u5224\u65ad\u300c\u5408\u4e0d\u5408\u7406\u300d\u7684\u4f9d\u636e\u662f<\/strong>\u4e1a\u52a1\u89c4\u5219**\uff1a\u5206\u6570 0100\u3001\u5e74\u9f84 0<\/del>120\u2014\u2014\u5148\u5b9a\u89c4\u5219\uff0c\u518d\u5199\u6761\u4ef6\u3002<\/p>\n\n\n\n
            \u751f\u6d3b\u7c7b\u6bd4<\/strong><\/p>\n\n\n\n
            \u50cf\u4f53\u68c0\u62a5\u544a\u91cc\u7684\u300c\u590d\u67e5\u9879\u300d<\/strong>\uff1a\u8840\u538b 300 \u660e\u663e\u662f\u673a\u5668\u6545\u969c\uff08\u5f02\u5e38\u503c\uff09\uff0c\u533b\u751f\u5148\u6807\u51fa\u6765\uff08\u53d1\u73b0\uff09\uff0c\u518d\u5224\u65ad\u662f\u91cd\u6d4b\uff08\u5220\u9664\uff09\u8fd8\u662f\u6309\u53c2\u8003\u503c\u4fee\u6b63\uff08\u66ff\u6362\uff09\u2014\u2014\u7edd\u4e0d\u80fd\u76f4\u63a5\u62ff\u5f02\u5e38\u503c\u53bb\u7b97\u5e73\u5747\u3002<\/p>\n\n\n\n
            \u6ce8\u610f\u4e8b\u9879\u603b\u7ed3<\/strong><\/p>\n\n\n\n
            1 \u53d1\u73b0\u5f02\u5e38 \u2014\u2014 describe \u6458\u8981 + \u5e03\u5c14\u7b5b\u9009<\/strong>\uff1a<\/p>\n\n\n\n
            import pandas as pd\n\ndf = pd.DataFrame({\n    \"\u59d3\u540d\": [\"\u5f20\u4e09\", \"\u674e\u56db\", \"\u738b\u4e94\", \"\u8d75\u516d\"],\n    \"\u5206\u6570\": [88, 150, 95, -5],\n})\nprint(\"=== \u7edf\u8ba1\u6458\u8981\uff08max\/min \u53ef\u7591\uff09===\")\nprint(df.describe())\n\nmask = (df[\"\u5206\u6570\"] > 100) | (df[\"\u5206\u6570\"] < 0)\nprint(\"\\n=== \u5f02\u5e38\u8bb0\u5f55 ===\")\nprint(df[mask])\n<\/code><\/pre>\n\n\n\n
            \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
            === \u7edf\u8ba1\u6458\u8981\uff08max\/min \u53ef\u7591\uff09===\n              \u5206\u6570\ncount    4.00000\nmean    82.00000\nstd     64.28582\nmin     -5.00000\n25%     64.75000\n50%     91.50000\n75%    108.75000\nmax    150.00000\n\n=== \u5f02\u5e38\u8bb0\u5f55 ===\n   \u59d3\u540d   \u5206\u6570\n1  \u674e\u56db  150\n3  \u8d75\u516d   -5\n<\/code><\/pre>\n\n\n\n
            describe()<\/code> \u91cc max=150<\/code>\u3001min=-5<\/code> \u4e00\u773c\u53ef\u7591\uff1bmask<\/code> \u7528\u5e03\u5c14\u7b5b\u9009\u628a\u4e24\u6761\u5f02\u5e38\u8bb0\u5f55\u5355\u72ec\u62ce\u51fa\u6765<\/strong>\u2014\u2014\u5148\u770b\u6e05\u662f\u8c01\u3001\u4ec0\u4e48\u503c\uff0c\u518d\u51b3\u5b9a\u600e\u4e48\u5904\u7406\u3002<\/p>\n\n\n\n
            2 \u5904\u7406\u5f02\u5e38 \u2014\u2014 \u5220\u9664 or \u66ff\u6362\u6210\u4e2d\u4f4d\u6570<\/strong>\uff1a<\/p>\n\n\n\n
            mask = (df[\"\u5206\u6570\"] > 100) | (df[\"\u5206\u6570\"] < 0)\n\n# \u65b9\u5f0f\u4e00\uff1a\u5220\u9664\u5f02\u5e38\u884c\uff08~ \u53d6\u53cd = \u4fdd\u7559\"\u4e0d\u662f\u5f02\u5e38\"\u7684\u884c\uff09\nclean = df[~mask]\nprint(\"=== \u5220\u9664\u5f02\u5e38\u540e ===\")\nprint(clean)\n\n# \u65b9\u5f0f\u4e8c\uff1a\u66ff\u6362\u6210\u4e2d\u4f4d\u6570\ndf[\"\u5206\u6570\"] = df[\"\u5206\u6570\"].astype(float)\ndf.loc[mask, \"\u5206\u6570\"] = df[\"\u5206\u6570\"].median()\nprint(\"\\n=== \u66ff\u6362\u6210\u4e2d\u4f4d\u6570\u540e ===\")\nprint(df)\n<\/code><\/pre>\n\n\n\n
            \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
            === \u5220\u9664\u5f02\u5e38\u540e ===\n   \u59d3\u540d  \u5206\u6570\n0  \u5f20\u4e09  88\n2  \u738b\u4e94  95\n\n=== \u66ff\u6362\u6210\u4e2d\u4f4d\u6570\u540e ===\n   \u59d3\u540d    \u5206\u6570\n0  \u5f20\u4e09  88.0\n1  \u674e\u56db  91.5\n2  \u738b\u4e94  95.0\n3  \u8d75\u516d  91.5\n<\/code><\/pre>\n\n\n\n
            df[~mask]<\/code> \u91cc ~<\/code> \u662f\u300c\u53d6\u53cd\u300d\u2014\u2014mask \u662f True \u7684\u662f\u5f02\u5e38\uff0c\u53d6\u53cd\u540e\u53ea\u7559\u6b63\u5e38\u884c\uff1b\u66ff\u6362\u65b9\u6848\uff1a\u5148 astype(float)<\/code>\uff0c\u518d df.loc[mask, \"\u5206\u6570\"] = df[\"\u5206\u6570\"].median()<\/code>\u2014\u2014**loc[mask, \u5217]<\/code> \u7cbe\u786e\u5b9a\u4f4d\u5230\u300c\u5f02\u5e38\u7684\u683c\u5b50\u300d\u586b\u4e2d\u4f4d\u6570**\u3002\u5220\u9664\u7b80\u5355\u76f4\u63a5\uff0c\u66ff\u6362\u4fdd\u7559\u6570\u636e\u91cf\uff0c\u770b\u4e1a\u52a1\u573a\u666f\u9009\u3002<\/p>\n\n\n\n
            \u6613\u9519\u70b9<\/strong><\/p>\n\n\n\n
              \n
            • \u9519\u8bef\u5199\u6cd5\uff1a\u53ea\u9760\u773c\u775b\u770b describe \u4e0d\u5199\u7b5b\u9009\u6761\u4ef6 \u2192 \u95ee\u9898\uff1a\u6570\u636e\u591a\u65f6\u770b\u4e0d\u51fa\u6765 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u7528\u5e03\u5c14\u6761\u4ef6\u628a\u5f02\u5e38\u7b5b\u51fa\u6765\u6253\u5370<\/strong>\u3002<\/li>\n\n\n\n
            • \u9519\u8bef\u5199\u6cd5\uff1a\u7528 and<\/code> \/ or<\/code> \u8fde\u5f02\u5e38\u6761\u4ef6 \u2192 \u95ee\u9898\uff1a\u6570\u7ec4\/Series \u4e0d\u652f\u6301\uff0c\u62a5\u9519 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u7528 &<\/code> \/ |<\/code>\uff0c\u6bcf\u4e2a\u6761\u4ef6\u52a0\u62ec\u53f7\u3002<\/li>\n\n\n\n
            • \u9519\u8bef\u5199\u6cd5\uff1a\u66ff\u6362\u65f6\u5199 df[\"\u5206\u6570\"][mask] = ...<\/code> \u2192 \u95ee\u9898\uff1a\u53ef\u80fd\u89e6\u53d1 SettingWithCopyWarning\uff0c\u6539\u4e0d\u5230\u539f\u8868 \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u7528 df.loc[mask, \"\u5206\u6570\"] = ...<\/code>\u3002<\/li>\n\n\n\n
            • \u9519\u8bef\u5199\u6cd5\uff1a\u4e0d\u5148\u5b9a\u300c\u5408\u7406\u8303\u56f4\u300d\u5c31\u5904\u7406 \u2192 \u95ee\u9898\uff1a\u5220\u9519\/\u6539\u9519\u6b63\u5e38\u6570\u636e \u2192 \u6b63\u786e\u5199\u6cd5\uff1a\u5148\u6309\u4e1a\u52a1\u5b9a\u89c4\u5219\uff08\u5206\u6570 0~100\uff09\uff0c\u518d\u5199\u6761\u4ef6\u3002<\/li>\n<\/ul>\n\n\n\n
              \u8bb0\u5fc6\u53e3\u8bc0<\/strong>\uff1adescribe \u770b\u6458\u8981\uff0c\u5e03\u5c14\u7b5b\u5f02\u5e38\uff1b\u5220\u9664\u7528\u53d6\u53cd\uff0c\u66ff\u6362\u7528 loc\uff1b\u8303\u56f4\u5148\u5b9a\u597d\uff0c\u518d\u52a8\u624b\u5904\u7406\u3002<\/p>\n\n\n\n
              6 \u52a8\u624b\u5b9e\u8df5\uff1a\u6e05\u6d17\u810f\u6570\u636e\u811a\u672c<\/h3>\n\n\n\n
              \u628a\u4eca\u5929\u5b66\u7684\u4e94\u62db\u7ec4\u5408\u6210\u4e00\u4e2a\u5b8c\u6574\u7684\u6e05\u6d17\u811a\u672c\uff0c\u5904\u7406\u4e00\u4efd\u300c\u96c6\u9f50\u4e86\u5404\u79cd\u810f\u6570\u636e\u300d\u7684 CSV\uff1a\u7f3a\u5931\u503c\u3001\u91cd\u590d\u884c\u3001\u5e26\u7a7a\u683c\u548c\u5927\u5c0f\u5199\u6df7\u4e71\u7684\u6587\u672c\u3001\u6587\u672c\u6570\u5b57\u3001\u5f02\u5e38\u503c\u3002<\/p>\n\n\n\n
              # clean_data.py \u2014\u2014 \u6e05\u6d17\u810f\u6570\u636e\u811a\u672c\nimport pandas as pd\n\n# 1. \u51c6\u5907\u4e00\u4efd\u810f\u6570\u636e\uff08\u6a21\u62df\u4ece\u5916\u90e8\u62ff\u5230\u7684\u6587\u4ef6\uff09\ncsv_text = \"\"\"\u59d3\u540d,\u5e74\u9f84,\u57ce\u5e02,\u6536\u5165\n\u5f20\u4e09,28,\u5317\u4eac,12000\n\u674e\u56db,,\u4e0a\u6d77,9000\n\u674e\u56db,,\u4e0a\u6d77,9000\n\u738b\u4e94,25, \u5e7f\u5dde ,8000\n\u8d75\u516d,150,\u5317\u4eac,99999\n\u5b59\u4e03,-5,\u6df1\u5733,6500\n\u5468\u516b,35,beijing,\n\"\"\"\nwith open(\"dirty.csv\", \"w\", encoding=\"utf-8\") as f:\n    f.write(csv_text)\n\ndf = pd.read_csv(\"dirty.csv\")\nprint(\"=== \u539f\u59cb\u6570\u636e ===\")\nprint(df)\nprint(\"\u5f62\u72b6:\", df.shape)\n\n# 2. \u53bb\u91cd\ndf = df.drop_duplicates()\nprint(\"\\n=== \u53bb\u91cd\u540e ===\")\nprint(df)\n\n# 3. \u6e05\u6d17\u6587\u672c\u5217\uff1a\u53bb\u7a7a\u683c + \u7edf\u4e00\u9996\u5b57\u6bcd\u5927\u5199\ndf[\"\u57ce\u5e02\"] = df[\"\u57ce\u5e02\"].apply(lambda s: s.strip().title())\n\n# 4. \u586b\u5145\u7f3a\u5931\u503c\uff1a\u5e74\u9f84\u586b\u4e2d\u4f4d\u6570\uff0c\u6536\u5165\u586b\u5e73\u5747\u503c\ndf[\"\u5e74\u9f84\"] = df[\"\u5e74\u9f84\"].fillna(df[\"\u5e74\u9f84\"].median())\ndf[\"\u6536\u5165\"] = df[\"\u6536\u5165\"].fillna(df[\"\u6536\u5165\"].mean())\n\n# 5. \u7c7b\u578b\u8f6c\u6362\uff1a\u6536\u5165\u662f\u6587\u672c\uff0c\u8f6c\u6210\u6d6e\u70b9\u6570\ndf[\"\u6536\u5165\"] = df[\"\u6536\u5165\"].astype(float)\n\n# 6. \u5904\u7406\u5f02\u5e38\u503c\uff1a\u5e74\u9f84\u5fc5\u987b 0~120\uff0c\u6536\u5165\u4e0d\u8d85\u8fc7 50000\nmask = (df[\"\u5e74\u9f84\"] > 120) | (df[\"\u5e74\u9f84\"] < 0) | (df[\"\u6536\u5165\"] > 50000)\nprint(\"\\n=== \u53d1\u73b0\u7684\u5f02\u5e38\u8bb0\u5f55 ===\")\nprint(df[mask])\ndf = df[~mask]          # \u5220\u9664\u5f02\u5e38\u884c\n\nprint(\"\\n=== \u6e05\u6d17\u5b8c\u6210 ===\")\nprint(df)\nprint(\"\u5f62\u72b6:\", df.shape)\nprint(\"\\n=== \u6e05\u6d17\u540e\u7edf\u8ba1 ===\")\nprint(df.describe())\n<\/code><\/pre>\n\n\n\n
              \u8fd0\u884c\u7ed3\u679c\uff1a<\/p>\n\n\n\n
              === \u539f\u59cb\u6570\u636e ===\n   \u59d3\u540d     \u5e74\u9f84       \u57ce\u5e02       \u6536\u5165\n0  \u5f20\u4e09   28.0       \u5317\u4eac  12000.0\n1  \u674e\u56db    NaN       \u4e0a\u6d77   9000.0\n2  \u674e\u56db    NaN       \u4e0a\u6d77   9000.0\n3  \u738b\u4e94   25.0      \u5e7f\u5dde    8000.0\n4  \u8d75\u516d  150.0       \u5317\u4eac  99999.0\n5  \u5b59\u4e03   -5.0       \u6df1\u5733   6500.0\n6  \u5468\u516b   35.0  beijing      NaN\n\n\u5f62\u72b6: (7, 4)\n\n=== \u53bb\u91cd\u540e ===\n   \u59d3\u540d     \u5e74\u9f84       \u57ce\u5e02       \u6536\u5165\n0  \u5f20\u4e09   28.0       \u5317\u4eac  12000.0\n1  \u674e\u56db    NaN       \u4e0a\u6d77   9000.0\n3  \u738b\u4e94   25.0      \u5e7f\u5dde    8000.0\n4  \u8d75\u516d  150.0       \u5317\u4eac  99999.0\n5  \u5b59\u4e03   -5.0       \u6df1\u5733   6500.0\n6  \u5468\u516b   35.0  beijing      NaN\n\n=== \u53d1\u73b0\u7684\u5f02\u5e38\u8bb0\u5f55 ===\n   \u59d3\u540d     \u5e74\u9f84  \u57ce\u5e02       \u6536\u5165\n4  \u8d75\u516d  150.0  \u5317\u4eac  99999.0\n5  \u5b59\u4e03   -5.0  \u6df1\u5733   6500.0\n\n=== \u6e05\u6d17\u5b8c\u6210 ===\n   \u59d3\u540d    \u5e74\u9f84       \u57ce\u5e02       \u6536\u5165\n0  \u5f20\u4e09  28.0       \u5317\u4eac  12000.0\n1  \u674e\u56db  28.0       \u4e0a\u6d77   9000.0\n3  \u738b\u4e94  25.0       \u5e7f\u5dde   8000.0\n6  \u5468\u516b  35.0  Beijing  27099.8\n\n\u5f62\u72b6: (4, 4)\n\n=== \u6e05\u6d17\u540e\u7edf\u8ba1 ===\n              \u5e74\u9f84            \u6536\u5165\ncount   4.000000      4.000000\nmean   29.000000  14024.950000\nstd     4.242641   8880.733266\nmin    25.000000   8000.000000\n25%    27.250000   8750.000000\n50%    28.000000  10500.000000\n75%    29.750000  15774.950000\nmax    35.000000  27099.800000\n<\/code><\/pre>\n\n\n\n
              \u6e05\u6d17\u8fc7\u7a0b\u62c6\u89e3\uff1a\u674e\u56db\u91cd\u590d\u884c\u88ab\u53bb\u91cd\uff1b\u57ce\u5e02\u7a7a\u683c\u53bb\u6389\u3001beijing<\/code> \u7edf\u4e00\u6210 Beijing<\/code>\uff1b\u674e\u56db\u7f3a\u5931\u5e74\u9f84\u7528\u4e2d\u4f4d\u6570 28 \u8865\u4e0a\uff0c\u5468\u516b\u7f3a\u5931\u6536\u5165\u7528\u5e73\u5747\u503c 27099.8 \u8865\u4e0a\uff1b\u6536\u5165\u4ece\u6587\u672c\u8f6c\u6570\u5b57\uff1b\u5e74\u9f84 150 \u548c -5 \u4e24\u6761\u5f02\u5e38\u88ab\u5220\u9664\u30027 \u884c\u810f\u6570\u636e\u53d8 4 \u884c\u5e72\u51c0\u6570\u636e\u3002<\/p>\n\n\n\n
              \u5347\u7ea7\u6311\u6218<\/strong>\uff1a\u628a\u6e05\u6d17\u597d\u7684\u6570\u636e\u4fdd\u5b58\u56de CSV\uff08df.to_csv(\"clean.csv\", index=False, encoding=\"utf-8-sig\")<\/code>\uff09\uff1b\u518d\u8bd5\u8bd5\u628a\u5f02\u5e38\u503c\u6539\u6210\u300c\u66ff\u6362\u300d\u800c\u4e0d\u662f\u300c\u5220\u9664\u300d\uff0c\u5bf9\u6bd4\u4e24\u79cd\u7ed3\u679c\u3002<\/p>\n\n\n\n
              \u4eca\u65e5\u603b\u7ed3<\/strong><\/p>\n\n\n\n
              \u80fd\u4e00\u773c\u770b\u51fa\u6570\u636e\u300c\u810f\u300d\u5728\u54ea\uff1a\u7f3a\u5931\u3001\u91cd\u590d\u3001\u7c7b\u578b\u9519\u3001\u5f02\u5e38\u503c\uff0c\u56db\u79cd\u810f\u6570\u636e\u5404\u6709\u5404\u7684\u4fee\u6cd5\nisnull<\/code> \u53d1\u73b0\u7f3a\u5931\uff0cdropna<\/code> \u5220\u9664\u3001fillna<\/code> \u586b\u5145\uff08\u6570\u503c\u5217\u586b\u5e73\u5747\u503c\/\u4e2d\u4f4d\u6570\uff0c\u6587\u672c\u5217\u586b\u56fa\u5b9a\u503c\uff0c\u7f3a\u5931\u591a\u5c31\u5220\uff09\ndrop_duplicates<\/code> \u53bb\u91cd\uff0csubset<\/code> \u63a7\u5236\u6309\u54ea\u51e0\u5217\u5224\u65ad\u91cd\u590d\uff0ckeep<\/code> \u7559\u9996\u6216\u7559\u5c3e\nastype<\/code> \u8f6c\u6570\u503c\u3001pd.to_datetime<\/code> \u8f6c\u65e5\u671f\uff1b\u6587\u672c\u4e0d\u80fd\u7b97\u3001\u65e5\u671f\u4e0d\u80fd\u6bd4\uff0c\u8f6c\u5b8c\u8bb0\u5f97\u91cd\u65b0\u8d4b\u503c\napply<\/code> \u6309\u884c\/\u5217\u5904\u7406\uff08axis=1<\/code> \u6309\u884c\uff09\uff0cmap<\/code> \u6309\u683c\u5b50\u5904\u7406\uff1b\u4e00\u884c lambda\uff0c\u590d\u6742\u5c31 def\ndescribe<\/code> \u770b\u6458\u8981\u53d1\u73b0\u5f02\u5e38\u503c\uff0c\u5e03\u5c14\u7b5b\u9009\u63ea\u51fa\u6765\uff1b\u5220\u9664\u7528 df[~mask]<\/code>\uff0c\u66ff\u6362\u7528 df.loc[mask, \u5217] = \u503c<\/code>\n\u6e05\u6d17\u810f\u6570\u636e\u811a\u672c\u8fd0\u884c\u6b63\u5e38\uff1a7 \u884c\u810f\u6570\u636e\u53d8 4 \u884c\u5e72\u51c0\u6570\u636e<\/p>\n\n\n\n
              \u62a5\u9519<\/th>\u539f\u56e0<\/th>\u4fee\u590d<\/th><\/tr><\/thead>
              ValueError: invalid literal for int() with base 10: '12.5'<\/code><\/td>astype(int)<\/code> \u8f6c\u4e0d\u4e86\u5e26\u5c0f\u6570\u70b9\u7684\u6587\u672c<\/td>\u5148\u7528 astype(float)<\/code>\uff0c\u518d\u51b3\u5b9a\u8981\u4e0d\u8981\u53d6\u6574<\/td><\/tr>
              KeyError: '\u57ce\u5e02'<\/code><\/td>\u5217\u540d\u5199\u9519\uff08\u6bd4\u5982\u591a\u4e86\u7a7a\u683c\uff09<\/td>print(df.columns)<\/code> \u67e5\u770b\u5b9e\u9645\u5217\u540d\uff0c\u5fc5\u8981\u65f6 df.columns = [c.strip() for c in df.columns]<\/code><\/td><\/tr>
              AttributeError: Can only use .dt accessor with datetimelike values<\/code><\/td>dt.day<\/code> \u7528\u5728\u6ca1\u8f6c\u6210\u65e5\u671f\u7684\u5217\u4e0a<\/td>\u5148 pd.to_datetime<\/code> \u8f6c\u6362\u518d\u8bbf\u95ee .dt<\/code><\/td><\/tr>
              TypeError: unsupported operand type(s) for +: 'str' and 'float'<\/code><\/td>\u6587\u672c\u5217\u6ca1\u8f6c\u7c7b\u578b\u5c31\u505a\u8fd0\u7b97<\/td>\u5148 astype(float)<\/code> \u6216 to_numeric<\/code><\/td><\/tr>
              AttributeError: 'DataFrame' object has no attribute 'applymap'<\/code><\/td>\u65b0\u7248 pandas \u79fb\u9664\u4e86 applymap<\/code><\/td>\u6539\u7528 df.map()<\/code>\uff08\u65e7\u7248\u672c\u4e24\u4e2a\u90fd\u652f\u6301\uff09<\/td><\/tr>
              SettingWithCopyWarning<\/code><\/td>\u5728\u7b5b\u9009\u51fa\u6765\u7684\u526f\u672c\u4e0a\u505a\u4fee\u6539<\/td>\u7528 .loc<\/code> \u4fee\u6539\uff0c\u6216\u5148 df = df.copy()<\/code><\/td><\/tr><\/tbody><\/table><\/figure>\n","protected":false},"excerpt":{"rendered":"
              1 \u7f3a\u5931\u503c \u2014\u2014 \u6570\u636e\u91cc\u7684\u300c\u7a7a\u4f4d\u300d \u662f\u4ec0\u4e48\uff1f\u4e3a\u4ec0\u4e48\u5b66\uff1f \u7f3a\u5931\u503c\u5c31\u662f\u8868\u683c\u91cc\u300c\u6ca1\u586b\u300d\u7684\u683c\u5b50\uff0cPandas \u91cc\u663e\u793a\u4e3a […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_import_markdown_pro_load_document_selector":0,"_import_markdown_pro_submit_text_textarea":"","footnotes":""},"categories":[16,9],"tags":[17],"class_list":["post-115","post","type-post","status-publish","format-standard","hentry","category-python","category-code-dev","tag-python-day"],"_links":{"self":[{"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/posts\/115","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/comments?post=115"}],"version-history":[{"count":2,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/posts\/115\/revisions"}],"predecessor-version":[{"id":153,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/posts\/115\/revisions\/153"}],"wp:attachment":[{"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/media?parent=115"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/categories?post=115"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ltrainyard.icu\/index.php\/wp-json\/wp\/v2\/tags?post=115"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}