{"id":118017,"date":"2026-10-26T03:24:38","date_gmt":"2026-10-25T19:24:38","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/118017.html"},"modified":"2026-10-26T03:24:38","modified_gmt":"2026-10-25T19:24:38","slug":"%e5%a4%a7%e6%a8%a1%e5%9e%8b%e5%ad%a6%e4%b9%a0%e4%b9%8b%e6%95%b0%e6%8d%ae%e5%88%86%e6%9e%90-%e9%9d%a2%e5%90%91%e5%a4%a7%e6%a8%a1%e5%9e%8b%e7%9a%84pandas-%e4%b8%8e-matplotlib%e5%9f%ba","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/118017.html","title":{"rendered":"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840"},"content":{"rendered":"<\/p>\n

\u6458\u8981：Pandas \u662f Python \u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u5e93，\u63d0\u4f9b Series（\u4e00\u7ef4）\u548c DataFrame（\u4e8c\u7ef4）\u4e24\u79cd\u6570\u636e\u7ed3\u6784，\u652f\u6301\u6570\u636e\u8bfb\u5199（CSV\u3001JSON）\u3001\u7d22\u5f15\u9009\u62e9（loc\/iloc）\u3001\u6570\u636e\u6e05\u6d17（\u53bb\u91cd\u3001\u7f3a\u5931\u503c）\u3001\u6570\u636e\u5408\u5e76（merge\/concat）\u3001\u5206\u7ec4\u805a\u5408（groupby）\u7b49\u5b8c\u6574\u7684\u6570\u636e\u5904\u7406\u80fd\u529b\u3002\u5728\u5927\u6a21\u578b\u5e94\u7528\u4e2d，Pandas \u8d2f\u7a7f\u8bc4\u6d4b\u96c6\u6784\u5efa\u3001\u6a21\u578b\u8f93\u51fa\u89e3\u6790\u3001badcase \u5206\u6790\u3001\u7ed3\u679c\u7edf\u8ba1\u5168\u6d41\u7a0b，\u662f\u6570\u636e\u5904\u7406\u7684\u9996\u9009\u5de5\u5177\u3002<\/p>\n

Matplotlib \u662f Python \u6700\u57fa\u7840\u7684\u53ef\u89c6\u5316\u5e93，\u5927\u6a21\u578b\u89d2\u8272\u53ea\u9700\u7b80\u8981\u4e86\u89e3，\u638c\u63e1\u6298\u7ebf\u56fe（\u770b loss \u66f2\u7ebf）\u3001\u76f4\u65b9\u56fe（\u770b\u6570\u636e\u5206\u5e03）\u3001\u67f1\u72b6\u56fe（\u770b\u7c7b\u522b\u5206\u5e03）\u4e09\u79cd\u6700\u5e38\u7528\u7684\u56fe\u5373\u53ef\u3002\u901a\u8fc7 df.plot() \u4e00\u884c\u51fa\u56fe，\u914d\u5408 savefig() \u4fdd\u5b58\u5b9e\u9a8c\u8bb0\u5f55，\u8db3\u4ee5\u6ee1\u8db3\u65e5\u5e38\u5206\u6790\u9700\u6c42\u3002<\/p>\n

\u4e24\u8005\u7ed3\u5408，\u6784\u6210\u5927\u6a21\u578b\u5e94\u7528\u5f00\u53d1\u4e2d\u300c\u6570\u636e\u5904\u7406 + \u53ef\u89c6\u5316\u300d\u7684\u5b8c\u6574\u5de5\u5177\u94fe：Pandas \u8d1f\u8d23\u6570\u636e\u6e05\u6d17\u4e0e\u5206\u6790，Matplotlib \u8d1f\u8d23\u5feb\u901f\u53ef\u89c6\u5316，\u5171\u540c\u652f\u6491\u8bc4\u6d4b\u3001\u8c03\u8bd5\u3001\u5b9e\u9a8c\u8bb0\u5f55\u7b49\u6838\u5fc3\u5de5\u4f5c\u6d41\u3002<\/p>\n

\u76ee\u5f55<\/p>\n

\u4e8c\u3001pandas\u57fa\u7840<\/p>\n

2.1 Pandas \u4ecb\u7ecd<\/p>\n

2.1.1 Pandas \u662f\u4ec0\u4e48<\/p>\n

2.1.2 Pandas \u6570\u636e\u7ed3\u6784<\/p>\n

2.1.3 \u4e3a\u4ec0\u4e48\u5927\u6a21\u578b\u89d2\u8272\u8981\u5b66 Pandas<\/p>\n

2.2 Series \u4e0e DataFrame \u57fa\u7840<\/p>\n

2.2.1 \u521b\u5efa\u5bf9\u8c61<\/p>\n

2.2.2 \u7d22\u5f15\u4e0e\u9009\u62e9<\/p>\n

2.2.3 \u8d4b\u503c\u4e0e\u65b0\u589e\u5217<\/p>\n

2.2.4 \u6392\u5e8f<\/p>\n

2.2.5 \u91cd\u547d\u540d\u4e0e\u5217\u64cd\u4f5c<\/p>\n

2.2.6 \u7c7b\u578b\u8f6c\u6362 astype<\/p>\n

2.3 \u6570\u636e\u8bfb\u5199（\u6570\u636e\u8fdb\u51fa）<\/p>\n

2.3.1 \u6a21\u578b\u8f93\u51fa \u2192 DataFrame<\/p>\n

2.3.2 CSV<\/p>\n

2.3.3 JSON<\/p>\n

2.3.4 MySQL（\u4e86\u89e3）<\/p>\n

2.3.5 \u4e0e NumPy \/ PyTorch \u4e92\u8f6c<\/p>\n

2.4 DataFrame \u8fd0\u7b97<\/p>\n

2.4.1 \u903b\u8f91\u8fd0\u7b97<\/p>\n

2.4.2 \u7edf\u8ba1\u8fd0\u7b97<\/p>\n

2.4.2.1 describe \/ mean \/ std \/ sum<\/p>\n

2.4.2.2 value_counts<\/p>\n

2.5 \u6570\u636e\u5904\u7406<\/p>\n

2.5.1 \u53bb\u91cd drop_duplicates<\/p>\n

2.5.2 \u6dfb\u52a0 \/ \u5220\u9664\u5217<\/p>\n

2.5.3 \u7f3a\u5931\u503c\u5904\u7406<\/p>\n

2.5.4 \u6570\u636e\u5408\u5e76<\/p>\n

2.5.4.1 merge（\u952e\u8fde\u63a5）<\/p>\n

2.5.5 \u6570\u636e\u5206\u7ec4\u4e0e\u805a\u5408<\/p>\n

2.5.6 \u6587\u672c\u5904\u7406<\/p>\n

2.5.7 \u968f\u673a\u91c7\u6837<\/p>\n

2.5.8 \u65f6\u95f4\u5e8f\u5217\u57fa\u7840<\/p>\n

2.5.9 \u900f\u89c6\u8868\u4e0e\u4ea4\u53c9\u8868（\u4e86\u89e3）<\/p>\n

2.6 \u7efc\u5408\u5b9e\u6218\u6848\u4f8b<\/p>\n

2.6.1 \u6a21\u578b\u8f93\u51fa\u89e3\u6790（JSON \u2192 DataFrame）<\/p>\n

2.6.2 \u8bc4\u6d4b\u96c6\u7edf\u8ba1（\u51c6\u786e\u7387\u3001badcase）<\/p>\n

2.6.3 \u6570\u636e\u6e05\u6d17\u4e0e\u5bfc\u51fa<\/p>\n

\u4e09\u3001Matplotlib \u5165\u95e8（\u7b80\u8981\u4e86\u89e3） \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0<\/p>\n

3.1 Matplotlib \u4ecb\u7ecd<\/p>\n

3.1.1 Matplotlib \u662f\u4ec0\u4e48<\/p>\n

3.1.2 \u4e3a\u4ec0\u4e48\u5927\u6a21\u578b\u89d2\u8272\u53ea\u9700\u7b80\u8981\u4e86\u89e3<\/p>\n

3.2 \u5927\u6a21\u578b\u573a\u666f\u4e0b\u6700\u5e38\u7528\u7684\u56fe<\/p>\n

3.2.1 loss \/ accuracy \u66f2\u7ebf（\u6298\u7ebf\u56fe）<\/p>\n

3.2.2 \u6570\u636e\u5206\u5e03（\u76f4\u65b9\u56fe）<\/p>\n

3.2.3 \u7c7b\u522b\u5206\u5e03（\u67f1\u72b6\u56fe）<\/p>\n

3.2.4 \u6563\u70b9\u56fe（\u4e86\u89e3）<\/p>\n

3.2.5 \u56fe\u8868\u9009\u62e9\u6307\u5357（\u4e86\u89e3）<\/p>\n

3.3 \u57fa\u7840\u7ed8\u56fe\u529f\u80fd<\/p>\n

3.3.1 \u4e0e pandas \u8054\u52a8 df.plot()\u00a0<\/p>\n

3.3.2 \u4e0e NumPy \u6570\u7ec4\u76f4\u63a5\u7ed8\u56fe\u00a0<\/p>\n

3.3.3 pyplot \u98ce\u683c plt.plot \/ scatter \/ hist<\/p>\n

3.3.4 \u6807\u9898\u3001\u5750\u6807\u8f74\u6807\u7b7e\u3001\u56fe\u4f8b<\/p>\n

3.3.5 \u4fdd\u5b58\u56fe\u7247 savefig<\/p>\n

\n

\u4e8c\u3001pandas\u57fa\u7840<\/h2>\n

\"\"<\/p>\n

2.1 Pandas \u4ecb\u7ecd<\/h3>\n

2.1.1 Pandas \u662f\u4ec0\u4e48<\/h4>\n

Pandas \u662f Python \u6700\u6d41\u884c\u7684\u6570\u636e\u5206\u6790\u5e93，\u6838\u5fc3\u662f\u4e24\u79cd\u6570\u636e\u7ed3\u6784：Series（\u4e00\u7ef4）\u548c DataFrame（\u4e8c\u7ef4）\u3002\u5b83\u5efa\u7acb\u5728 NumPy \u4e4b\u4e0a，\u63d0\u4f9b\u4e86\u66f4\u5f3a\u5927\u7684\u6570\u636e\u64cd\u4f5c\u80fd\u529b\u3002<\/p>\n

\u4e00\u53e5\u8bdd\u7406\u89e3： Pandas \u8ba9 Python \u80fd\u50cf Excel \u4e00\u6837\u5904\u7406\u8868\u683c\u6570\u636e，\u4f46\u529f\u80fd\u66f4\u5f3a\u5927\u3001\u66f4\u7075\u6d3b\u3002<\/p>\n

\u6838\u5fc3\u80fd\u529b：<\/p>\n\n\u80fd\u529b\u8bf4\u660e<\/tr>\n\n\n\n\n\n\n
\u6570\u636e\u8bfb\u5199<\/td>\nCSV\u3001JSON\u3001Excel\u3001MySQL \u7b49<\/td>\n<\/tr>\n
\u6570\u636e\u6e05\u6d17<\/td>\n\u7f3a\u5931\u503c\u3001\u53bb\u91cd\u3001\u7c7b\u578b\u8f6c\u6362<\/td>\n<\/tr>\n
\u6570\u636e\u7b5b\u9009<\/td>\n\u6761\u4ef6\u7b5b\u9009\u3001\u6392\u5e8f\u3001\u5206\u7ec4<\/td>\n<\/tr>\n
\u6570\u636e\u805a\u5408<\/td>\ngroupby\u3001\u900f\u89c6\u8868<\/td>\n<\/tr>\n
\u6570\u636e\u5408\u5e76<\/td>\nmerge\u3001concat<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

2.1.2 Pandas \u6570\u636e\u7ed3\u6784<\/h4>\n

\"\"<\/p>\n

2.1.2.1 Series \u5bf9\u8c61<\/h5>\n

Series \u662f Pandas \u7684\u4e00\u7ef4\u6570\u636e\u7ed3\u6784，\u7531\u7d22\u5f15（index）\u548c\u503c（values）\u7ec4\u6210\u3002\u53ef\u4ee5\u7406\u89e3\u4e3a"\u5e26\u6807\u7b7e\u7684\u4e00\u7ef4\u6570\u7ec4"\u3002<\/p>\n

2.1.2.2 DataFrame \u5bf9\u8c61<\/h5>\n

DataFrame \u662f Pandas \u7684\u4e8c\u7ef4\u6570\u636e\u7ed3\u6784，\u7531\u884c\u7d22\u5f15（index）\u548c\u5217\u7d22\u5f15（columns）\u7ec4\u6210\u3002\u53ef\u4ee5\u7406\u89e3\u4e3a"\u5e26\u6807\u7b7e\u7684\u4e8c\u7ef4\u8868\u683c"\u3002<\/p>\n

2.1.2.3 \u4e0e NumPy \u7684\u6838\u5fc3\u533a\u522b<\/h5>\n\n\u7ef4\u5ea6NumPyPandas<\/tr>\n\n\n\n\n\n\n\n
\u7ef4\u5ea6<\/td>\n\u591a\u7ef4\u6570\u7ec4<\/td>\n\u4e00\u7ef4 Series\u3001\u4e8c\u7ef4 DataFrame<\/td>\n<\/tr>\n
\u7d22\u5f15<\/td>\n\u9690\u5f0f\u6574\u6570\u7d22\u5f15<\/td>\n\u652f\u6301\u6807\u7b7e\u7d22\u5f15<\/td>\n<\/tr>\n
\u7c7b\u578b<\/td>\n\u8981\u6c42\u7edf\u4e00 dtype<\/td>\n\u5217\u95f4\u53ef\u4e0d\u540c dtype<\/td>\n<\/tr>\n
\u7f3a\u5931\u503c<\/td>\nnan（\u4ec5\u6d6e\u70b9）<\/td>\nNone \/ NaN \/ NaT<\/td>\n<\/tr>\n
\u5217\u540d<\/td>\n\u65e0<\/td>\n\u6709\u5217\u540d<\/td>\n<\/tr>\n
\u5e94\u7528<\/td>\n\u6570\u503c\u8ba1\u7b97<\/td>\n\u6570\u636e\u5904\u7406<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

2.1.3 \u4e3a\u4ec0\u4e48\u5927\u6a21\u578b\u89d2\u8272\u8981\u5b66 Pandas<\/h4>\n

Pandas \u662f\u5927\u6a21\u578b\u5e94\u7528\u5f00\u53d1\u4e2d\u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u5de5\u5177，\u8d2f\u7a7f\u8bc4\u6d4b\u3001\u6e05\u6d17\u3001\u5206\u6790\u5168\u6d41\u7a0b\u3002<\/p>\n

\u5e94\u7528\u573a\u666f<\/p>\n\n\u573a\u666f\u8bf4\u660e<\/tr>\n\n\n\n\n\n\n
\u6570\u636e\u6e05\u6d17<\/td>\n\u6e05\u6d17\u8bad\u7ec3\u6570\u636e\u3001\u8bc4\u6d4b\u6570\u636e<\/td>\n<\/tr>\n
\u8bc4\u6d4b\u96c6<\/td>\n\u6784\u5efa\u3001\u7ba1\u7406\u8bc4\u6d4b\u96c6<\/td>\n<\/tr>\n
badcase<\/td>\n\u5206\u6790\u6a21\u578b\u9519\u8bef\u6848\u4f8b<\/td>\n<\/tr>\n
\u6a21\u578b\u8f93\u51fa\u89e3\u6790<\/td>\nJSON \u2192 DataFrame，\u6279\u91cf\u5206\u6790<\/td>\n<\/tr>\n
\u7ed3\u679c\u7edf\u8ba1<\/td>\n\u51c6\u786e\u7387\u3001\u53ec\u56de\u7387\u3001\u5206\u5e03\u7edf\u8ba1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

2.2 Series \u4e0e DataFrame \u57fa\u7840<\/h3>\n

2.2.1 \u521b\u5efa\u5bf9\u8c61<\/h4>\n

Pandas \u652f\u6301\u4ece\u591a\u79cd\u6570\u636e\u6e90\u521b\u5efa Series \u548c DataFrame\u3002<\/p>\n

# \u521b\u5efa Series \u548c DataFrame<\/p>\n

import pandas as pd
\nimport numpy as np<\/p>\n

# ===== Series =====
\n# \u4ece\u5217\u8868
\ns1 = pd.Series([1, 2, 3, 4, 5])<\/p>\n

# \u4ece\u5b57\u5178
\ns2 = pd.Series({"a": 1, "b": 2, "c": 3})<\/p>\n

# \u4ece ndarray
\ns3 = pd.Series(np.array([10, 20, 30]))<\/p>\n

# \u6307\u5b9a\u7d22\u5f15
\ns4 = pd.Series([1, 2, 3], index=["x", "y", "z"])<\/p>\n

print(f"\u4ece\u5217\u8868：\\\\n{s1}")
\nprint(f"\\\\n\u4ece\u5b57\u5178：\\\\n{s2}")
\nprint(f"\\\\n\u6307\u5b9a\u7d22\u5f15：\\\\n{s4}")<\/p>\n

# ===== DataFrame =====
\n# \u4ece\u5b57\u5178（\u6700\u5e38\u89c1）
\ndf1 = pd.DataFrame({
\n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
\n "age": [25, 30, 28],
\n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"]
\n})<\/p>\n

# \u4ece\u5217\u8868\u7684\u5b57\u5178
\ndf2 = pd.DataFrame([
\n {"name": "\u5f20\u4e09", "age": 25},
\n {"name": "\u674e\u56db", "age": 30},
\n])<\/p>\n

# \u4ece ndarray
\ndf3 = pd.DataFrame(np.array([[1, 2], [3, 4]]), columns=["A", "B"])<\/p>\n

print(f"\\\\nDataFrame \u4ece\u5b57\u5178：\\\\n{df1}")
\nprint(f"\\\\nDataFrame \u4ece\u5217\u8868\u7684\u5b57\u5178：\\\\n{df2}")
\nprint(f"\\\\nDataFrame \u4ece ndarray：\\\\n{df3}") <\/p>\n

\u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
pd.Series()<\/td>\n\u521b\u5efa Series\u3002\u53c2\u6570\u4e3a\u6570\u636e，index \u6307\u5b9a\u7d22\u5f15<\/td>\n<\/tr>\n
pd.DataFrame()<\/td>\n\u521b\u5efa DataFrame\u3002\u53c2\u6570\u4e3a\u5b57\u5178\u3001\u5217\u8868\u3001ndarray，columns \u6307\u5b9a\u5217\u540d<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\u8bf4\u660e：<\/p>\n

    \n
  • \n

    Series \u4ece\u5217\u8868\u3001\u5b57\u5178\u3001ndarray \u521b\u5efa<\/p>\n<\/li>\n

  • \n

    DataFrame \u4ece\u5b57\u5178（\u6700\u5e38\u89c1）\u3001\u5217\u8868\u7684\u5b57\u5178\u3001ndarray \u521b\u5efa<\/p>\n<\/li>\n

  • \n

    columns \u6307\u5b9a\u5217\u540d<\/p>\n<\/li>\n<\/ul>\n

    2.2.2 \u7d22\u5f15\u4e0e\u9009\u62e9<\/h4>\n

    2.2.2.1 loc（\u6807\u7b7e\u7d22\u5f15）<\/h5>\n

    loc \u57fa\u4e8e\u6807\u7b7e\u7d22\u5f15，\u4f7f\u7528\u884c\u7d22\u5f15\u540d\u548c\u5217\u540d\u3002<\/p>\n

    # loc：\u6807\u7b7e\u7d22\u5f15<\/p>\n

    import pandas as pd<\/p>\n

    df = pd.DataFrame({
    \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
    \n "age": [25, 30, 28],
    \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"]
    \n}, index=["a", "b", "c"])<\/p>\n

    # \u53d6\u5355\u884c
    \nprint(f"df.loc['a']：\\\\n{df.loc['a']}")<\/p>\n

    # \u53d6\u5355\u5217
    \nprint(f"\\\\ndf.loc[:, 'name']：\\\\n{df.loc[:, 'name']}")<\/p>\n

    # \u53d6\u5355\u503c
    \nprint(f"\\\\ndf.loc['a', 'name']：{df.loc['a', 'name']}")<\/p>\n

    # \u53d6\u591a\u884c\u591a\u5217
    \nprint(f"\\\\ndf.loc['a':'b', ['name', 'age']]：\\\\n{df.loc['a':'b', ['name', 'age']]}") <\/p>\n

    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u8bed\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
    df.loc[\u884c\u6807\u7b7e]<\/td>\n\u53d6\u884c<\/td>\n<\/tr>\n
    df.loc[:, \u5217\u6807\u7b7e]<\/td>\n\u53d6\u5217<\/td>\n<\/tr>\n
    df.loc[\u884c\u6807\u7b7e, \u5217\u6807\u7b7e]<\/td>\n\u53d6\u5355\u503c<\/td>\n<\/tr>\n
    df.loc[\u884c\u6807\u7b7e\u5217\u8868, \u5217\u6807\u7b7e\u5217\u8868]<\/td>\n\u53d6\u591a\u884c\u591a\u5217<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

    \u8bf4\u660e：<\/p>\n

      \n
    • \n

      loc \u57fa\u4e8e\u6807\u7b7e<\/p>\n<\/li>\n

    • \n

      \u5207\u7247 'a':'b' \u5305\u542b\u53f3\u7aef<\/p>\n<\/li>\n<\/ul>\n

      2.2.2.2 iloc（\u4f4d\u7f6e\u7d22\u5f15）<\/h5>\n

      iloc \u57fa\u4e8e\u4f4d\u7f6e\u7d22\u5f15，\u4f7f\u7528\u6574\u6570\u4f4d\u7f6e\u3002<\/p>\n

      # iloc：\u4f4d\u7f6e\u7d22\u5f15<\/p>\n

      import pandas as pd<\/p>\n

      df = pd.DataFrame({
      \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
      \n "age": [25, 30, 28],
      \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"]
      \n})<\/p>\n

      # \u53d6\u5355\u884c
      \nprint(f"df.iloc[0]：\\\\n{df.iloc[0]}")<\/p>\n

      # \u53d6\u5355\u5217
      \nprint(f"\\\\ndf.iloc[:, 0]：\\\\n{df.iloc[:, 0]}")<\/p>\n

      # \u53d6\u5355\u503c
      \nprint(f"\\\\ndf.iloc[0, 0]：{df.iloc[0, 0]}")<\/p>\n

      # \u53d6\u591a\u884c\u591a\u5217
      \nprint(f"\\\\ndf.iloc[0:2, 0:2]：\\\\n{df.iloc[0:2, 0:2]}") <\/p>\n

      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u8bed\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
      df.iloc[\u884c\u4f4d\u7f6e]<\/td>\n\u53d6\u884c<\/td>\n<\/tr>\n
      df.iloc[:, \u5217\u4f4d\u7f6e]<\/td>\n\u53d6\u5217<\/td>\n<\/tr>\n
      df.iloc[\u884c\u4f4d\u7f6e, \u5217\u4f4d\u7f6e]<\/td>\n\u53d6\u5355\u503c<\/td>\n<\/tr>\n
      df.iloc[\u884c\u4f4d\u7f6e\u5207\u7247, \u5217\u4f4d\u7f6e\u5207\u7247]<\/td>\n\u53d6\u591a\u884c\u591a\u5217<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

      \u8bf4\u660e：<\/p>\n

        \n
      • \n

        iloc \u57fa\u4e8e\u4f4d\u7f6e<\/p>\n<\/li>\n

      • \n

        \u5207\u7247 0:2 \u4e0d\u542b\u53f3\u7aef<\/p>\n<\/li>\n<\/ul>\n

        2.2.2.3 \u5e03\u5c14\u7d22\u5f15<\/h5>\n

        \u5e03\u5c14\u7d22\u5f15\u7528\u6761\u4ef6\u8868\u8fbe\u5f0f\u7b5b\u9009\u6570\u636e，\u8fd4\u56de\u6ee1\u8db3\u6761\u4ef6\u7684\u884c\u3002<\/p>\n

        # \u5e03\u5c14\u7d22\u5f15<\/p>\n

        import pandas as pd<\/p>\n

        df = pd.DataFrame({
        \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
        \n "age": [25, 30, 28],
        \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"]
        \n})<\/p>\n

        # \u5355\u6761\u4ef6
        \nresult = df[df["age"] > 26]
        \nprint(f"age > 26：\\\\n{result}")<\/p>\n

        # \u591a\u6761\u4ef6
        \nresult = df[(df["age"] > 26) & (df["city"] == "\u4e0a\u6d77")]
        \nprint(f"\\\\nage > 26 \u4e14 city == \u4e0a\u6d77：\\\\n{result}")<\/p>\n

        # isin
        \nresult = df[df["city"].isin(["\u676d\u5dde", "\u5317\u4eac"])]
        \nprint(f"\\\\ncity in [\u676d\u5dde, \u5317\u4eac]：\\\\n{result}") <\/p>\n

        \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u8bed\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n
        df[\u6761\u4ef6]<\/td>\n\u5e03\u5c14\u7d22\u5f15\u7b5b\u9009<\/td>\n<\/tr>\n
        df[\u6761\u4ef61 & \u6761\u4ef62]<\/td>\n\u591a\u6761\u4ef6，\u7528 & \u4e14\u52a0\u62ec\u53f7<\/td>\n<\/tr>\n
        df[\u5217].isin(\u5217\u8868)<\/td>\n\u5224\u65ad\u662f\u5426\u5728\u5217\u8868\u4e2d<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

        \u8bf4\u660e：<\/p>\n

          \n
        • \n

          \u5e03\u5c14\u7d22\u5f15\u8fd4\u56de\u6ee1\u8db3\u6761\u4ef6\u7684\u884c<\/p>\n<\/li>\n

        • \n

          \u591a\u6761\u4ef6\u7528 &\u3001|，\u4e0d\u80fd\u7528 and\u3001or<\/p>\n<\/li>\n<\/ul>\n

          2.2.2.4 \u6613\u9519\u70b9：loc \u542b\u53f3\u7aef，iloc \u4e0d\u542b\u53f3\u7aef<\/h5>\n

          # \u9519\u8bef：\u4ee5\u4e3a loc \u548c iloc \u5207\u7247\u884c\u4e3a\u4e00\u81f4<\/p>\n

          import pandas as pd<\/p>\n

          df = pd.DataFrame({
          \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
          \n "age": [25, 30, 28]
          \n}, index=["a", "b", "c"])<\/p>\n

          # loc：\u542b\u53f3\u7aef
          \nprint(f"loc['a':'b']：\\\\n{df.loc['a':'b']}") # \u5305\u542b a \u548c b<\/p>\n

          # iloc：\u4e0d\u542b\u53f3\u7aef
          \nprint(f"\\\\niloc[0:2]：\\\\n{df.iloc[0:2]}") # \u5305\u542b 0 \u548c 1，\u4e0d\u542b 2 <\/p>\n

          \u8bf4\u660e：<\/p>\n\n\u7d22\u5f15\u65b9\u5f0f\u5207\u7247\u884c\u4e3a<\/tr>\n\n\n\n
          loc<\/td>\n\u542b\u53f3\u7aef<\/td>\n<\/tr>\n
          iloc<\/td>\n\u4e0d\u542b\u53f3\u7aef<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
          2.2.2.5 \u6613\u9519\u70b9：\u94fe\u5f0f\u8d4b\u503c SettingWithCopy<\/h5>\n

          # \u9519\u8bef：\u94fe\u5f0f\u8d4b\u503c<\/p>\n

          import pandas as pd<\/p>\n

          df = pd.DataFrame({
          \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
          \n "age": [25, 30, 28]
          \n})<\/p>\n

          # \u94fe\u5f0f\u8d4b\u503c：\u5148\u7b5b\u9009，\u518d\u8d4b\u503c
          \n# df[df["age"] > 26]["age"] = 100 # SettingWithCopyWarning<\/p>\n

          # \u6b63\u786e：\u7528 .loc \u4e00\u6b65\u5230\u4f4d
          \ndf.loc[df["age"] > 26, "age"] = 100
          \nprint(df) <\/p>\n

          \u8bf4\u660e：<\/p>\n

            \n
          • \n

            \u94fe\u5f0f\u8d4b\u503c\u53ef\u80fd\u4fee\u6539\u7684\u662f\u526f\u672c，\u800c\u975e\u539f DataFrame<\/p>\n<\/li>\n

          • \n

            \u7528 .loc[\u6761\u4ef6, \u5217] = \u503c \u4e00\u6b65\u5230\u4f4d<\/p>\n<\/li>\n

          • \n

            \u907f\u514d\u94fe\u5f0f\u8d4b\u503c<\/p>\n<\/li>\n<\/ul>\n

            2.2.3 \u8d4b\u503c\u4e0e\u65b0\u589e\u5217<\/h4>\n

            Pandas \u652f\u6301\u76f4\u63a5\u8d4b\u503c\u65b0\u589e\u5217\u3001\u4fee\u6539\u5217\u503c\u3001\u6761\u4ef6\u8d4b\u503c\u3002<\/p>\n

            # \u8d4b\u503c\u4e0e\u65b0\u589e\u5217<\/p>\n

            import pandas as pd<\/p>\n

            df = pd.DataFrame({
            \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
            \n "age": [25, 30, 28]
            \n})<\/p>\n

            # ===== \u65b0\u589e\u5217：\u5e38\u91cf =====
            \ndf["city"] = "\u676d\u5dde"
            \nprint(f"\u65b0\u589e\u5e38\u91cf\u5217：\\\\n{df}")<\/p>\n

            # ===== \u65b0\u589e\u5217：\u8ba1\u7b97 =====
            \ndf["age_plus_10"] = df["age"] + 10
            \nprint(f"\\\\n\u65b0\u589e\u8ba1\u7b97\u5217：\\\\n{df}")<\/p>\n

            # ===== \u65b0\u589e\u5217：\u6761\u4ef6\u8d4b\u503c =====
            \ndf["is_adult"] = df["age"] >= 18
            \nprint(f"\\\\n\u65b0\u589e\u6761\u4ef6\u5217：\\\\n{df}")<\/p>\n

            # ===== \u4fee\u6539\u5217\u503c =====
            \ndf["age"] = df["age"] + 1
            \nprint(f"\\\\n\u4fee\u6539\u5217\u503c：\\\\n{df}")<\/p>\n

            # ===== \u5220\u9664\u5217 =====
            \ndf = df.drop(columns=["age_plus_10"])
            \nprint(f"\\\\n\u5220\u9664\u5217：\\\\n{df}") <\/p>\n

            \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
            df["\u65b0\u5217"] = \u503c<\/td>\n\u65b0\u589e\u5217\u6216\u4fee\u6539\u5217<\/td>\n<\/tr>\n
            df.drop(columns=[…])<\/td>\n\u5220\u9664\u5217\u3002columns \u6307\u5b9a\u5217\u540d<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

            \u8bf4\u660e：<\/p>\n

              \n
            • \n

              \u65b0\u589e\u5217\u76f4\u63a5\u8d4b\u503c<\/p>\n<\/li>\n

            • \n

              \u6761\u4ef6\u8d4b\u503c\u7528\u5e03\u5c14\u8868\u8fbe\u5f0f<\/p>\n<\/li>\n

            • \n

              drop \u8fd4\u56de\u65b0 DataFrame，\u4e0d\u4fee\u6539\u539f\u6570\u636e<\/p>\n<\/li>\n<\/ul>\n

              2.2.4 \u6392\u5e8f<\/h4>\n

              2.2.4.1 sort_values \/ sort_index [L3]<\/h3>\n

              # \u6392\u5e8f<\/p>\n

              import pandas as pd<\/p>\n

              df = pd.DataFrame({
              \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
              \n "age": [25, 30, 28]
              \n}, index=["c", "a", "b"])<\/p>\n

              # \u6309\u503c\u6392\u5e8f：\u9ed8\u8ba4\u5347\u5e8f
              \nprint(f"\u6309 age \u5347\u5e8f：\\\\n{df.sort_values('age')}")<\/p>\n

              # \u964d\u5e8f
              \nprint(f"\\\\n\u6309 age \u964d\u5e8f：\\\\n{df.sort_values('age', ascending=False)}")<\/p>\n

              # \u6309\u7d22\u5f15\u6392\u5e8f
              \nprint(f"\\\\n\u6309\u7d22\u5f15\u6392\u5e8f：\\\\n{df.sort_index()}") <\/p>\n

              \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
              df.sort_values()<\/td>\n\u6309\u503c\u6392\u5e8f\u3002by \u6307\u5b9a\u5217，ascending \u63a7\u5236\u5347\u964d\u5e8f<\/td>\n<\/tr>\n
              df.sort_index()<\/td>\n\u6309\u7d22\u5f15\u6392\u5e8f<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

              \u8bf4\u660e：<\/p>\n

                \n
              • \n

                \u9ed8\u8ba4\u5347\u5e8f<\/p>\n<\/li>\n

              • \n

                \u8fd4\u56de\u65b0 DataFrame，\u4e0d\u4fee\u6539\u539f\u6570\u636e<\/p>\n<\/li>\n<\/ul>\n

                2.2.4.2 \u6613\u9519\u70b9：\u9ed8\u8ba4\u4e0d\u539f\u5730，\u9700 inplace \u6216\u91cd\u65b0\u8d4b\u503c [L2]<\/h3>\n

                # \u9519\u8bef：\u4ee5\u4e3a\u6392\u5e8f\u4f1a\u4fee\u6539\u539f DataFrame<\/p>\n

                import pandas as pd<\/p>\n

                df = pd.DataFrame({"age": [30, 25, 28]})<\/p>\n

                df.sort_values("age") # \u8fd4\u56de\u65b0 DataFrame，\u539f df \u672a\u53d8
                \nprint(f"\u539f df：\\\\n{df}") # \u987a\u5e8f\u672a\u53d8<\/p>\n

                # \u6b63\u786e：\u91cd\u65b0\u8d4b\u503c
                \ndf = df.sort_values("age")
                \nprint(f"\\\\n\u91cd\u65b0\u8d4b\u503c\u540e：\\\\n{df}")<\/p>\n

                # \u6216\u4f7f\u7528 inplace
                \ndf.sort_values("age", inplace=True) <\/p>\n

                \u8bf4\u660e：<\/p>\n

                  \n
                • \n

                  sort_values \u9ed8\u8ba4\u8fd4\u56de\u65b0 DataFrame<\/p>\n<\/li>\n

                • \n

                  \u9700\u91cd\u65b0\u8d4b\u503c\u6216 inplace=True<\/p>\n<\/li>\n

                • \n

                  \u63a8\u8350\u91cd\u65b0\u8d4b\u503c<\/p>\n<\/li>\n<\/ul>\n

                  2.2.5 \u91cd\u547d\u540d\u4e0e\u5217\u64cd\u4f5c<\/h4>\n

                  # \u91cd\u547d\u540d\u4e0e\u5217\u64cd\u4f5c<\/p>\n

                  import pandas as pd<\/p>\n

                  df = pd.DataFrame({
                  \n "name": ["\u5f20\u4e09", "\u674e\u56db"],
                  \n "age": [25, 30]
                  \n})<\/p>\n

                  # \u65b9\u5f0f\u4e00：rename
                  \ndf1 = df.rename(columns={"name": "\u59d3\u540d", "age": "\u5e74\u9f84"})
                  \nprint(f"rename：\\\\n{df1}")<\/p>\n

                  # \u65b9\u5f0f\u4e8c：\u76f4\u63a5\u8d4b\u503c columns
                  \ndf2 = df.copy()
                  \ndf2.columns = ["\u59d3\u540d", "\u5e74\u9f84"]
                  \nprint(f"\\\\n\u76f4\u63a5\u8d4b\u503c columns：\\\\n{df2}")<\/p>\n

                  # \u67e5\u770b\u5217\u540d
                  \nprint(f"\\\\n\u5217\u540d：{df.columns.tolist()}") <\/p>\n

                  \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                  df.rename()<\/td>\n\u91cd\u547d\u540d\u3002columns \u6307\u5b9a\u5217\u6620\u5c04，index \u6307\u5b9a\u7d22\u5f15\u6620\u5c04<\/td>\n<\/tr>\n
                  df.columns<\/td>\n\u5217\u540d，\u53ef\u8d4b\u503c\u4fee\u6539<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                  \u8bf4\u660e：<\/p>\n

                    \n
                  • \n

                    rename \u8fd4\u56de\u65b0 DataFrame<\/p>\n<\/li>\n

                  • \n

                    columns \u76f4\u63a5\u8d4b\u503c\u4fee\u6539<\/p>\n<\/li>\n<\/ul>\n

                    2.2.6 \u7c7b\u578b\u8f6c\u6362 astype<\/h4>\n

                    astype() \u7528\u4e8e\u8f6c\u6362\u6570\u636e\u7c7b\u578b，\u8fd4\u56de\u65b0 DataFrame\u3002<\/p>\n

                    # \u7c7b\u578b\u8f6c\u6362 astype<\/p>\n

                    import pandas as pd<\/p>\n

                    df = pd.DataFrame({
                    \n "age": ["25", "30", "28"],
                    \n "score": [85.5, 92.3, 78.8]
                    \n})<\/p>\n

                    print(f"\u539f\u59cb\u7c7b\u578b：\\\\n{df.dtypes}")<\/p>\n

                    # \u5b57\u7b26\u4e32\u8f6c\u6574\u6570
                    \ndf["age"] = df["age"].astype(int)
                    \nprint(f"\\\\n\u8f6c\u6362\u540e\u7c7b\u578b：\\\\n{df.dtypes}")<\/p>\n

                    # \u6d6e\u70b9\u8f6c\u6574\u6570
                    \ndf["score_int"] = df["score"].astype(int)
                    \nprint(f"\\\\n\u8f6c\u6362\u540e：\\\\n{df}")<\/p>\n

                    # \u8f6c\u4e3a category（\u8282\u7701\u5185\u5b58）
                    \ndf["age_cat"] = df["age"].astype("category")
                    \nprint(f"\\\\ncategory \u7c7b\u578b：{df['age_cat'].dtype}") <\/p>\n

                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n
                    df[\u5217].astype()<\/td>\n\u8f6c\u6362\u5217\u7c7b\u578b\u3002\u53c2\u6570\u4e3a\u76ee\u6807\u7c7b\u578b，\u5982 int\u3001float\u3001str\u3001category<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                    \u8bf4\u660e：<\/p>\n

                      \n
                    • \n

                      astype \u8fd4\u56de\u65b0 Series，\u9700\u91cd\u65b0\u8d4b\u503c<\/p>\n<\/li>\n

                    • \n

                      \u6d6e\u70b9\u8f6c\u6574\u6570\u4f1a\u622a\u65ad\u5c0f\u6570<\/p>\n<\/li>\n

                    • \n

                      category \u7c7b\u578b\u8282\u7701\u5185\u5b58<\/p>\n<\/li>\n<\/ul>\n

                      2.3 \u6570\u636e\u8bfb\u5199（\u6570\u636e\u8fdb\u51fa）<\/h3>\n

                      2.3.1 \u6a21\u578b\u8f93\u51fa \u2192 DataFrame<\/h4>\n

                      \u5927\u6a21\u578b\u8f93\u51fa\u901a\u5e38\u4e3a JSON \u683c\u5f0f（\u5217\u8868\u7684\u5b57\u5178\u3001\u5d4c\u5957\u5b57\u5178），\u9700\u8f6c\u4e3a DataFrame \u8fdb\u884c\u5206\u6790\u3002\u8fd9\u662f\u5927\u6a21\u578b\u8bc4\u6d4b\u3001badcase \u5206\u6790\u7684\u7b2c\u4e00\u6b65\u3002<\/p>\n

                      2.3.2 CSV<\/h4>\n

                      CSV \u662f\u6700\u5e38\u7528\u7684\u6570\u636e\u5b58\u50a8\u683c\u5f0f，Pandas \u63d0\u4f9b read_csv \u548c to_csv \u8bfb\u5199\u3002<\/p>\n

                      # CSV \u8bfb\u5199<\/p>\n

                      import pandas as pd<\/p>\n

                      # ===== \u5199\u5165 CSV =====
                      \ndf = pd.DataFrame({
                      \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
                      \n "age": [25, 30, 28],
                      \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"]
                      \n})<\/p>\n

                      df.to_csv("data.csv", index=False, encoding="utf-8")<\/p>\n

                      # ===== \u8bfb\u53d6 CSV =====
                      \ndf2 = pd.read_csv("data.csv", encoding="utf-8")
                      \nprint(f"\\\\n\u8bfb\u53d6\u7ed3\u679c：\\\\n{df2}") <\/p>\n

                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                      df.to_csv()<\/td>\n\u5199\u5165 CSV\u3002index=False \u4e0d\u5199\u7d22\u5f15，encoding \u6307\u5b9a\u7f16\u7801<\/td>\n<\/tr>\n
                      pd.read_csv()<\/td>\n\u8bfb\u53d6 CSV\u3002encoding \u6307\u5b9a\u7f16\u7801，sep \u6307\u5b9a\u5206\u9694\u7b26<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      \u5e38\u7528\u53c2\u6570：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                      index<\/td>\n\u662f\u5426\u5199\u5165\u7d22\u5f15，\u9ed8\u8ba4 True<\/td>\n<\/tr>\n
                      encoding<\/td>\n\u7f16\u7801，\u5e38\u7528 utf-8\u3001gbk<\/td>\n<\/tr>\n
                      sep<\/td>\n\u5206\u9694\u7b26，\u9ed8\u8ba4 ,<\/td>\n<\/tr>\n
                      header<\/td>\n\u662f\u5426\u6709\u8868\u5934，\u9ed8\u8ba4 0<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      2.3.3 JSON<\/h4>\n

                      JSON \u662f API \u548c\u6a21\u578b\u8f93\u51fa\u7684\u5e38\u89c1\u683c\u5f0f，Pandas \u652f\u6301\u8bfb\u5199\u3002<\/p>\n

                      # JSON \u8bfb\u5199<\/p>\n

                      import pandas as pd<\/p>\n

                      df = pd.DataFrame({
                      \n "name": ["\u5f20\u4e09", "\u674e\u56db"],
                      \n "age": [25, 30]
                      \n})<\/p>\n

                      # \u5199\u5165 JSON
                      \ndf.to_json("data.json", orient="records", force_ascii=False)<\/p>\n

                      # \u8bfb\u53d6 JSON
                      \ndf2 = pd.read_json("data.json")
                      \nprint(f"\\\\n\u8bfb\u53d6\u7ed3\u679c：\\\\n{df2}") <\/p>\n

                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                      df.to_json()<\/td>\n\u5199\u5165 JSON\u3002orient \u6307\u5b9a\u65b9\u5411，force_ascii=False \u4fdd\u7559\u4e2d\u6587<\/td>\n<\/tr>\n
                      pd.read_json()<\/td>\n\u8bfb\u53d6 JSON<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      orient \u53c2\u6570：<\/p>\n\n\u503c\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                      "records"<\/td>\n\u5217\u8868\u7684\u5b57\u5178（\u63a8\u8350）<\/td>\n<\/tr>\n
                      "index"<\/td>\n\u5b57\u5178\u7684\u5b57\u5178<\/td>\n<\/tr>\n
                      "columns"<\/td>\n\u5217\u5bfc\u5411<\/td>\n<\/tr>\n
                      "values"<\/td>\n\u4ec5\u503c<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      2.3.4 MySQL（\u4e86\u89e3）<\/h4>\n

                      Pandas \u53ef\u4ee5\u8bfb\u5199 MySQL \u6570\u636e\u5e93，\u9700\u914d\u5408 sqlalchemy \u6216 pymysql\u3002<\/p>\n

                      \u5b9e\u9645\u9879\u76ee\u4e2d\u4f7f\u7528\u8f83\u5c11，\u4e86\u89e3\u5373\u53ef\u3002<\/p>\n

                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                      pd.read_sql()<\/td>\n\u8bfb\u53d6 SQL \u67e5\u8be2\u7ed3\u679c<\/td>\n<\/tr>\n
                      df.to_sql()<\/td>\n\u5199\u5165\u6570\u636e\u5e93\u8868<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      2.3.5 \u4e0e NumPy \/ PyTorch \u4e92\u8f6c<\/h4>\n

                      2.3.5.1 .values \/ .to_numpy()<\/h5>\n

                      DataFrame \u548c Series \u53ef\u4ee5\u8f6c\u4e3a NumPy \u6570\u7ec4，\u7528\u4e8e\u6570\u503c\u8ba1\u7b97\u3002<\/p>\n

                      # \u4e0e NumPy \u4e92\u8f6c<\/p>\n

                      import pandas as pd
                      \nimport numpy as np<\/p>\n

                      df = pd.DataFrame({
                      \n "age": [25, 30, 28],
                      \n "score": [85.5, 92.3, 78.8]
                      \n})<\/p>\n

                      # \u65b9\u5f0f\u4e00：.values
                      \narr1 = df.values
                      \nprint(f".values：\\\\n{arr1}")
                      \nprint(f"\u7c7b\u578b：{type(arr1)}，dtype：{arr1.dtype}")<\/p>\n

                      # \u65b9\u5f0f\u4e8c：.to_numpy()（\u63a8\u8350）
                      \narr2 = df.to_numpy()
                      \nprint(f"\\\\n.to_numpy()：\\\\n{arr2}")<\/p>\n

                      # Series \u8f6c NumPy
                      \ns = df["age"]
                      \narr3 = s.to_numpy()
                      \nprint(f"\\\\nSeries.to_numpy()：{arr3}") <\/p>\n

                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                      df.values<\/td>\n\u8fd4\u56de NumPy \u6570\u7ec4（\u65e7\u7248，\u4e0d\u63a8\u8350）<\/td>\n<\/tr>\n
                      df.to_numpy()<\/td>\n\u8fd4\u56de NumPy \u6570\u7ec4（\u63a8\u8350）\u3002dtype \u6307\u5b9a\u7c7b\u578b<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                      \u8bf4\u660e：<\/p>\n

                        \n
                      • \n

                        .to_numpy() \u662f\u63a8\u8350\u65b9\u6cd5，.values \u662f\u65e7\u7248<\/p>\n<\/li>\n

                      • \n

                        \u6df7\u5408 dtype \u7684 DataFrame \u8f6c\u4e3a NumPy \u4f1a\u7edf\u4e00\u4e3a object \u6216 float<\/p>\n<\/li>\n<\/ul>\n

                        2.3.5.2 \u4e0e PyTorch \u5bf9\u63a5\u524d\u7684 dtype \u6ce8\u610f<\/h5>\n

                        Pandas \u2192 NumPy \u2192 PyTorch \u662f\u5927\u6a21\u578b\u6570\u636e\u5904\u7406\u7684\u5e38\u89c1\u94fe\u8def\u3002\u5173\u952e\u6ce8\u610f：NumPy \u9ed8\u8ba4 float64，PyTorch \u9ed8\u8ba4 float32\u3002<\/p>\n

                        # \u4e0e PyTorch \u5bf9\u63a5<\/p>\n

                        import pandas as pd
                        \nimport numpy as np
                        \nimport torch<\/p>\n

                        df = pd.DataFrame({
                        \n "age": [25, 30, 28],
                        \n "score": [85.5, 92.3, 78.8]
                        \n})<\/p>\n

                        # \u95ee\u9898：\u9ed8\u8ba4 float64
                        \narr = df.to_numpy()
                        \ntensor = torch.from_numpy(arr)
                        \nprint(f"\u9ed8\u8ba4 dtype：{tensor.dtype}") # torch.float64<\/p>\n

                        # \u6b63\u786e：\u663e\u5f0f\u8f6c float32
                        \narr = df.to_numpy(dtype=np.float32)
                        \ntensor = torch.from_numpy(arr)
                        \nprint(f"\u8f6c\u6362\u540e dtype：{tensor.dtype}") # torch.float32 <\/p>\n

                        \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                        df.to_numpy(dtype=…)<\/td>\n\u8f6c\u4e3a\u6307\u5b9a dtype \u7684 NumPy \u6570\u7ec4<\/td>\n<\/tr>\n
                        torch.from_numpy()<\/td>\nNumPy \u6570\u7ec4\u8f6c PyTorch \u5f20\u91cf（\u5171\u4eab\u5185\u5b58）<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                        \u8bf4\u660e：<\/p>\n

                          \n
                        • \n

                          NumPy \u9ed8\u8ba4 float64，PyTorch \u9ed8\u8ba4 float32<\/p>\n<\/li>\n

                        • \n

                          \u5bf9\u63a5\u524d\u9700\u663e\u5f0f\u8f6c\u6362<\/p>\n<\/li>\n

                        • \n

                          \u5927\u6a21\u578b\u4e2d\u7edf\u4e00\u4f7f\u7528 float32<\/p>\n<\/li>\n<\/ul>\n

                          2.4 DataFrame \u8fd0\u7b97<\/h3>\n

                          2.4.1 \u903b\u8f91\u8fd0\u7b97<\/h4>\n

                          Pandas \u7684\u903b\u8f91\u8fd0\u7b97\u7528\u4e8e\u7ec4\u5408\u591a\u4e2a\u6761\u4ef6，\u8fd4\u56de\u5e03\u5c14 Series\u3002<\/p>\n\n\u8fd0\u7b97\u7b26\u8bf4\u660e<\/tr>\n\n\n\n\n
                          &<\/td>\n\u4e0e（and）<\/td>\n<\/tr>\n
                          |<\/td>\n\u6216（or）<\/td>\n<\/tr>\n
                          ~<\/td>\n\u975e（not）<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                          \u6ce8\u610f： \u5fc5\u987b\u7528 &\u3001|\u3001~，\u4e0d\u80fd\u7528 and\u3001or\u3001not；\u4e14\u6bcf\u4e2a\u6761\u4ef6\u5fc5\u987b\u52a0\u62ec\u53f7\u3002<\/p>\n

                          # \u903b\u8f91\u8fd0\u7b97<\/p>\n

                          import pandas as pd<\/p>\n

                          df = pd.DataFrame({
                          \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94", "\u8d75\u516d"],
                          \n "age": [25, 30, 28, 35],
                          \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac", "\u4e0a\u6d77"]
                          \n})<\/p>\n

                          # \u4e0e：&
                          \nresult = df[(df["age"] > 26) & (df["city"] == "\u4e0a\u6d77")]
                          \n# result = df[(df["age"] > 26) and (df["city"] == "\u4e0a\u6d77")] # ValueError
                          \nprint(f"age > 26 \u4e14 city == \u4e0a\u6d77：\\\\n{result}")<\/p>\n

                          # \u6216：|
                          \nresult = df[(df["age"] > 30) | (df["city"] == "\u676d\u5dde")]
                          \nprint(f"\\\\nage > 30 \u6216 city == \u676d\u5dde：\\\\n{result}")<\/p>\n

                          # \u975e：~
                          \nresult = df[~(df["city"] == "\u4e0a\u6d77")]
                          \nprint(f"\\\\ncity != \u4e0a\u6d77：\\\\n{result}") <\/p>\n

                          \u8bf4\u660e：<\/p>\n

                            \n
                          • \n

                            & \u4e0e\u3001| \u6216\u3001~ \u975e<\/p>\n<\/li>\n

                          • \n

                            \u6bcf\u4e2a\u6761\u4ef6\u5fc5\u987b\u52a0\u62ec\u53f7<\/p>\n<\/li>\n

                          • \n

                            \u4e0d\u80fd\u7528 and\u3001or\u3001not<\/p>\n<\/li>\n<\/ul>\n

                            2.4.2 \u7edf\u8ba1\u8fd0\u7b97<\/h4>\n

                            2.4.2.1 describe \/ mean \/ std \/ sum<\/h4>\n

                            Pandas \u63d0\u4f9b\u5e38\u7528\u7edf\u8ba1\u51fd\u6570，\u53ef\u5bf9\u6574\u8868\u6216\u6307\u5b9a\u5217\u8ba1\u7b97\u3002<\/p>\n

                            # \u7edf\u8ba1\u8fd0\u7b97<\/p>\n

                            import pandas as pd<\/p>\n

                            df = pd.DataFrame({
                            \n "age": [25, 30, 28, 35, 22],
                            \n "score": [85.5, 92.3, 78.8, 88.0, 95.5]
                            \n})<\/p>\n

                            # describe：\u7efc\u5408\u7edf\u8ba1
                            \nprint(f"describe：\\\\n{df.describe()}")<\/p>\n

                            # \u5355\u5217\u7edf\u8ba1
                            \nprint(f"\\\\nage \u5747\u503c：{df['age'].mean():.2f}")
                            \nprint(f"age \u6807\u51c6\u5dee：{df['age'].std():.2f}")
                            \nprint(f"age \u6c42\u548c：{df['age'].sum()}")<\/p>\n

                            # \u6309\u5217\u7edf\u8ba1
                            \nprint(f"\\\\n\u5404\u5217\u5747\u503c：\\\\n{df.mean()}") <\/p>\n

                            \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                            df.describe()<\/td>\n\u7efc\u5408\u7edf\u8ba1，\u8fd4\u56de\u8ba1\u6570\u3001\u5747\u503c\u3001\u6807\u51c6\u5dee\u3001\u5206\u4f4d\u6570<\/td>\n<\/tr>\n
                            df.mean()<\/td>\n\u5747\u503c<\/td>\n<\/tr>\n
                            df.std()<\/td>\n\u6807\u51c6\u5dee<\/td>\n<\/tr>\n
                            df.sum()<\/td>\n\u6c42\u548c<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                            \u8bf4\u660e：<\/p>\n

                              \n
                            • \n

                              describe \u4e00\u6b21\u6027\u8f93\u51fa\u591a\u4e2a\u7edf\u8ba1\u91cf<\/p>\n<\/li>\n

                            • \n

                              \u5355\u5217\u7edf\u8ba1\u7528 df["\u5217"].\u65b9\u6cd5()<\/p>\n<\/li>\n

                            • \n

                              \u6309\u5217\u7edf\u8ba1\u7528 df.\u65b9\u6cd5()<\/p>\n<\/li>\n<\/ul>\n

                              2.4.2.2 value_counts<\/h4>\n

                              value_counts() \u7528\u4e8e\u7edf\u8ba1\u6bcf\u4e2a\u503c\u51fa\u73b0\u7684\u6b21\u6570，\u5e38\u7528\u4e8e\u5206\u7c7b\u6570\u636e\u5206\u5e03\u5206\u6790\u3002<\/p>\n

                              # value_counts<\/p>\n

                              import pandas as pd<\/p>\n

                              df = pd.DataFrame({
                              \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac", "\u4e0a\u6d77", "\u676d\u5dde", "\u4e0a\u6d77"]
                              \n})<\/p>\n

                              # \u7edf\u8ba1\u9891\u6b21
                              \nprint(f"value_counts：\\\\n{df['city'].value_counts()}")<\/p>\n

                              # \u8f6c\u4e3a\u6bd4\u4f8b
                              \nprint(f"\\\\n\u6bd4\u4f8b：\\\\n{df['city'].value_counts(normalize=True)}") <\/p>\n

                              \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n
                              Series.value_counts()<\/td>\n\u7edf\u8ba1\u6bcf\u4e2a\u503c\u7684\u51fa\u73b0\u6b21\u6570\u3002normalize=True \u8fd4\u56de\u6bd4\u4f8b<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                              \u8bf4\u660e：<\/p>\n

                                \n
                              • \n

                                \u9ed8\u8ba4\u6309\u9891\u6b21\u964d\u5e8f<\/p>\n<\/li>\n

                              • \n

                                normalize=True \u8fd4\u56de\u6bd4\u4f8b<\/p>\n<\/li>\n

                              • \n

                                \u5e38\u7528\u4e8e\u5206\u7c7b\u6570\u636e\u5206\u5e03\u5206\u6790<\/p>\n<\/li>\n<\/ul>\n

                                2.5 \u6570\u636e\u5904\u7406<\/h3>\n

                                2.5.1 \u53bb\u91cd drop_duplicates<\/h4>\n

                                drop_duplicates() \u7528\u4e8e\u53bb\u9664\u91cd\u590d\u884c，\u53ef\u6307\u5b9a\u5217\u6216\u5168\u90e8\u5217\u3002<\/p>\n

                                # \u53bb\u91cd<\/p>\n

                                import pandas as pd<\/p>\n

                                df = pd.DataFrame({
                                \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u5f20\u4e09", "\u738b\u4e94", "\u674e\u56db"],
                                \n "city": ["\u676d\u5dde", "\u5317\u4eac", "\u676d\u5dde", "\u5317\u4eac", "\u4e0a\u6d77"]
                                \n})<\/p>\n

                                # \u5168\u90e8\u5217\u53bb\u91cd
                                \nprint(f"\u539f\u59cb\u6570\u636e：\\\\n{df}")
                                \nprint(f"\\\\n\u5168\u90e8\u5217\u53bb\u91cd：\\\\n{df.drop_duplicates()}") # \u884cSeries\u5bf9\u8c61\u5b8c\u5168\u76f8\u540c\u624d\u4f1a\u53bb\u9664<\/p>\n

                                # \u6309\u6307\u5b9a\u5217\u53bb\u91cd（\u4fdd\u7559\u7b2c\u4e00\u6761）
                                \nprint(f"\\\\n\u6309 name \u53bb\u91cd：\\\\n{df.drop_duplicates(subset=['name'])}") # \u6307\u5b9a\u5217\u5bf9\u5e94\u5143\u7d20\u76f8\u540c\u5c31\u53bb\u9664<\/p>\n

                                # \u6309\u6307\u5b9a\u5217\u53bb\u91cd（\u4fdd\u7559\u6700\u540e\u4e00\u6761）
                                \nprint(f"\\\\n\u6309 name \u53bb\u91cd（\u4fdd\u7559\u6700\u540e）：\\\\n{df.drop_duplicates(subset=['name'], keep='last')}") <\/p>\n

                                \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                subset<\/td>\n\u6307\u5b9a\u5217，\u9ed8\u8ba4\u5168\u90e8\u5217<\/td>\n<\/tr>\n
                                keep<\/td>\n'first'（\u9ed8\u8ba4）\u4fdd\u7559\u7b2c\u4e00\u6761，'last' \u4fdd\u7559\u6700\u540e\u4e00\u6761，False \u5168\u90e8\u5220\u9664<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                \u8bf4\u660e：<\/p>\n

                                  \n
                                • \n

                                  \u9ed8\u8ba4\u5168\u90e8\u5217\u53bb\u91cd<\/p>\n<\/li>\n

                                • \n

                                  subset \u6307\u5b9a\u53bb\u91cd\u5217<\/p>\n<\/li>\n

                                • \n

                                  keep \u63a7\u5236\u4fdd\u7559\u54ea\u6761<\/p>\n<\/li>\n<\/ul>\n

                                  2.5.2 \u6dfb\u52a0 \/ \u5220\u9664\u5217<\/h4>\n

                                  # \u6dfb\u52a0 \/ \u5220\u9664\u5217<\/p>\n

                                  import pandas as pd<\/p>\n

                                  df = pd.DataFrame({
                                  \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
                                  \n "age": [25, 30, 28]
                                  \n})<\/p>\n

                                  # assign：\u65b0\u589e\u5217
                                  \ndf1 = df.assign(age_plus_10=df["age"] + 10)
                                  \nprint(f"assign \u65b0\u589e\u5217：\\\\n{df1}")<\/p>\n

                                  # \u94fe\u5f0f assign
                                  \ndf2 = df.assign(
                                  \n age_plus_10=df["age"] + 10,
                                  \n is_adult=df["age"] >= 18
                                  \n)
                                  \nprint(f"\\\\n\u94fe\u5f0f assign：\\\\n{df2}")<\/p>\n

                                  # drop：\u5220\u9664\u5217
                                  \ndf3 = df.drop(columns=["age"])
                                  \nprint(f"\\\\ndrop \u5220\u9664\u5217：\\\\n{df3}")<\/p>\n

                                  # drop：\u5220\u9664\u884c
                                  \ndf4 = df.drop(index=[0, 2])
                                  \nprint(f"\\\\ndrop \u5220\u9664\u884c：\\\\n{df4}") <\/p>\n

                                  \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n
                                  df.assign()<\/td>\n\u65b0\u589e\u5217\u3002\u53c2\u6570\u4e3a \u5217\u540d=\u503c，\u8fd4\u56de\u65b0 DataFrame<\/td>\n<\/tr>\n
                                  df.drop()<\/td>\n\u5220\u9664\u5217\u6216\u884c\u3002columns \u6307\u5b9a\u5217，index \u6307\u5b9a\u884c<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                  \u8bf4\u660e：<\/p>\n

                                    \n
                                  • \n

                                    assign \u8fd4\u56de\u65b0 DataFrame，\u53ef\u94fe\u5f0f\u8c03\u7528<\/p>\n<\/li>\n

                                  • \n

                                    drop \u8fd4\u56de\u65b0 DataFrame<\/p>\n<\/li>\n<\/ul>\n

                                    2.5.3 \u7f3a\u5931\u503c\u5904\u7406<\/h4>\n

                                    # \u7f3a\u5931\u503c\u5904\u7406<\/p>\n

                                    import pandas as pd
                                    \nimport numpy as np<\/p>\n

                                    df = pd.DataFrame({
                                    \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94", "\u8d75\u516d"],
                                    \n "age": [25, np.nan, 28, 35],
                                    \n "score": [85, 92, np.nan, 88]
                                    \n})<\/p>\n

                                    print(f"\u539f\u59cb\u6570\u636e：\\\\n{df}")<\/p>\n

                                    # isnull：\u5224\u65ad\u7f3a\u5931
                                    \nprint(f"\\\\n\u662f\u5426\u7f3a\u5931：\\\\n{df.isnull()}")<\/p>\n

                                    # \u7edf\u8ba1\u7f3a\u5931\u6570
                                    \nprint(f"\\\\n\u7f3a\u5931\u6570：\\\\n{df.isnull().sum()}")<\/p>\n

                                    # dropna：\u5220\u9664\u7f3a\u5931\u884c
                                    \nprint(f"\\\\n\u5220\u9664\u7f3a\u5931\u884c：\\\\n{df.dropna()}")<\/p>\n

                                    # fillna：\u586b\u5145\u7f3a\u5931\u503c
                                    \nprint(f"\\\\n\u586b\u5145 0：\\\\n{df.fillna(0)}")<\/p>\n

                                    # \u7528\u5747\u503c\u586b\u5145
                                    \nprint(f"\\\\n\u7528\u5747\u503c\u586b\u5145：\\\\n{df.fillna(df.mean(numeric_only=True))}")<\/p>\n

                                    print(f"\u539f\u59cb\u6570\u636e：\\\\n{df}") # \u4e0d\u4f1a\u4fee\u6539\u539f\u6570\u636e <\/p>\n

                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n
                                    df.isnull()<\/td>\n\u5224\u65ad\u662f\u5426\u7f3a\u5931，\u8fd4\u56de\u5e03\u5c14 DataFrame<\/td>\n<\/tr>\n
                                    df.dropna()<\/td>\n\u5220\u9664\u7f3a\u5931\u503c\u3002axis \u6307\u5b9a\u8f74，how \u6307\u5b9a\u6761\u4ef6<\/td>\n<\/tr>\n
                                    df.fillna()<\/td>\n\u586b\u5145\u7f3a\u5931\u503c\u3002\u53c2\u6570\u4e3a\u586b\u5145\u503c<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                    \u8bf4\u660e：<\/p>\n

                                      \n
                                    • \n

                                      isnull() \u8fd4\u56de\u5e03\u5c14 DataFrame<\/p>\n<\/li>\n

                                    • \n

                                      dropna() \u5220\u9664\u542b\u7f3a\u5931\u503c\u7684\u884c<\/p>\n<\/li>\n

                                    • \n

                                      fillna() \u586b\u5145\u7f3a\u5931\u503c<\/p>\n<\/li>\n

                                    • \n

                                      \u4e0d\u4f1a\u4fee\u6539\u539f\u6570\u636e，\u8fd4\u56de\u65b0\u7684DataFrame\u5bf9\u8c61<\/p>\n<\/li>\n<\/ul>\n

                                      2.5.4 \u6570\u636e\u5408\u5e76<\/h4>\n

                                      2.5.4.1 merge（\u952e\u8fde\u63a5）<\/h4>\n

                                      merge \u7528\u4e8e\u6309\u952e\u8fde\u63a5\u4e24\u4e2a DataFrame，\u7c7b\u4f3c SQL \u7684 JOIN\u3002<\/p>\n

                                      # merge：\u952e\u8fde\u63a5<\/p>\n

                                      import pandas as pd<\/p>\n

                                      df1 = pd.DataFrame({
                                      \n "id": [1, 2, 3],
                                      \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"]
                                      \n})<\/p>\n

                                      df2 = pd.DataFrame({
                                      \n "id": [2, 3, 4],
                                      \n "score": [85, 92, 78]
                                      \n})<\/p>\n

                                      # \u5185\u8fde\u63a5：\u53ea\u4fdd\u7559\u4e24\u8868\u90fd\u6709\u7684\u952e
                                      \ninner = pd.merge(df1, df2, on="id", how="inner")
                                      \nprint(f"\u5185\u8fde\u63a5：\\\\n{inner}")<\/p>\n

                                      # \u5de6\u8fde\u63a5：\u4fdd\u7559\u5de6\u8868\u6240\u6709\u952e
                                      \nleft = pd.merge(df1, df2, on="id", how="left")
                                      \nprint(f"\\\\n\u5de6\u8fde\u63a5：\\\\n{left}")<\/p>\n

                                      # \u5916\u8fde\u63a5：\u4fdd\u7559\u6240\u6709\u952e
                                      \nouter = pd.merge(df1, df2, on="id", how="outer")
                                      \nprint(f"\\\\n\u5916\u8fde\u63a5：\\\\n{outer}") <\/p>\n

                                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                      on<\/td>\n\u8fde\u63a5\u952e，\u5217\u540d<\/td>\n<\/tr>\n
                                      how<\/td>\n\u8fde\u63a5\u65b9\u5f0f：inner\u3001left\u3001right\u3001outer<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                      \u8bf4\u660e：<\/p>\n

                                        \n
                                      • \n

                                        inner：\u53ea\u4fdd\u7559\u4e24\u8868\u90fd\u6709\u7684\u952e<\/p>\n<\/li>\n

                                      • \n

                                        left：\u4fdd\u7559\u5de6\u8868\u6240\u6709\u952e<\/p>\n<\/li>\n

                                      • \n

                                        outer：\u4fdd\u7559\u6240\u6709\u952e<\/p>\n<\/li>\n<\/ul>\n

                                        2.5.4.2 concat（\u5806\u53e0）<\/h5>\n

                                        concat \u7528\u4e8e\u5806\u53e0\u591a\u4e2a DataFrame，\u6cbf\u884c\u6216\u5217\u65b9\u5411\u3002<\/p>\n

                                        # concat：\u5806\u53e0<\/p>\n

                                        import pandas as pd<\/p>\n

                                        df1 = pd.DataFrame({
                                        \n "name": ["\u5f20\u4e09", "\u674e\u56db"],
                                        \n "age": [25, 30]
                                        \n})<\/p>\n

                                        df2 = pd.DataFrame({
                                        \n "name": ["\u738b\u4e94", "\u8d75\u516d"],
                                        \n "age": [28, 35]
                                        \n})<\/p>\n

                                        # \u6cbf\u884c\u5806\u53e0（\u4e0a\u4e0b）
                                        \nrow_concat = pd.concat([df1, df2], axis=0, ignore_index=True)
                                        \nprint(f"\u6cbf\u884c\u5806\u53e0：\\\\n{row_concat}")<\/p>\n

                                        # \u6cbf\u5217\u5806\u53e0（\u5de6\u53f3）
                                        \ncol_concat = pd.concat([df1, df2], axis=1)
                                        \nprint(f"\\\\n\u6cbf\u5217\u5806\u53e0：\\\\n{col_concat}") <\/p>\n

                                        \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                        axis<\/td>\n0 \u6cbf\u884c\u5806\u53e0，1 \u6cbf\u5217\u5806\u53e0<\/td>\n<\/tr>\n
                                        ignore_index<\/td>\n\u662f\u5426\u91cd\u7f6e\u7d22\u5f15<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                        \u8bf4\u660e：<\/p>\n

                                          \n
                                        • \n

                                          axis=0 \u4e0a\u4e0b\u5806\u53e0<\/p>\n<\/li>\n

                                        • \n

                                          axis=1 \u5de6\u53f3\u5806\u53e0<\/p>\n<\/li>\n

                                        • \n

                                          ignore_index=True \u91cd\u7f6e\u7d22\u5f15<\/p>\n<\/li>\n<\/ul>\n

                                          2.5.5 \u6570\u636e\u5206\u7ec4\u4e0e\u805a\u5408<\/h4>\n

                                          2.5.5.1 groupby \u57fa\u7840<\/h5>\n

                                          groupby \u662f Pandas \u6700\u5f3a\u5927\u7684\u529f\u80fd\u4e4b\u4e00，\u7528\u4e8e\u6309\u67d0\u4e2a\u952e\u5206\u7ec4，\u7136\u540e\u5bf9\u6bcf\u7ec4\u505a\u805a\u5408\u8fd0\u7b97\u3002<\/p>\n

                                          \u6838\u5fc3\u6d41\u7a0b： \u62c6\u5206（split）\u2192 \u5e94\u7528（apply）\u2192 \u5408\u5e76（combine）<\/p>\n

                                          # groupby \u57fa\u7840<\/p>\n

                                          import pandas as pd<\/p>\n

                                          df = pd.DataFrame({
                                          \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac", "\u676d\u5dde"],
                                          \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94", "\u8d75\u516d", "\u94b1\u4e03", "\u5b59\u516b"],
                                          \n "score": [85, 92, 78, 88, 95, 90]
                                          \n})<\/p>\n

                                          # \u6309 city \u5206\u7ec4，\u6c42 score \u5747\u503c
                                          \ngrouped = df.groupby("city")["score"].mean()
                                          \nprint(f"\u6309 city \u5206\u7ec4\u6c42\u5747\u503c：\\\\n{grouped}")<\/p>\n

                                          # \u6309 city \u5206\u7ec4，\u6c42\u591a\u4e2a\u7edf\u8ba1\u91cf
                                          \nprint(f"\\\\n\u6309 city \u5206\u7ec4，\u591a\u4e2a\u7edf\u8ba1\u91cf：\\\\n{df.groupby('city')['score'].agg(['mean', 'max', 'min'])}")<\/p>\n

                                          # \u6309 city \u5206\u7ec4，\u7edf\u8ba1\u4eba\u6570
                                          \nprint(f"\\\\n\u6309 city \u5206\u7ec4，\u7edf\u8ba1\u4eba\u6570：\\\\n{df.groupby('city').size()}") <\/p>\n

                                          \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n
                                          df.groupby("\u5217")<\/td>\n\u6309\u5217\u5206\u7ec4，\u8fd4\u56de GroupBy \u5bf9\u8c61<\/td>\n<\/tr>\n
                                          .agg([…])<\/td>\n\u591a\u4e2a\u805a\u5408\u51fd\u6570<\/td>\n<\/tr>\n
                                          .size()<\/td>\n\u7edf\u8ba1\u6bcf\u7ec4\u6570\u91cf<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                          \u8bf4\u660e：<\/p>\n

                                            \n
                                          • \n

                                            groupby \u8fd4\u56de GroupBy \u5bf9\u8c61，\u9700\u63a5\u805a\u5408\u51fd\u6570<\/p>\n<\/li>\n

                                          • \n

                                            agg \u53ef\u4e00\u6b21\u8ba1\u7b97\u591a\u4e2a\u7edf\u8ba1\u91cf<\/p>\n<\/li>\n

                                          • \n

                                            size \u7edf\u8ba1\u6bcf\u7ec4\u884c\u6570<\/p>\n<\/li>\n<\/ul>\n

                                            2.5.5.2 agg \u591a\u805a\u5408\u00a0<\/h5>\n

                                            agg \u7528\u4e8e\u5bf9\u5206\u7ec4\u6570\u636e\u540c\u65f6\u5e94\u7528\u591a\u4e2a\u805a\u5408\u51fd\u6570\u3002<\/p>\n

                                            # agg \u591a\u805a\u5408<\/p>\n

                                            import pandas as pd<\/p>\n

                                            df = pd.DataFrame({
                                            \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"],
                                            \n "score": [85, 92, 78, 88, 95],
                                            \n "age": [25, 30, 28, 35, 22]
                                            \n})<\/p>\n

                                            # \u591a\u4e2a\u805a\u5408\u51fd\u6570
                                            \nresult = df.groupby("city").agg({
                                            \n "score": ["mean", "max", "min"],
                                            \n "age": "mean"
                                            \n})
                                            \nprint(f"\u591a\u805a\u5408：\\\\n{result}")<\/p>\n

                                            # \u547d\u540d\u805a\u5408（\u63a8\u8350）
                                            \nresult2 = df.groupby("city").agg(
                                            \n score_mean=("score", "mean"),
                                            \n score_max=("score", "max"),
                                            \n age_mean=("age", "mean")
                                            \n)
                                            \nprint(f"\\\\n\u547d\u540d\u805a\u5408：\\\\n{result2}") <\/p>\n

                                            \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                            agg(\u5b57\u5178)<\/td>\n\u5bf9\u4e0d\u540c\u5217\u5e94\u7528\u4e0d\u540c\u805a\u5408\u51fd\u6570<\/td>\n<\/tr>\n
                                            agg(\u547d\u540d\u5143\u7ec4)<\/td>\n\u547d\u540d\u805a\u5408，\u63a8\u8350<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                            \u8bf4\u660e：<\/p>\n

                                              \n
                                            • \n

                                              \u5b57\u5178\u5f62\u5f0f：{"\u5217": ["\u51fd\u65701", "\u51fd\u65702"]}<\/p>\n<\/li>\n

                                            • \n

                                              \u547d\u540d\u805a\u5408：\u65b0\u5217\u540d=("\u5217", "\u51fd\u6570")<\/p>\n<\/li>\n

                                            • \n

                                              \u547d\u540d\u805a\u5408\u66f4\u6e05\u6670<\/p>\n<\/li>\n<\/ul>\n

                                              2.5.5.3 reset_index\u00a0<\/h5>\n

                                              groupby \u7684\u7ed3\u679c\u7d22\u5f15\u662f\u5206\u7ec4\u952e，\u7528 reset_index() \u53ef\u5c06\u5206\u7ec4\u952e\u8f6c\u4e3a\u5217\u3002<\/p>\n

                                              # reset_index<\/p>\n

                                              import pandas as pd<\/p>\n

                                              df = pd.DataFrame({
                                              \n "city": ["\u676d\u5dde", "\u4e0a\u6d77", "\u676d\u5dde", "\u4e0a\u6d77", "\u5317\u4eac"],
                                              \n "score": [85, 92, 78, 88, 95]
                                              \n})<\/p>\n

                                              # groupby \u7ed3\u679c：\u7d22\u5f15\u662f city
                                              \ngrouped = df.groupby("city")["score"].mean()
                                              \nprint(f"\u672a reset_index：\\\\n{grouped}")
                                              \nprint(f"\u7d22\u5f15：{grouped.index.tolist()}")<\/p>\n

                                              # reset_index：\u7d22\u5f15\u8f6c\u4e3a\u5217
                                              \ngrouped_reset = grouped.reset_index()
                                              \nprint(f"\\\\nreset_index \u540e：\\\\n{grouped_reset}")
                                              \nprint(f"\u5217\u540d：{grouped_reset.columns.tolist()}") <\/p>\n

                                              \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n
                                              .reset_index()<\/td>\n\u5c06\u7d22\u5f15\u8f6c\u4e3a\u5217，\u8fd4\u56de\u65b0 DataFrame<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                              \u8bf4\u660e：<\/p>\n

                                                \n
                                              • \n

                                                groupby \u7ed3\u679c\u7d22\u5f15\u662f\u5206\u7ec4\u952e<\/p>\n<\/li>\n

                                              • \n

                                                reset_index() \u5c06\u7d22\u5f15\u8f6c\u4e3a\u5217<\/p>\n<\/li>\n

                                              • \n

                                                \u65b9\u4fbf\u540e\u7eed\u5904\u7406<\/p>\n<\/li>\n<\/ul>\n

                                                2.5.6 \u6587\u672c\u5904\u7406<\/h4>\n

                                                2.5.6.1 str \u8bbf\u95ee\u5668<\/h5>\n

                                                str \u8bbf\u95ee\u5668\u7528\u4e8e\u5bf9\u5b57\u7b26\u4e32\u5217\u8fdb\u884c\u5411\u91cf\u5316\u64cd\u4f5c，\u5982\u5927\u5c0f\u5199\u3001\u66ff\u6362\u3001\u5207\u5206\u3002<\/p>\n

                                                # str \u8bbf\u95ee\u5668<\/p>\n

                                                import pandas as pd<\/p>\n

                                                df = pd.DataFrame({
                                                \n "name": ["\u5f20\u4e09", "\u674e\u56db", "\u738b\u4e94"],
                                                \n "email": ["ZhangSan@example.com", "LiSi@test.com", "WangWu@demo.com"]
                                                \n})<\/p>\n

                                                # \u5927\u5199
                                                \nprint(f"\u5927\u5199：\\\\n{df['email'].str.upper()}")<\/p>\n

                                                # \u5c0f\u5199
                                                \nprint(f"\\\\n\u5c0f\u5199：\\\\n{df['email'].str.lower()}")<\/p>\n

                                                # \u5305\u542b\u5224\u65ad
                                                \nprint(f"\\\\n\u5305\u542b example：\\\\n{df['email'].str.contains('example')}")<\/p>\n

                                                # \u66ff\u6362
                                                \nprint(f"\\\\n\u66ff\u6362：\\\\n{df['email'].str.replace('.com', '.cn')}")<\/p>\n

                                                # \u5207\u5206
                                                \nprint(f"\\\\n\u5207\u5206：\\\\n{df['email'].str.split('@')}") <\/p>\n

                                                \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n\n
                                                .str.upper()<\/td>\n\u8f6c\u5927\u5199<\/td>\n<\/tr>\n
                                                .str.lower()<\/td>\n\u8f6c\u5c0f\u5199<\/td>\n<\/tr>\n
                                                .str.contains()<\/td>\n\u5305\u542b\u5224\u65ad<\/td>\n<\/tr>\n
                                                .str.replace()<\/td>\n\u66ff\u6362<\/td>\n<\/tr>\n
                                                .str.split()<\/td>\n\u5207\u5206<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                \u8bf4\u660e：<\/p>\n

                                                  \n
                                                • \n

                                                  str \u8bbf\u95ee\u5668\u7528\u4e8e\u5b57\u7b26\u4e32\u5217<\/p>\n<\/li>\n

                                                • \n

                                                  \u5411\u91cf\u5316\u64cd\u4f5c，\u6bd4 apply \u5feb<\/p>\n<\/li>\n<\/ul>\n

                                                  2.5.6.2 apply \/ map \/ applymap \u533a\u5206<\/h5>\n\n\u65b9\u6cd5\u4f5c\u7528\u5bf9\u8c61\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                  apply<\/td>\n\u884c\/\u5217<\/td>\n\u5bf9\u6bcf\u884c\u6216\u6bcf\u5217\u5e94\u7528\u51fd\u6570<\/td>\n<\/tr>\n
                                                  map<\/td>\nSeries<\/td>\n\u5bf9\u6bcf\u4e2a\u5143\u7d20\u5e94\u7528\u51fd\u6570<\/td>\n<\/tr>\n
                                                  applymap<\/td>\nDataFrame<\/td>\n\u5bf9\u6bcf\u4e2a\u5143\u7d20\u5e94\u7528\u51fd\u6570<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                  # apply \/ map \/ applymap<\/p>\n

                                                  import pandas as pd<\/p>\n

                                                  df = pd.DataFrame({
                                                  \n "name": ["\u5f20\u4e09", "\u674e\u56db"],
                                                  \n "score": [85, 92]
                                                  \n})<\/p>\n

                                                  # apply：\u5bf9\u5217\u5e94\u7528
                                                  \nprint(f"apply \u5bf9\u5217：\\\\n{df[['score']].apply(lambda x: x.max())}")<\/p>\n

                                                  # map：\u5bf9 Series \u6bcf\u4e2a\u5143\u7d20
                                                  \nprint(f"\\\\nmap：\\\\n{df['name'].map(lambda x: f'\u5b66\u751f-{x}')}")<\/p>\n

                                                  # applymap：\u5bf9 DataFrame \u6bcf\u4e2a\u5143\u7d20
                                                  \nprint(f"\\\\napplymap：\\\\n{df.applymap(lambda x: str(x)+"test")}") <\/p>\n

                                                  \u8bf4\u660e：<\/p>\n

                                                    \n
                                                  • \n

                                                    apply \u4f5c\u7528\u4e8e\u884c\/\u5217<\/p>\n<\/li>\n

                                                  • \n

                                                    map \u4f5c\u7528\u4e8e Series \u5143\u7d20<\/p>\n<\/li>\n

                                                  • \n

                                                    applymap \u4f5c\u7528\u4e8e DataFrame \u5143\u7d20<\/p>\n<\/li>\n<\/ul>\n

                                                    2.5.6.3 \u62c6\u5206\u5217 split \/ str.split<\/h5>\n

                                                    str.split() \u7528\u4e8e\u62c6\u5206\u5b57\u7b26\u4e32\u5217，\u53ef\u914d\u5408 expand=True \u62c6\u4e3a\u591a\u5217\u3002<\/p>\n

                                                    # \u62c6\u5206\u5217<\/p>\n

                                                    import pandas as pd<\/p>\n

                                                    df = pd.DataFrame({
                                                    \n "info": ["\u5f20\u4e09-25-\u676d\u5dde", "\u674e\u56db-30-\u4e0a\u6d77", "\u738b\u4e94-28-\u5317\u4eac"]
                                                    \n})<\/p>\n

                                                    # \u62c6\u5206\u4e3a\u591a\u5217
                                                    \nsplit_df = df["info"].str.split("-", expand=True)
                                                    \nsplit_df.columns = ["name", "age", "city"]
                                                    \nprint(f"\u62c6\u5206\u540e：\\\\n{split_df}")<\/p>\n

                                                    # \u5408\u5e76\u5230\u539f DataFrame
                                                    \ndf = pd.concat([df, split_df], axis=1)
                                                    \nprint(f"\\\\n\u5408\u5e76\u540e：\\\\n{df}") <\/p>\n

                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                                    str.split(sep)<\/td>\n\u6309\u5206\u9694\u7b26\u62c6\u5206<\/td>\n<\/tr>\n
                                                    expand=True<\/td>\n\u62c6\u4e3a\u591a\u5217<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                    \u8bf4\u660e： expand=True \u62c6\u4e3a\u591a\u5217，\u5426\u5219\u8fd4\u56de\u5217\u8868\u3002<\/p>\n

                                                    2.5.7 \u968f\u673a\u91c7\u6837<\/h4>\n

                                                    sample() \u7528\u4e8e\u968f\u673a\u91c7\u6837，\u5e38\u7528\u4e8e\u6570\u636e\u96c6\u5212\u5206\u3001\u62bd\u6837\u5206\u6790\u3002<\/p>\n

                                                    # \u968f\u673a\u91c7\u6837<\/p>\n

                                                    import pandas as pd<\/p>\n

                                                    df = pd.DataFrame({
                                                    \n "id": range(1, 11),
                                                    \n "score": [85, 92, 78, 88, 95, 90, 82, 87, 91, 86]
                                                    \n})<\/p>\n

                                                    # \u968f\u673a\u53d6 3 \u6761
                                                    \nprint(f"\u968f\u673a\u53d6 3 \u6761：\\\\n{df.sample(n=3, random_state=42)}")<\/p>\n

                                                    # \u968f\u673a\u53d6 30%
                                                    \nprint(f"\\\\n\u968f\u673a\u53d6 30%：\\\\n{df.sample(frac=0.3, random_state=42)}") <\/p>\n

                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                    n<\/td>\n\u91c7\u6837\u6570\u91cf<\/td>\n<\/tr>\n
                                                    frac<\/td>\n\u91c7\u6837\u6bd4\u4f8b<\/td>\n<\/tr>\n
                                                    random_state<\/td>\n\u968f\u673a\u79cd\u5b50，\u4fdd\u8bc1\u53ef\u590d\u73b0<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                    \u8bf4\u660e： random_state \u4fdd\u8bc1\u7ed3\u679c\u53ef\u590d\u73b0\u3002<\/p>\n

                                                    2.5.8 \u65f6\u95f4\u5e8f\u5217\u57fa\u7840<\/h4>\n

                                                    to_datetime() \u7528\u4e8e\u5c06\u5b57\u7b26\u4e32\u8f6c\u4e3a\u65f6\u95f4\u7c7b\u578b，dt \u8bbf\u95ee\u5668\u7528\u4e8e\u63d0\u53d6\u65f6\u95f4\u5c5e\u6027\u3002<\/p>\n

                                                    # \u65f6\u95f4\u5e8f\u5217\u57fa\u7840<\/p>\n

                                                    import pandas as pd<\/p>\n

                                                    df = pd.DataFrame({
                                                    \n "date": ["2024-01-15", "2024-02-20", "2024-03-25"],
                                                    \n "score": [85, 92, 78]
                                                    \n})<\/p>\n

                                                    # \u8f6c\u4e3a\u65f6\u95f4\u7c7b\u578b
                                                    \ndf["date"] = pd.to_datetime(df["date"])
                                                    \nprint(f"\u8f6c\u6362\u540e\u7c7b\u578b：{df['date'].dtype}")<\/p>\n

                                                    # dt \u8bbf\u95ee\u5668
                                                    \ndf["year"] = df["date"].dt.year
                                                    \ndf["month"] = df["date"].dt.month
                                                    \ndf["day"] = df["date"].dt.day
                                                    \nprint(f"\\\\n\u63d0\u53d6\u5e74\u6708\u65e5：\\\\n{df}") <\/p>\n

                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                                                    pd.to_datetime()<\/td>\n\u8f6c\u4e3a\u65f6\u95f4\u7c7b\u578b<\/td>\n<\/tr>\n
                                                    .dt.year<\/td>\n\u63d0\u53d6\u5e74\u4efd<\/td>\n<\/tr>\n
                                                    .dt.month<\/td>\n\u63d0\u53d6\u6708\u4efd<\/td>\n<\/tr>\n
                                                    .dt.day<\/td>\n\u63d0\u53d6\u65e5<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                    2.5.9 \u900f\u89c6\u8868\u4e0e\u4ea4\u53c9\u8868（\u4e86\u89e3）<\/h4>\n\n\u6982\u5ff5\u8bf4\u660e\u7c7b\u6bd4<\/tr>\n\n\n\n
                                                    \u900f\u89c6\u8868（pivot_table）<\/td>\n\u5bf9\u6570\u636e\u8fdb\u884c\u591a\u7ef4\u6c47\u603b，\u53ef\u6307\u5b9a\u503c\u5217\u3001\u884c\u3001\u5217\u548c\u805a\u5408\u51fd\u6570<\/td>\n\u7c7b\u4f3c Excel \u6570\u636e\u900f\u89c6\u8868<\/td>\n<\/tr>\n
                                                    \u4ea4\u53c9\u8868（crosstab）<\/td>\n\u7edf\u8ba1\u4e24\u4e2a\u5206\u7c7b\u53d8\u91cf\u7684\u9891\u6b21，\u662f\u900f\u89c6\u8868\u7684\u7279\u4f8b<\/td>\n\u7c7b\u4f3c\u5217\u8054\u8868<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                    \"\"<\/p>\n

                                                    \"\"<\/p>\n

                                                    \"\"<\/p>\n<\/p>\n

                                                    2.6 \u7efc\u5408\u5b9e\u6218\u6848\u4f8b<\/h3>\n

                                                    2.6.1 \u6a21\u578b\u8f93\u51fa\u89e3\u6790（JSON \u2192 DataFrame）<\/h4>\n

                                                    \u8c03\u7528\u672c\u5730 Ollama \u6a21\u578b，\u83b7\u53d6 JSON \u8f93\u51fa，\u89e3\u6790\u4e3a DataFrame\u3002\u8fd9\u662f\u5927\u6a21\u578b\u5e94\u7528\u5f00\u53d1\u7684\u7b2c\u4e00\u6b65\u3002<\/p>\n

                                                    # \u6a21\u578b\u8f93\u51fa\u89e3\u6790：JSON \u2192 DataFrame<\/p>\n

                                                    import httpx
                                                    \nimport json
                                                    \nimport pandas as pd<\/p>\n

                                                    # \u8c03\u7528 Ollama
                                                    \ndef call_ollama(prompt):
                                                    \n """\u8c03\u7528\u672c\u5730 Ollama \u6a21\u578b"""
                                                    \n response = httpx.post(
                                                    \n "http:\/\/localhost:11434\/v1\/chat\/completions",
                                                    \n headers={"Authorization": "Bearer ollama"},
                                                    \n json={
                                                    \n "model": "qwen2.5:1.5b",
                                                    \n "messages": [{"role": "user", "content": prompt}],
                                                    \n "stream": False
                                                    \n },
                                                    \n timeout=60
                                                    \n )
                                                    \n return response.json()["choices"][0]["message"]["content"]<\/p>\n

                                                    # \u6279\u91cf\u8c03\u7528
                                                    \nprompts = [
                                                    \n "\u7528\u4e00\u53e5\u8bdd\u4ecb\u7ecd Python",
                                                    \n "\u7528\u4e00\u53e5\u8bdd\u4ecb\u7ecd Java",
                                                    \n "\u7528\u4e00\u53e5\u8bdd\u4ecb\u7ecd Go",
                                                    \n]<\/p>\n

                                                    results = []
                                                    \nfor i, prompt in enumerate(prompts):
                                                    \n output = call_ollama(prompt)
                                                    \n results.append({
                                                    \n "id": i + 1,
                                                    \n "prompt": prompt,
                                                    \n "output": output,
                                                    \n "length": len(output)
                                                    \n })<\/p>\n

                                                    # \u8f6c\u4e3a DataFrame
                                                    \ndf = pd.DataFrame(results)
                                                    \nprint(f"\u6a21\u578b\u8f93\u51fa：\\\\n{df}")<\/p>\n

                                                    # \u4fdd\u5b58
                                                    \ndf.to_csv("model_outputs.csv", index=False, encoding="utf-8")
                                                    \nprint("\\\\n\u5df2\u4fdd\u5b58 model_outputs.csv") <\/p>\n

                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                                                    httpx.post()<\/td>\n\u53d1\u9001 POST \u8bf7\u6c42\u3002json \u53c2\u6570\u4f20\u9012\u8bf7\u6c42\u4f53<\/td>\n<\/tr>\n
                                                    response.json()<\/td>\n\u89e3\u6790 JSON \u54cd\u5e94<\/td>\n<\/tr>\n
                                                    pd.DataFrame()<\/td>\n\u4ece\u5217\u8868\u7684\u5b57\u5178\u521b\u5efa DataFrame<\/td>\n<\/tr>\n
                                                    df.to_csv()<\/td>\n\u4fdd\u5b58\u4e3a CSV\u3002index=False \u4e0d\u5199\u7d22\u5f15<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                    \u8bf4\u660e：<\/p>\n

                                                      \n
                                                    • \n

                                                      \u8c03\u7528 Ollama \u83b7\u53d6\u6a21\u578b\u8f93\u51fa<\/p>\n<\/li>\n

                                                    • \n

                                                      \u6279\u91cf\u8c03\u7528，\u6536\u96c6\u7ed3\u679c<\/p>\n<\/li>\n

                                                    • \n

                                                      \u8f6c\u4e3a DataFrame，\u4fdd\u5b58<\/p>\n<\/li>\n<\/ul>\n

                                                      2.6.2 \u8bc4\u6d4b\u96c6\u7edf\u8ba1（\u51c6\u786e\u7387\u3001badcase）<\/h4>\n

                                                      \u6784\u5efa\u8bc4\u6d4b\u96c6，\u8c03\u7528\u6a21\u578b，\u7edf\u8ba1\u51c6\u786e\u7387，\u7b5b\u9009 badcase\u3002\u8fd9\u662f\u5927\u6a21\u578b\u8bc4\u6d4b\u7684\u6838\u5fc3\u6d41\u7a0b\u3002<\/p>\n

                                                      # \u8bc4\u6d4b\u96c6\u7edf\u8ba1<\/p>\n

                                                      import httpx
                                                      \nimport json
                                                      \nimport pandas as pd<\/p>\n

                                                      def call_ollama(prompt):
                                                      \n """\u8c03\u7528\u672c\u5730 Ollama \u6a21\u578b"""
                                                      \n response = httpx.post(
                                                      \n "http:\/\/localhost:11434\/v1\/chat\/completions",
                                                      \n headers={"Authorization": "Bearer ollama"},
                                                      \n json={
                                                      \n "model": "qwen2.5:1.5b",
                                                      \n "messages": [
                                                      \n {"role": "system", "content": "\u4f60\u662f\u4e00\u4e2a\u5206\u7c7b\u52a9\u624b，\u53ea\u56de\u7b54\u7c7b\u522b\u540d\u79f0\u3002"},
                                                      \n {"role": "user", "content": prompt}
                                                      \n ],
                                                      \n "stream": False
                                                      \n },
                                                      \n timeout=60
                                                      \n )
                                                      \n return response.json()["choices"][0]["message"]["content"].strip()<\/p>\n

                                                      # \u8bc4\u6d4b\u96c6
                                                      \ntest_cases = [
                                                      \n {"id": 1, "text": "\u4eca\u5929\u5929\u6c14\u771f\u597d", "label": "\u79ef\u6781"},
                                                      \n {"id": 2, "text": "\u8fd9\u4e2a\u4ea7\u54c1\u592a\u5dee\u4e86", "label": "\u6d88\u6781"},
                                                      \n {"id": 3, "text": "\u6211\u559c\u6b22\u8fd9\u4e2a\u7535\u5f71", "label": "\u79ef\u6781"},
                                                      \n {"id": 4, "text": "\u670d\u52a1\u6001\u5ea6\u5f88\u5dee", "label": "\u6d88\u6781"},
                                                      \n {"id": 5, "text": "\u4e00\u822c\u822c\u5427", "label": "\u4e2d\u6027"},
                                                      \n]<\/p>\n

                                                      # \u6279\u91cf\u8bc4\u6d4b
                                                      \nfor case in test_cases:
                                                      \n prompt = f"\u5224\u65ad\u4ee5\u4e0b\u6587\u672c\u7684\u60c5\u611f（\u79ef\u6781\/\u6d88\u6781\/\u4e2d\u6027）：{case['text']}"
                                                      \n case["pred"] = call_ollama(prompt)<\/p>\n

                                                      # \u8f6c\u4e3a DataFrame
                                                      \ndf = pd.DataFrame(test_cases)
                                                      \nprint(f"\u8bc4\u6d4b\u7ed3\u679c：\\\\n{df}")<\/p>\n

                                                      # \u7edf\u8ba1\u51c6\u786e\u7387
                                                      \ndf["correct"] = df["pred"] == df["label"]
                                                      \naccuracy = df["correct"].mean()
                                                      \nprint(f"\\\\n\u51c6\u786e\u7387：{accuracy:.2%}")<\/p>\n

                                                      # \u7b5b\u9009 badcase
                                                      \nbadcases = df[~df["correct"]]
                                                      \nprint(f"\\\\nbadcase：\\\\n{badcases}") <\/p>\n

                                                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                      df["correct"] = df["pred"] == df["label"]<\/td>\n\u65b0\u589e\u5e03\u5c14\u5217<\/td>\n<\/tr>\n
                                                      df["correct"].mean()<\/td>\n\u8ba1\u7b97\u51c6\u786e\u7387<\/td>\n<\/tr>\n
                                                      df[~df["correct"]]<\/td>\n\u7b5b\u9009 badcase<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                      \u8bf4\u660e：<\/p>\n

                                                        \n
                                                      • \n

                                                        \u6279\u91cf\u8c03\u7528\u6a21\u578b，\u6536\u96c6\u9884\u6d4b<\/p>\n<\/li>\n

                                                      • \n

                                                        \u7edf\u8ba1\u51c6\u786e\u7387<\/p>\n<\/li>\n

                                                      • \n

                                                        \u7b5b\u9009 badcase \u5206\u6790<\/p>\n<\/li>\n<\/ul>\n

                                                        2.6.3 \u6570\u636e\u6e05\u6d17\u4e0e\u5bfc\u51fa<\/h4>\n

                                                        \u5bf9\u6a21\u578b\u8f93\u51fa\u8fdb\u884c\u6570\u636e\u6e05\u6d17（\u53bb\u91cd\u3001\u7f3a\u5931\u503c\u3001\u6587\u672c\u5904\u7406），\u7136\u540e\u5bfc\u51fa\u3002<\/p>\n

                                                        # \u6570\u636e\u6e05\u6d17\u4e0e\u5bfc\u51fa<\/p>\n

                                                        import pandas as pd
                                                        \nimport numpy as np<\/p>\n

                                                        # \u6a21\u62df\u6a21\u578b\u8f93\u51fa（\u542b\u810f\u6570\u636e）
                                                        \ndf = pd.DataFrame({
                                                        \n "id": [1, 2, 3, 4, 5, 5],
                                                        \n "prompt": ["\u4ecb\u7ecdPython", "\u4ecb\u7ecdJava", "\u4ecb\u7ecdGo", None, "\u4ecb\u7ecdC++", "\u4ecb\u7ecdC++"],
                                                        \n "output": ["Python\u662f\u4e00\u79cd…", "Java\u662f\u4e00\u79cd…", None, "Go\u662f\u4e00\u79cd…", "C++\u662f\u4e00\u79cd…", "C++\u662f\u4e00\u79cd…"],
                                                        \n "score": [0.95, 0.87, 0.92, 0.88, np.nan, 0.88]
                                                        \n})<\/p>\n

                                                        print(f"\u539f\u59cb\u6570\u636e：\\\\n{df}")
                                                        \nprint(f"\\\\n\u7f3a\u5931\u503c：\\\\n{df.isnull().sum()}")<\/p>\n

                                                        # \u53bb\u91cd
                                                        \ndf = df.drop_duplicates(subset=["prompt"])
                                                        \nprint(f"\\\\n\u53bb\u91cd\u540e：\\\\n{df}")<\/p>\n

                                                        # \u5220\u9664\u7f3a\u5931\u503c
                                                        \ndf = df.dropna(subset=["prompt", "output"])
                                                        \nprint(f"\\\\n\u5220\u9664\u7f3a\u5931\u540e：\\\\n{df}")<\/p>\n

                                                        # \u586b\u5145\u7f3a\u5931\u503c
                                                        \ndf["score"] = df["score"].fillna(df["score"].mean())
                                                        \nprint(f"\\\\n\u586b\u5145\u540e：\\\\n{df}")<\/p>\n

                                                        # \u6587\u672c\u5904\u7406：\u622a\u65ad\u8f93\u51fa
                                                        \ndf["output_short"] = df["output"].str[:20]
                                                        \nprint(f"\\\\n\u622a\u65ad\u540e：\\\\n{df}")<\/p>\n

                                                        # \u5bfc\u51fa
                                                        \ndf.to_csv("cleaned_data.csv", index=False, encoding="utf-8")
                                                        \ndf.to_json("cleaned_data.json", orient="records", force_ascii=False)
                                                        \nprint("\\\\n\u5df2\u5bfc\u51fa cleaned_data.csv \u548c cleaned_data.json")<\/p>\n

                                                        \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n\n\n
                                                        df.drop_duplicates()<\/td>\n\u53bb\u91cd<\/td>\n<\/tr>\n
                                                        df.dropna()<\/td>\n\u5220\u9664\u7f3a\u5931\u503c<\/td>\n<\/tr>\n
                                                        df.fillna()<\/td>\n\u586b\u5145\u7f3a\u5931\u503c<\/td>\n<\/tr>\n
                                                        df["\u5217"].str[:20]<\/td>\n\u6587\u672c\u622a\u65ad<\/td>\n<\/tr>\n
                                                        df.to_csv()<\/td>\n\u5bfc\u51fa CSV<\/td>\n<\/tr>\n
                                                        df.to_json()<\/td>\n\u5bfc\u51fa JSON<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                        \u8bf4\u660e：<\/p>\n

                                                          \n
                                                        • \n

                                                          \u53bb\u91cd\u3001\u5220\u9664\u7f3a\u5931\u3001\u586b\u5145\u7f3a\u5931<\/p>\n<\/li>\n

                                                        • \n

                                                          \u6587\u672c\u5904\u7406<\/p>\n<\/li>\n

                                                        • \n

                                                          \u5bfc\u51fa\u4e3a CSV \/ JSON<\/p>\n<\/li>\n<\/ul>\n

                                                          \u4e09\u3001Matplotlib \u5165\u95e8（\u7b80\u8981\u4e86\u89e3）<\/h2>\n

                                                          \"\"<\/p>\n

                                                          3.1 Matplotlib \u4ecb\u7ecd<\/h3>\n

                                                          3.1.1 Matplotlib \u662f\u4ec0\u4e48<\/h4>\n

                                                          Matplotlib \u662f Python \u6700\u57fa\u7840\u3001\u6700\u5e38\u7528\u7684\u6570\u636e\u53ef\u89c6\u5316\u5e93，\u7528\u4e8e\u7ed8\u5236\u5404\u79cd\u56fe\u8868（\u6298\u7ebf\u56fe\u3001\u76f4\u65b9\u56fe\u3001\u67f1\u72b6\u56fe\u7b49）\u3002<\/p>\n

                                                          \u4e00\u53e5\u8bdd\u7406\u89e3： Matplotlib \u8ba9 Python \u80fd\u50cf Excel \u4e00\u6837\u753b\u56fe，\u4f46\u66f4\u7075\u6d3b\u3001\u66f4\u53ef\u5b9a\u5236\u3002<\/p>\n

                                                          3.1.2 \u4e3a\u4ec0\u4e48\u5927\u6a21\u578b\u89d2\u8272\u53ea\u9700\u7b80\u8981\u4e86\u89e3<\/h4>\n

                                                          \u5927\u6a21\u578b\u5e94\u7528\u5f00\u53d1\u4e2d，\u53ef\u89c6\u5316\u4e0d\u662f\u6838\u5fc3\u5de5\u4f5c，\u4f46\u9700\u8981\u5feb\u901f\u51fa\u56fe\u8f85\u52a9\u5206\u6790\u3002<\/p>\n

                                                          \u5927\u6a21\u578b\u573a\u666f\u4e0b\u7684\u5178\u578b\u9700\u6c42<\/p>\n\n\u9700\u6c42\u56fe\u8868\u7c7b\u578b\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                                                          \u770b loss \u66f2\u7ebf<\/td>\n\u6298\u7ebf\u56fe<\/td>\n\u89c2\u5bdf\u8bad\u7ec3\u662f\u5426\u6536\u655b<\/td>\n<\/tr>\n
                                                          \u770b\u6570\u636e\u5206\u5e03<\/td>\n\u76f4\u65b9\u56fe<\/td>\n\u89c2\u5bdf\u6570\u636e\u5206\u5e03\u662f\u5426\u5747\u5300<\/td>\n<\/tr>\n
                                                          \u770b\u7c7b\u522b\u5206\u5e03<\/td>\n\u67f1\u72b6\u56fe<\/td>\n\u89c2\u5bdf\u7c7b\u522b\u662f\u5426\u5e73\u8861<\/td>\n<\/tr>\n
                                                          \u8bb0\u5f55\u5b9e\u9a8c<\/td>\n\u4fdd\u5b58\u56fe\u7247<\/td>\n\u4fdd\u5b58\u5230\u5b9e\u9a8c\u76ee\u5f55，\u4fbf\u4e8e\u5bf9\u6bd4<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                          3.2 \u5927\u6a21\u578b\u573a\u666f\u4e0b\u6700\u5e38\u7528\u7684\u56fe<\/h3>\n

                                                          3.2.1 loss \/ accuracy \u66f2\u7ebf（\u6298\u7ebf\u56fe）<\/h4>\n

                                                          \u6298\u7ebf\u56fe\u7528\u4e8e\u5c55\u793a\u968f\u65f6\u95f4\u53d8\u5316\u7684\u8d8b\u52bf，\u662f\u5927\u6a21\u578b\u8bad\u7ec3\u4e2d\u6700\u5e38\u7528\u7684\u56fe\u3002\u901a\u8fc7 loss \u66f2\u7ebf\u5224\u65ad\u8bad\u7ec3\u662f\u5426\u6536\u655b\u3001\u662f\u5426\u8fc7\u62df\u5408\u3002<\/p>\n\n\u573a\u666f\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                          \u770b loss \u66f2\u7ebf<\/td>\n\u5224\u65ad\u8bad\u7ec3\u662f\u5426\u6536\u655b<\/td>\n<\/tr>\n
                                                          \u770b accuracy \u66f2\u7ebf<\/td>\n\u5224\u65ad\u6a21\u578b\u662f\u5426\u63d0\u5347<\/td>\n<\/tr>\n
                                                          \u5bf9\u6bd4\u591a\u6b21\u5b9e\u9a8c<\/td>\n\u5bf9\u6bd4\u4e0d\u540c\u8d85\u53c2\u6570\u7684\u8bad\u7ec3\u6548\u679c<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                            \n
                                                          • \n

                                                            X \u8f74：epoch（\u8bad\u7ec3\u8f6e\u6b21）<\/p>\n<\/li>\n

                                                          • \n

                                                            Y \u8f74：loss（\u635f\u5931\u503c）<\/p>\n<\/li>\n

                                                          • \n

                                                            \u5224\u65ad：loss \u6301\u7eed\u4e0b\u964d\u5e76\u8d8b\u4e8e\u5e73\u7a33 \u2192 \u6536\u655b<\/p>\n<\/li>\n

                                                          • \n

                                                            \u5224\u65ad：loss \u5148\u964d\u540e\u5347 \u2192 \u8fc7\u62df\u5408<\/p>\n<\/li>\n<\/ul>\n

                                                            3.2.2 \u6570\u636e\u5206\u5e03（\u76f4\u65b9\u56fe）<\/h4>\n

                                                            \u76f4\u65b9\u56fe\u7528\u4e8e\u5c55\u793a\u6570\u636e\u7684\u5206\u5e03\u60c5\u51b5，\u89c2\u5bdf\u6570\u636e\u662f\u5426\u5747\u5300\u3001\u662f\u5426\u504f\u659c\u3002<\/p>\n\n\u573a\u666f\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                            \u770b\u6570\u636e\u5206\u5e03<\/td>\n\u89c2\u5bdf\u6570\u636e\u662f\u5426\u5747\u5300<\/td>\n<\/tr>\n
                                                            \u770b\u957f\u5c3e\u5206\u5e03<\/td>\n\u89c2\u5bdf\u662f\u5426\u5b58\u5728\u957f\u5c3e<\/td>\n<\/tr>\n
                                                            \u6570\u636e\u6e05\u6d17\u524d<\/td>\n\u5224\u65ad\u662f\u5426\u9700\u8981\u5f52\u4e00\u5316<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                              \n
                                                            • \n

                                                              X \u8f74：\u6570\u636e\u503c（\u5206\u7bb1）<\/p>\n<\/li>\n

                                                            • \n

                                                              Y \u8f74：\u9891\u6b21<\/p>\n<\/li>\n

                                                            • \n

                                                              \u5224\u65ad：\u4e2d\u95f4\u9ad8\u4e24\u8fb9\u4f4e \u2192 \u6b63\u6001\u5206\u5e03<\/p>\n<\/li>\n

                                                            • \n

                                                              \u5224\u65ad：\u504f\u5de6\u6216\u504f\u53f3 \u2192 \u504f\u659c\u5206\u5e03<\/p>\n<\/li>\n<\/ul>\n

                                                              3.2.3 \u7c7b\u522b\u5206\u5e03（\u67f1\u72b6\u56fe）<\/h4>\n

                                                              \u67f1\u72b6\u56fe\u7528\u4e8e\u5c55\u793a\u4e0d\u540c\u7c7b\u522b\u7684\u6570\u91cf\u5bf9\u6bd4，\u89c2\u5bdf\u7c7b\u522b\u662f\u5426\u5e73\u8861\u3002<\/p>\n\n\u573a\u666f\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                              \u770b\u7c7b\u522b\u5206\u5e03<\/td>\n\u89c2\u5bdf\u7c7b\u522b\u662f\u5426\u5e73\u8861<\/td>\n<\/tr>\n
                                                              badcase \u7edf\u8ba1<\/td>\n\u7edf\u8ba1\u5404\u7c7b badcase \u6570\u91cf<\/td>\n<\/tr>\n
                                                              \u8bc4\u6d4b\u7ed3\u679c<\/td>\n\u7edf\u8ba1\u5404\u6807\u7b7e\u7684\u51c6\u786e\u7387<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                                \n
                                                              • \n

                                                                X \u8f74：\u7c7b\u522b<\/p>\n<\/li>\n

                                                              • \n

                                                                Y \u8f74：\u6570\u91cf<\/p>\n<\/li>\n

                                                              • \n

                                                                \u5224\u65ad：\u67f1\u5b50\u9ad8\u5ea6\u5dee\u5f02\u5927 \u2192 \u7c7b\u522b\u4e0d\u5e73\u8861<\/p>\n<\/li>\n

                                                              • \n

                                                                \u5e94\u7528：badcase \u7edf\u8ba1\u3001\u8bc4\u6d4b\u7ed3\u679c\u5206\u6790<\/p>\n<\/li>\n<\/ul>\n

                                                                3.2.4 \u6563\u70b9\u56fe（\u4e86\u89e3）<\/h4>\n

                                                                \u6563\u70b9\u56fe\u7528\u4e8e\u5c55\u793a\u4e24\u4e2a\u53d8\u91cf\u4e4b\u95f4\u7684\u5173\u7cfb，\u89c2\u5bdf\u662f\u5426\u5b58\u5728\u76f8\u5173\u6027\u3002<\/p>\n\n\u573a\u666f\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                                \u770b\u76f8\u5173\u6027<\/td>\n\u89c2\u5bdf\u4e24\u4e2a\u53d8\u91cf\u662f\u5426\u76f8\u5173<\/td>\n<\/tr>\n
                                                                \u770b\u805a\u7c7b<\/td>\n\u89c2\u5bdf\u6570\u636e\u662f\u5426\u5206\u7fa4<\/td>\n<\/tr>\n
                                                                \u5f02\u5e38\u503c\u68c0\u6d4b<\/td>\n\u89c2\u5bdf\u662f\u5426\u6709\u79bb\u7fa4\u70b9<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                                                                  \n
                                                                • \n

                                                                  X \u8f74：\u53d8\u91cf 1<\/p>\n<\/li>\n

                                                                • \n

                                                                  Y \u8f74：\u53d8\u91cf 2<\/p>\n<\/li>\n

                                                                • \n

                                                                  \u5224\u65ad：\u70b9\u5448\u4e0a\u5347\u8d8b\u52bf \u2192 \u6b63\u76f8\u5173<\/p>\n<\/li>\n

                                                                • \n

                                                                  \u5224\u65ad：\u70b9\u5206\u6563 \u2192 \u65e0\u76f8\u5173<\/p>\n<\/li>\n<\/ul>\n

                                                                  \u8bf4\u660e： \u6563\u70b9\u56fe\u5728\u5927\u6a21\u578b\u573a\u666f\u4e2d\u8f83\u5c11\u7528，\u4e86\u89e3\u5373\u53ef\u3002<\/p>\n

                                                                  3.2.5 \u56fe\u8868\u9009\u62e9\u6307\u5357（\u4e86\u89e3）<\/h4>\n

                                                                  \u6839\u636e\u5206\u6790\u76ee\u7684\u9009\u62e9\u5408\u9002\u7684\u56fe\u8868\u7c7b\u578b\u3002<\/p>\n\n\u5206\u6790\u76ee\u7684\u63a8\u8350\u56fe\u8868\u793a\u4f8b<\/tr>\n\n\n\n\n\n
                                                                  \u770b\u8d8b\u52bf<\/td>\n\u6298\u7ebf\u56fe<\/td>\nloss \u66f2\u7ebf<\/td>\n<\/tr>\n
                                                                  \u770b\u5206\u5e03<\/td>\n\u76f4\u65b9\u56fe<\/td>\n\u6570\u636e\u5206\u5e03<\/td>\n<\/tr>\n
                                                                  \u770b\u5bf9\u6bd4<\/td>\n\u67f1\u72b6\u56fe<\/td>\n\u7c7b\u522b\u5206\u5e03<\/td>\n<\/tr>\n
                                                                  \u770b\u5173\u7cfb<\/td>\n\u6563\u70b9\u56fe<\/td>\n\u76f8\u5173\u6027<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                  \u5927\u6a21\u578b\u573a\u666f\u901f\u67e5<\/p>\n\n\u9700\u6c42\u56fe\u8868<\/tr>\n\n\n\n\n\n
                                                                  \u8bad\u7ec3\u662f\u5426\u6536\u655b<\/td>\n\u6298\u7ebf\u56fe<\/td>\n<\/tr>\n
                                                                  \u6570\u636e\u662f\u5426\u5747\u5300<\/td>\n\u76f4\u65b9\u56fe<\/td>\n<\/tr>\n
                                                                  \u7c7b\u522b\u662f\u5426\u5e73\u8861<\/td>\n\u67f1\u72b6\u56fe<\/td>\n<\/tr>\n
                                                                  \u53d8\u91cf\u662f\u5426\u76f8\u5173<\/td>\n\u6563\u70b9\u56fe<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                  \u8bf4\u660e： \u5927\u6a21\u578b\u89d2\u8272\u4e3b\u8981\u7528\u6298\u7ebf\u56fe\u3001\u76f4\u65b9\u56fe\u3001\u67f1\u72b6\u56fe，\u6563\u70b9\u56fe\u4e86\u89e3\u5373\u53ef\u3002<\/p>\n<\/p>\n

                                                                  3.3 \u57fa\u7840\u7ed8\u56fe\u529f\u80fd<\/h3>\n

                                                                  3.3.1 \u4e0e pandas \u8054\u52a8 df.plot()\u00a0<\/h4>\n

                                                                  Pandas \u5185\u7f6e plot() \u65b9\u6cd5，\u57fa\u4e8e Matplotlib，\u53ef\u4e00\u884c\u4ee3\u7801\u51fa\u56fe，\u662f\u6700\u4fbf\u6377\u7684\u65b9\u5f0f\u3002<\/p>\n

                                                                  # \u4e0e pandas \u8054\u52a8 df.plot()<\/p>\n

                                                                  import pandas as pd
                                                                  \nimport matplotlib.pyplot as plt<\/p>\n

                                                                  # \u6a21\u62df\u8bad\u7ec3\u65e5\u5fd7
                                                                  \ndf = pd.DataFrame({
                                                                  \n "epoch": [1, 2, 3, 4, 5, 6, 7, 8],
                                                                  \n "loss": [1.0, 0.8, 0.6, 0.5, 0.4, 0.35, 0.33, 0.32],
                                                                  \n "accuracy": [0.5, 0.6, 0.7, 0.75, 0.8, 0.83, 0.85, 0.86]
                                                                  \n})<\/p>\n

                                                                  # \u6298\u7ebf\u56fe：loss \u66f2\u7ebf
                                                                  \ndf.plot(x="epoch", y="loss", title="Loss Curve")
                                                                  \nplt.savefig("loss_curve.png")
                                                                  \nplt.show()<\/p>\n

                                                                  # \u591a\u6761\u66f2\u7ebf
                                                                  \ndf.plot(x="epoch", y=["loss", "accuracy"], title="Training Metrics")
                                                                  \nplt.show() <\/p>\n

                                                                  \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                                                                  x<\/td>\nX \u8f74\u5217\u540d<\/td>\n<\/tr>\n
                                                                  y<\/td>\nY \u8f74\u5217\u540d（\u53ef\u4e3a\u5217\u8868）<\/td>\n<\/tr>\n
                                                                  kind<\/td>\n\u56fe\u8868\u7c7b\u578b：line\u3001bar\u3001hist\u3001scatter<\/td>\n<\/tr>\n
                                                                  title<\/td>\n\u56fe\u8868\u6807\u9898<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                  \u8bf4\u660e：<\/p>\n

                                                                    \n
                                                                  • \n

                                                                    df.plot() \u4e00\u884c\u51fa\u56fe<\/p>\n<\/li>\n

                                                                  • \n

                                                                    kind \u6307\u5b9a\u56fe\u8868\u7c7b\u578b<\/p>\n<\/li>\n

                                                                  • \n

                                                                    \u8fd4\u56de Axes \u5bf9\u8c61，\u53ef\u7ee7\u7eed\u5b9a\u5236<\/p>\n<\/li>\n<\/ul>\n

                                                                    3.3.2 \u4e0e NumPy \u6570\u7ec4\u76f4\u63a5\u7ed8\u56fe\u00a0<\/h4>\n

                                                                    Matplotlib \u53ef\u76f4\u63a5\u63a5\u6536 NumPy \u6570\u7ec4，\u65e0\u9700\u8f6c DataFrame\u3002<\/p>\n

                                                                    # \u4e0e NumPy \u6570\u7ec4\u76f4\u63a5\u7ed8\u56fe<\/p>\n

                                                                    import numpy as np
                                                                    \nimport matplotlib.pyplot as plt<\/p>\n

                                                                    # \u751f\u6210\u6570\u636e
                                                                    \nx = np.linspace(0, 10, 100)
                                                                    \ny = np.sin(x)<\/p>\n

                                                                    # \u76f4\u63a5\u7ed8\u56fe
                                                                    \nplt.plot(x, y)
                                                                    \nplt.title("Sin Wave")
                                                                    \nplt.show()<\/p>\n

                                                                    # \u591a\u6761\u66f2\u7ebf
                                                                    \ny2 = np.cos(x)
                                                                    \nplt.plot(x, y, label="sin")
                                                                    \nplt.plot(x, y2, label="cos")
                                                                    \nplt.legend()
                                                                    \nplt.show() <\/p>\n

                                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n
                                                                    x, y<\/td>\n\u6570\u636e\u6570\u7ec4<\/td>\n<\/tr>\n
                                                                    label<\/td>\n\u66f2\u7ebf\u6807\u7b7e，\u7528\u4e8e\u56fe\u4f8b<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                    \u8bf4\u660e： Matplotlib \u53ef\u76f4\u63a5\u63a5\u6536 NumPy \u6570\u7ec4\u3002<\/p>\n

                                                                    3.3.3 pyplot \u98ce\u683c plt.plot \/ scatter \/ hist<\/h4>\n

                                                                    pyplot \u662f Matplotlib \u7684\u51fd\u6570\u5f0f\u63a5\u53e3，\u63d0\u4f9b plt.plot\u3001plt.scatter\u3001plt.hist \u7b49\u5feb\u6377\u65b9\u6cd5\u3002<\/p>\n

                                                                    # pyplot \u98ce\u683c<\/p>\n

                                                                    import numpy as np
                                                                    \nimport matplotlib.pyplot as plt<\/p>\n

                                                                    # ===== \u6298\u7ebf\u56fe =====
                                                                    \nx = np.linspace(0, 10, 100)
                                                                    \nplt.plot(x, np.sin(x))
                                                                    \nplt.title("Line Plot")
                                                                    \nplt.show()<\/p>\n

                                                                    # ===== \u6563\u70b9\u56fe =====
                                                                    \nx = np.random.rand(50)
                                                                    \ny = np.random.rand(50)
                                                                    \nplt.scatter(x, y)
                                                                    \nplt.title("Scatter Plot")
                                                                    \nplt.show()<\/p>\n

                                                                    # ===== \u76f4\u65b9\u56fe =====
                                                                    \ndata = np.random.randn(1000)
                                                                    \nplt.hist(data, bins=30)
                                                                    \nplt.title("Histogram")
                                                                    \nplt.show() <\/p>\n

                                                                    \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                                    plt.plot()<\/td>\n\u6298\u7ebf\u56fe<\/td>\n<\/tr>\n
                                                                    plt.scatter()<\/td>\n\u6563\u70b9\u56fe<\/td>\n<\/tr>\n
                                                                    plt.hist()<\/td>\n\u76f4\u65b9\u56fe\u3002bins \u6307\u5b9a\u5206\u7bb1\u6570<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                    \u8bf4\u660e：<\/p>\n

                                                                      \n
                                                                    • \n

                                                                      plt.plot \u6298\u7ebf\u56fe<\/p>\n<\/li>\n

                                                                    • \n

                                                                      plt.scatter \u6563\u70b9\u56fe<\/p>\n<\/li>\n

                                                                    • \n

                                                                      plt.hist \u76f4\u65b9\u56fe<\/p>\n<\/li>\n<\/ul>\n

                                                                      3.3.4 \u6807\u9898\u3001\u5750\u6807\u8f74\u6807\u7b7e\u3001\u56fe\u4f8b<\/h4>\n

                                                                      \u4e3a\u56fe\u8868\u6dfb\u52a0\u6807\u9898\u3001\u5750\u6807\u8f74\u6807\u7b7e\u3001\u56fe\u4f8b，\u4f7f\u56fe\u8868\u66f4\u6e05\u6670\u3002<\/p>\n

                                                                      # \u6807\u9898\u3001\u5750\u6807\u8f74\u6807\u7b7e\u3001\u56fe\u4f8b<\/p>\n

                                                                      import numpy as np
                                                                      \nimport matplotlib.pyplot as plt<\/p>\n

                                                                      x = np.linspace(0, 10, 100)<\/p>\n

                                                                      plt.plot(x, np.sin(x), label="sin")
                                                                      \nplt.plot(x, np.cos(x), label="cos")<\/p>\n

                                                                      # \u6807\u9898
                                                                      \nplt.title("test")<\/p>\n

                                                                      # \u5750\u6807\u8f74\u6807\u7b7e
                                                                      \nplt.xlabel("x")
                                                                      \nplt.ylabel("y")<\/p>\n

                                                                      # \u56fe\u4f8b
                                                                      \nplt.legend()<\/p>\n

                                                                      plt.show() <\/p>\n

                                                                      \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u65b9\u6cd5\u8bf4\u660e<\/tr>\n\n\n\n\n\n
                                                                      plt.title()<\/td>\n\u8bbe\u7f6e\u6807\u9898<\/td>\n<\/tr>\n
                                                                      plt.xlabel()<\/td>\n\u8bbe\u7f6e X \u8f74\u6807\u7b7e<\/td>\n<\/tr>\n
                                                                      plt.ylabel()<\/td>\n\u8bbe\u7f6e Y \u8f74\u6807\u7b7e<\/td>\n<\/tr>\n
                                                                      plt.legend()<\/td>\n\u663e\u793a\u56fe\u4f8b<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                      \u8bf4\u660e：<\/p>\n

                                                                        \n
                                                                      • \n

                                                                        title \u6807\u9898<\/p>\n<\/li>\n

                                                                      • \n

                                                                        xlabel \/ ylabel \u5750\u6807\u8f74\u6807\u7b7e<\/p>\n<\/li>\n

                                                                      • \n

                                                                        legend \u56fe\u4f8b，\u9700\u5728 plot \u65f6\u6307\u5b9a label<\/p>\n<\/li>\n<\/ul>\n

                                                                        3.3.5 \u4fdd\u5b58\u56fe\u7247 savefig<\/h4>\n

                                                                        savefig() \u7528\u4e8e\u4fdd\u5b58\u56fe\u7247\u5230\u6587\u4ef6，\u5e38\u7528\u4e8e\u8bb0\u5f55\u5b9e\u9a8c\u3002<\/p>\n

                                                                        # \u4fdd\u5b58\u56fe\u7247<\/p>\n

                                                                        import numpy as np
                                                                        \nimport matplotlib.pyplot as plt<\/p>\n

                                                                        x = np.linspace(0, 10, 100)<\/p>\n

                                                                        plt.plot(x, np.sin(x))
                                                                        \nplt.title("Sin Wave")<\/p>\n

                                                                        # \u4fdd\u5b58\u56fe\u7247
                                                                        \nplt.savefig("sin_wave.png", dpi=150, bbox_inches="tight")
                                                                        \nprint("\u5df2\u4fdd\u5b58 sin_wave.png") <\/p>\n

                                                                        \u65b9\u6cd5\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u8bf4\u660e<\/tr>\n\n\n\n\n
                                                                        fname<\/td>\n\u6587\u4ef6\u540d，\u652f\u6301 PNG\u3001JPG\u3001PDF<\/td>\n<\/tr>\n
                                                                        dpi<\/td>\n\u5206\u8fa8\u7387，\u9ed8\u8ba4 100<\/td>\n<\/tr>\n
                                                                        bbox_inches<\/td>\n"tight" \u53bb\u9664\u591a\u4f59\u7a7a\u767d<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                                                                        \u8bf4\u660e：<\/p>\n

                                                                          \n
                                                                        • \n

                                                                          \u652f\u6301\u591a\u79cd\u683c\u5f0f：PNG\u3001JPG\u3001PDF\u3001SVG<\/p>\n<\/li>\n

                                                                        • \n

                                                                          dpi \u63a7\u5236\u6e05\u6670\u5ea6<\/p>\n<\/li>\n

                                                                        • \n

                                                                          bbox_inches="tight" \u53bb\u9664\u7a7a\u767d<\/p>\n<\/li>\n<\/ul><\/p>\n","protected":false},"excerpt":{"rendered":"

                                                                          \u6458\u8981：Pandas \u662f Python \u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u5e93，\u63d0\u4f9b Series（\u4e00\u7ef4）\u548c DataFrame（\u4e8c\u7ef4）\u4e24\u79cd\u6570\u636e\u7ed3\u6784，\u652f\u6301\u6570\u636e\u8bfb\u5199（CSV\u3001JSON）\u3001\u7d22\u5f15\u9009\u62e9（loc\/iloc）\u3001\u6570\u636e\u6e05\u6d17（\u53bb\u91cd\u3001\u7f3a\u5931\u503c）\u3001\u6570\u636e\u5408\u5e76（merge\/concat）\u3001\u5206\u7ec4\u805a\u5408&#xf<\/p>\n","protected":false},"author":2,"featured_media":118011,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[4278,371,801],"topic":[],"class_list":["post-118017","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-server","tag-pandas","tag-371","tag-801"],"yoast_head":"\n\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wsisp.com\/helps\/118017.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"\u6458\u8981：Pandas \u662f Python \u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u5e93，\u63d0\u4f9b Series（\u4e00\u7ef4）\u548c DataFrame（\u4e8c\u7ef4）\u4e24\u79cd\u6570\u636e\u7ed3\u6784，\u652f\u6301\u6570\u636e\u8bfb\u5199（CSV\u3001JSON）\u3001\u7d22\u5f15\u9009\u62e9（loc\/iloc）\u3001\u6570\u636e\u6e05\u6d17（\u53bb\u91cd\u3001\u7f3a\u5931\u503c）\u3001\u6570\u636e\u5408\u5e76（merge\/concat）\u3001\u5206\u7ec4\u805a\u5408&#xf\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/118017.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-25T19:24:38+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/10\/20261025192435-6ade5773be470.png\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"18 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/118017.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/118017.html\",\"name\":\"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2026-10-25T19:24:38+00:00\",\"dateModified\":\"2026-10-25T19:24:38+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/118017.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/118017.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/118017.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wsisp.com\/helps\/118017.html","og_locale":"zh_CN","og_type":"article","og_title":"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"\u6458\u8981：Pandas \u662f Python \u6570\u636e\u5904\u7406\u7684\u6838\u5fc3\u5e93，\u63d0\u4f9b Series（\u4e00\u7ef4）\u548c DataFrame（\u4e8c\u7ef4）\u4e24\u79cd\u6570\u636e\u7ed3\u6784，\u652f\u6301\u6570\u636e\u8bfb\u5199（CSV\u3001JSON）\u3001\u7d22\u5f15\u9009\u62e9（loc\/iloc）\u3001\u6570\u636e\u6e05\u6d17（\u53bb\u91cd\u3001\u7f3a\u5931\u503c）\u3001\u6570\u636e\u5408\u5e76（merge\/concat）\u3001\u5206\u7ec4\u805a\u5408&#xf","og_url":"https:\/\/www.wsisp.com\/helps\/118017.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2026-10-25T19:24:38+00:00","og_image":[{"url":"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/10\/20261025192435-6ade5773be470.png"}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"18 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/118017.html","url":"https:\/\/www.wsisp.com\/helps\/118017.html","name":"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2026-10-25T19:24:38+00:00","dateModified":"2026-10-25T19:24:38+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/118017.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/118017.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/118017.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"\u5927\u6a21\u578b\u5b66\u4e60\u4e4b\u6570\u636e\u5206\u6790\u2014\u2014\u9762\u5411\u5927\u6a21\u578b\u7684Pandas \u4e0e Matplotlib\u57fa\u7840"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/118017","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=118017"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/118017\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media\/118011"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=118017"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=118017"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=118017"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=118017"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}