{"id":926998,"date":"2024-12-26T15:59:42","date_gmt":"2024-12-26T07:59:42","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/926998.html"},"modified":"2024-12-26T15:59:44","modified_gmt":"2024-12-26T07:59:44","slug":"python%e5%a6%82%e4%bd%95%e7%bc%a9%e6%94%be","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/926998.html","title":{"rendered":"Python\u5982\u4f55\u7f29\u653e"},"content":{"rendered":"

\"Python\u5982\u4f55\u7f29\u653e\"<\/p>\n

Python\u7f29\u653e\u7684\u65b9\u6cd5\u5305\u62ec\u4f7f\u7528\u5185\u7f6e\u5e93\u8fdb\u884c\u57fa\u672c\u7f29\u653e\u3001\u5229\u7528\u7b2c\u4e09\u65b9\u5e93\u5982NumPy\u8fdb\u884c\u77e9\u9635\u7f29\u653e\u3001\u4ee5\u53ca\u4f7f\u7528Pandas\u8fdb\u884c\u6570\u636e\u7f29\u653e\u3002\u5728\u6570\u636e\u79d1\u5b66\u548c\u673a\u5668\u5b66\u4e60<\/a>\u9886\u57df\uff0c\u6570\u636e\u7f29\u653e\u662f\u6570\u636e\u9884\u5904\u7406\u7684\u91cd\u8981\u6b65\u9aa4\uff0c\u80fd\u6709\u6548\u63d0\u5347\u6a21\u578b\u6027\u80fd\u3002<\/strong><\/p>\n<\/p>\n

\u5728\u6df1\u5165\u63a2\u8ba8\u4e4b\u524d\uff0c\u5148\u7b80\u5355\u4ecb\u7ecd\u4e00\u4e0b\u6570\u636e\u7f29\u653e\u3002\u6570\u636e\u7f29\u653e\u662f\u5c06\u6570\u636e\u7684\u503c\u8c03\u6574\u5230\u7279\u5b9a\u8303\u56f4\u6216\u5c3a\u5ea6\u4e0a\u7684\u8fc7\u7a0b\u3002\u8fd9\u5728\u673a\u5668\u5b66\u4e60\u4e2d\u5c24\u4e3a\u91cd\u8981\uff0c\u56e0\u4e3a\u8bb8\u591a\u7b97\u6cd5\u5bf9\u8f93\u5165\u6570\u636e\u7684\u5c3a\u5ea6\u654f\u611f\u3002\u901a\u8fc7\u7f29\u653e\u53ef\u4ee5\u63d0\u9ad8\u6a21\u578b\u7684\u51c6\u786e\u6027\u548c\u8bad\u7ec3\u901f\u5ea6\u3002<\/p>\n<\/p>\n

\u4e00\u3001\u4f7f\u7528\u5185\u7f6e\u5e93\u8fdb\u884c\u57fa\u672c\u7f29\u653e<\/p>\n<\/p>\n

Python\u5185\u7f6e\u5e93\u63d0\u4f9b\u4e86\u4e00\u4e9b\u7b80\u5355\u7684\u65b9\u6cd5\u53ef\u4ee5\u5bf9\u6570\u636e\u8fdb\u884c\u7f29\u653e\u3002\u53ef\u4ee5\u4f7f\u7528\u57fa\u672c\u7684\u6570\u5b66\u8fd0\u7b97\u5b9e\u73b0\u6570\u636e\u7684\u6700\u5c0f\u503c-\u6700\u5927\u503c\u7f29\u653e\uff08Min-Max Scaling\uff09\u548c\u6807\u51c6\u5316\uff08Standardization\uff09\u3002<\/p>\n<\/p>\n

    \n
  1. \u6700\u5c0f\u503c-\u6700\u5927\u503c\u7f29\u653e<\/strong><\/li>\n<\/ol>\n

    \u6700\u5c0f\u503c-\u6700\u5927\u503c\u7f29\u653e\u5c06\u6570\u636e\u7f29\u653e\u5230\u4e00\u4e2a\u56fa\u5b9a\u8303\u56f4\uff08\u901a\u5e38\u662f0\u52301\uff09\u3002\u516c\u5f0f\u4e3a\uff1a<\/p>\n<\/p>\n

    [ X' = \\frac{X – X_{min}}{X_{max} – X_{min}} ]<\/p>\n<\/p>\n

    \u901a\u8fc7\u8fd9\u79cd\u7f29\u653e\u65b9\u5f0f\uff0c\u53ef\u4ee5\u4fdd\u6301\u6570\u636e\u4e4b\u95f4\u7684\u6bd4\u4f8b\u5173\u7cfb\u3002\u5b9e\u73b0\u65f6\uff0c\u53ef\u4ee5\u7b80\u5355\u5730\u4f7f\u7528Python\u7684\u57fa\u672c\u8fd0\u7b97\u7b26\u3002<\/p>\n<\/p>\n

    def min_max_scaling(data):<\/p>\n
        min_val = min(data)<\/p>\n
        max_val = max(data)<\/p>\n
        return [(x - min_val) \/ (max_val - min_val) for x in data]<\/p>\n
    <\/code><\/pre>\n<\/p>\n
      \n
    1. \u6807\u51c6\u5316<\/strong><\/li>\n<\/ol>\n
      \u6807\u51c6\u5316\u662f\u5c06\u6570\u636e\u8c03\u6574\u4e3a\u5747\u503c\u4e3a0\uff0c\u6807\u51c6\u5dee\u4e3a1\u7684\u8fc7\u7a0b\uff0c\u516c\u5f0f\u4e3a\uff1a<\/p>\n<\/p>\n
      [ X' = \\frac{X – \\mu}{\\sigma} ]<\/p>\n<\/p>\n
      \u5176\u4e2d\uff0c(\\mu)\u662f\u5747\u503c\uff0c(\\sigma)\u662f\u6807\u51c6\u5dee\u3002\u6807\u51c6\u5316\u9002\u7528\u4e8e\u6570\u636e\u5177\u6709\u5f02\u5e38\u503c\u6216\u4e0d\u540c\u5355\u4f4d\u91cf\u7eb2\u7684\u60c5\u51b5\u3002<\/p>\n<\/p>\n
      def standardize(data):<\/p>\n
          mean = sum(data) \/ len(data)<\/p>\n
          variance = sum((x - mean)  2 for x in data) \/ len(data)<\/p>\n
          stddev = variance  0.5<\/p>\n
          return [(x - mean) \/ stddev for x in data]<\/p>\n
      <\/code><\/pre>\n<\/p>\n
      \u4e8c\u3001\u5229\u7528NumPy\u8fdb\u884c\u77e9\u9635\u7f29\u653e<\/p>\n<\/p>\n
      NumPy\u662fPython\u4e2d\u6700\u5f3a\u5927\u7684\u79d1\u5b66\u8ba1\u7b97\u5e93\u4e4b\u4e00\u3002\u5b83\u63d0\u4f9b\u4e86\u591a\u7ef4\u6570\u7ec4\u5bf9\u8c61\u548c\u4e00\u7cfb\u5217\u51fd\u6570\u6765\u8fdb\u884c\u590d\u6742\u7684\u6570\u5b66\u8fd0\u7b97\u3002<\/p>\n<\/p>\n
        \n
      1. NumPy\u5b9e\u73b0\u6700\u5c0f\u503c-\u6700\u5927\u503c\u7f29\u653e<\/strong><\/li>\n<\/ol>\n
        NumPy\u4f7f\u5f97\u5bf9\u5927\u578b\u6570\u636e\u96c6\u7684\u64cd\u4f5c\u66f4\u52a0\u9ad8\u6548\uff0c\u5c24\u5176\u662f\u6d89\u53ca\u77e9\u9635\u8fd0\u7b97\u65f6\u3002<\/p>\n<\/p>\n
        import numpy as np<\/p>\n
        def numpy_min_max_scaling(data):<\/p>\n
            min_val = np.min(data)<\/p>\n
            max_val = np.max(data)<\/p>\n
            return (data - min_val) \/ (max_val - min_val)<\/p>\n
        <\/code><\/pre>\n<\/p>\n
          \n
        1. NumPy\u5b9e\u73b0\u6807\u51c6\u5316<\/strong><\/li>\n<\/ol>\n
          \u540c\u6837\uff0cNumPy\u4e5f\u80fd\u8f7b\u677e\u5b9e\u73b0\u6570\u636e\u7684\u6807\u51c6\u5316\u3002<\/p>\n<\/p>\n
          def numpy_standardize(data):<\/p>\n
              mean = np.mean(data)<\/p>\n
              stddev = np.std(data)<\/p>\n
              return (data - mean) \/ stddev<\/p>\n
          <\/code><\/pre>\n<\/p>\n
          \u4f7f\u7528NumPy\u8fdb\u884c\u6570\u636e\u7f29\u653e\u53ef\u4ee5\u6781\u5927\u5730\u63d0\u9ad8\u8ba1\u7b97\u6548\u7387\uff0c\u5c24\u5176\u662f\u5f53\u6570\u636e\u91cf\u975e\u5e38\u5927\u65f6\u3002<\/strong><\/p>\n<\/p>\n
          \u4e09\u3001\u4f7f\u7528Pandas\u8fdb\u884c\u6570\u636e\u7f29\u653e<\/p>\n<\/p>\n
          Pandas\u662f\u4e00\u4e2a\u5f3a\u5927\u7684\u6570\u636e\u5206\u6790\u5e93\uff0c\u5c24\u5176\u9002\u5408\u5904\u7406\u7ed3\u6784\u5316\u6570\u636e\u3002\u5176DataFrame\u5bf9\u8c61\u80fd\u65b9\u4fbf\u5730\u5bf9\u6570\u636e\u8fdb\u884c\u64cd\u4f5c\u3002<\/p>\n<\/p>\n
            \n
          1. Pandas\u5b9e\u73b0\u6700\u5c0f\u503c-\u6700\u5927\u503c\u7f29\u653e<\/strong><\/li>\n<\/ol>\n
            Pandas\u7684DataFrame\u4f7f\u5f97\u5bf9\u5217\u6570\u636e\u7684\u64cd\u4f5c\u66f4\u52a0\u76f4\u89c2\u3002<\/p>\n<\/p>\n
            import pandas as pd<\/p>\n
            def pandas_min_max_scaling(df, column_name):<\/p>\n
                min_val = df[column_name].min()<\/p>\n
                max_val = df[column_name].max()<\/p>\n
                df[column_name] = (df[column_name] - min_val) \/ (max_val - min_val)<\/p>\n
                return df<\/p>\n
            <\/code><\/pre>\n<\/p>\n
              \n
            1. Pandas\u5b9e\u73b0\u6807\u51c6\u5316<\/strong><\/li>\n<\/ol>\n
              Pandas\u540c\u6837\u53ef\u4ee5\u7528\u4e8e\u6807\u51c6\u5316\u6570\u636e\u3002<\/p>\n<\/p>\n
              def pandas_standardize(df, column_name):<\/p>\n
                  mean = df[column_name].mean()<\/p>\n
                  stddev = df[column_name].std()<\/p>\n
                  df[column_name] = (df[column_name] - mean) \/ stddev<\/p>\n
                  return df<\/p>\n
              <\/code><\/pre>\n<\/p>\n
              \u5229\u7528Pandas\u8fdb\u884c\u6570\u636e\u7f29\u653e\u53ef\u4ee5\u8f7b\u677e\u5904\u7406\u7f3a\u5931\u503c\uff0c\u5e76\u76f4\u63a5\u5728DataFrame\u4e2d\u64cd\u4f5c\u6570\u636e\uff0c\u9002\u5408\u6570\u636e\u9884\u5904\u7406\u9636\u6bb5\u3002<\/strong><\/p>\n<\/p>\n
              \u56db\u3001\u5728\u673a\u5668\u5b66\u4e60\u4e2d\u7684\u5e94\u7528<\/p>\n<\/p>\n
              \u5728\u673a\u5668\u5b66\u4e60\u4e2d\uff0c\u6570\u636e\u7f29\u653e\u662f\u4e00\u4e2a\u5fc5\u8981\u7684\u6b65\u9aa4\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u5e94\u7528\u573a\u666f\uff1a<\/p>\n<\/p>\n
                \n
              1. K-Means\u805a\u7c7b<\/strong><\/li>\n<\/ol>\n
                K-Means\u7b97\u6cd5\u5bf9\u6570\u636e\u7684\u5c3a\u5ea6\u975e\u5e38\u654f\u611f\u3002\u672a\u7ecf\u7f29\u653e\u7684\u6570\u636e\u53ef\u80fd\u5bfc\u81f4\u805a\u7c7b\u7ed3\u679c\u4e0d\u51c6\u786e\u3002\u901a\u8fc7\u7f29\u653e\uff0c\u6240\u6709\u7279\u5f81\u90fd\u80fd\u5728\u540c\u4e00\u5c3a\u5ea6\u4e0a\u8fdb\u884c\u6bd4\u8f83\uff0c\u4ece\u800c\u63d0\u9ad8\u805a\u7c7b\u7684\u51c6\u786e\u6027\u3002<\/p>\n<\/p>\n
                  \n
                1. \u652f\u6301\u5411\u91cf\u673a\uff08SVM\uff09<\/strong><\/li>\n<\/ol>\n
                  SVM\u5bf9\u7279\u5f81\u7684\u5c3a\u5ea6\u540c\u6837\u654f\u611f\u3002\u6570\u636e\u7f29\u653e\u6709\u52a9\u4e8e\u52a0\u901fSVM\u7684\u6536\u655b\uff0c\u63d0\u9ad8\u5206\u7c7b\u7684\u7cbe\u5ea6\u3002<\/p>\n<\/p>\n
                    \n
                  1. \u795e\u7ecf\u7f51\u7edc<\/strong><\/li>\n<\/ol>\n
                    \u795e\u7ecf\u7f51\u7edc\u7684\u8bad\u7ec3\u8fc7\u7a0b\u5bf9\u8f93\u5165\u6570\u636e\u7684\u5c3a\u5ea6\u4e5f\u5f88\u654f\u611f\u3002\u672a\u7ecf\u7f29\u653e\u7684\u6570\u636e\u53ef\u80fd\u5bfc\u81f4\u7f51\u7edc\u7684\u6536\u655b\u901f\u5ea6\u53d8\u6162\uff0c\u751a\u81f3\u65e0\u6cd5\u6536\u655b\u3002<\/p>\n<\/p>\n
                      \n
                    1. \u68af\u5ea6\u4e0b\u964d<\/strong><\/li>\n<\/ol>\n
                      \u5728\u4f7f\u7528\u68af\u5ea6\u4e0b\u964d\u6cd5\u4f18\u5316\u6a21\u578b\u65f6\uff0c\u6570\u636e\u7684\u5c3a\u5ea6\u4f1a\u5f71\u54cd\u6536\u655b\u901f\u5ea6\u3002\u901a\u8fc7\u7f29\u653e\uff0c\u53ef\u4ee5\u4f7f\u68af\u5ea6\u4e0b\u964d\u66f4\u7a33\u5b9a\u4e14\u6536\u655b\u66f4\u5feb\u3002<\/p>\n<\/p>\n
                      \u4e94\u3001\u603b\u7ed3<\/p>\n<\/p>\n
                      Python\u4e2d\u8fdb\u884c\u6570\u636e\u7f29\u653e\u7684\u65b9\u6cd5\u591a\u79cd\u591a\u6837\uff0c\u53ef\u4ee5\u6839\u636e\u6570\u636e\u7684\u7279\u70b9\u548c\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u65b9\u6cd5\u3002\u4f7f\u7528\u5185\u7f6e\u5e93\u9002\u5408\u5c0f\u89c4\u6a21\u6570\u636e\u7684\u5feb\u901f\u5904\u7406\uff0cNumPy\u548cPandas\u5219\u9002\u5408\u5927\u89c4\u6a21\u6570\u636e\u7684\u9ad8\u6548\u5904\u7406\u3002\u5728\u673a\u5668\u5b66\u4e60\u4e2d\uff0c\u6570\u636e\u7f29\u653e\u662f\u63d0\u9ad8\u6a21\u578b\u6027\u80fd\u7684\u91cd\u8981\u6b65\u9aa4\uff0c\u5e94\u8be5\u5728\u6570\u636e\u9884\u5904\u7406\u9636\u6bb5\u4e88\u4ee5\u91cd\u89c6\u3002\u901a\u8fc7\u5408\u7406\u7684\u6570\u636e\u7f29\u653e\uff0c\u53ef\u4ee5\u663e\u8457\u63d0\u9ad8\u6a21\u578b\u7684\u8bad\u7ec3\u901f\u5ea6\u548c\u51c6\u786e\u6027\u3002<\/p>\n<\/p>\n
                      \u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
                       \u5982\u4f55\u5728Python\u4e2d\u5b9e\u73b0\u56fe\u50cf\u7f29\u653e\uff1f<\/strong>
                      \u5728Python\u4e2d\uff0c\u53ef\u4ee5\u4f7f\u7528PIL\uff08Python Imaging Library\uff09\u6216\u5176\u5206\u652fPillow\u6765\u5b9e\u73b0\u56fe\u50cf\u7f29\u653e\u3002\u901a\u8fc7Image.resize()<\/code>\u65b9\u6cd5\uff0c\u53ef\u4ee5\u6307\u5b9a\u65b0\u7684\u5bbd\u5ea6\u548c\u9ad8\u5ea6\u6765\u8c03\u6574\u56fe\u50cf\u7684\u5927\u5c0f\u3002\u793a\u4f8b\u5982\u4e0b\uff1a<\/p>\n
                      from PIL import Image\n\n# \u6253\u5f00\u56fe\u50cf\nimage = Image.open("path_to_image.jpg")\n\n# \u7f29\u653e\u56fe\u50cf\nresized_image = image.resize((new_width, new_height))\n\n# \u4fdd\u5b58\u7f29\u653e\u540e\u7684\u56fe\u50cf\nresized_image.save("path_to_resized_image.jpg")\n<\/code><\/pre>\n
                      Python\u4e2d\u662f\u5426\u6709\u5e93\u53ef\u4ee5\u7528\u4e8e\u6570\u636e\u7f29\u653e\uff1f<\/strong>
                      \u662f\u7684\uff0cPython\u4e2d\u6709\u591a\u4e2a\u5e93\u53ef\u4ee5\u7528\u4e8e\u6570\u636e\u7f29\u653e\uff0c\u5c24\u5176\u5728\u6570\u636e\u5904\u7406\u548c\u673a\u5668\u5b66\u4e60\u9886\u57df\u3002scikit-learn<\/code>\u5e93\u63d0\u4f9b\u4e86MinMaxScaler<\/code>\u548cStandardScaler<\/code>\u7b49\u5de5\u5177\uff0c\u53ef\u4ee5\u5c06\u6570\u636e\u7f29\u653e\u5230\u7279\u5b9a\u7684\u8303\u56f4\u6216\u8005\u6807\u51c6\u5316\u6570\u636e\u3002\u4f7f\u7528\u8fd9\u4e9b\u5de5\u5177\u53ef\u4ee5\u786e\u4fdd\u6a21\u578b\u8bad\u7ec3\u65f6\u6570\u636e\u7684\u4e00\u81f4\u6027\u548c\u6709\u6548\u6027\u3002<\/p>\n
                      from sklearn.preprocessing import MinMaxScaler\n\n# \u521b\u5efa\u7f29\u653e\u5668\nscaler = MinMaxScaler()\n\n# \u5047\u8bbe\u6709\u4e00\u4e2a\u6570\u636e\u96c6\ndata = [[1, 2], [3, 4], [5, 6]]\n\n# \u7f29\u653e\u6570\u636e\nscaled_data = scaler.fit_transform(data)\n<\/code><\/pre>\n
                      \u5982\u4f55\u5728Python\u4e2d\u8fdb\u884c\u56fe\u5f62\u7f29\u653e\uff1f<\/strong>
                      \u5728\u6570\u636e\u53ef\u89c6\u5316\u65b9\u9762\uff0c\u53ef\u4ee5\u4f7f\u7528Matplotlib\u5e93\u5b9e\u73b0\u56fe\u5f62\u7684\u7f29\u653e\u3002\u901a\u8fc7\u8bbe\u7f6e\u56fe\u5f62\u7684\u5750\u6807\u8f74\u8303\u56f4\uff0c\u53ef\u4ee5\u653e\u5927\u6216\u7f29\u5c0f\u56fe\u5f62\u663e\u793a\u7684\u533a\u57df\u3002\u4f7f\u7528set_xlim()<\/code>\u548cset_ylim()<\/code>\u65b9\u6cd5\u53ef\u4ee5\u63a7\u5236X\u548cY\u8f74\u7684\u7f29\u653e\u3002<\/p>\n
                      import matplotlib.pyplot as plt\n\n# \u521b\u5efa\u6570\u636e\nx = [1, 2, 3, 4, 5]\ny = [10, 20, 25, 30, 40]\n\n# \u7ed8\u5236\u56fe\u5f62\nplt.plot(x, y)\n\n# \u8bbe\u7f6e\u5750\u6807\u8f74\u8303\u56f4\nplt.xlim(0, 6)\nplt.ylim(0, 50)\n\n# \u663e\u793a\u56fe\u5f62\nplt.show()\n<\/code><\/pre>\n
                      \u8fd9\u4e9b\u65b9\u6cd5\u548c\u5de5\u5177\u53ef\u4ee5\u5e2e\u52a9\u60a8\u5728\u4e0d\u540c\u7684\u4e0a\u4e0b\u6587\u4e2d\u5b9e\u73b0\u7f29\u653e\uff0c\u65e0\u8bba\u662f\u56fe\u50cf\u3001\u6570\u636e\u8fd8\u662f\u56fe\u5f62\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"Python\u7f29\u653e\u7684\u65b9\u6cd5\u5305\u62ec\u4f7f\u7528\u5185\u7f6e\u5e93\u8fdb\u884c\u57fa\u672c\u7f29\u653e\u3001\u5229\u7528\u7b2c\u4e09\u65b9\u5e93\u5982NumPy\u8fdb\u884c\u77e9\u9635\u7f29\u653e\u3001\u4ee5\u53ca\u4f7f\u7528Pandas\u8fdb […]","protected":false},"author":3,"featured_media":927000,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/926998"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=926998"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/926998\/revisions"}],"predecessor-version":[{"id":927001,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/926998\/revisions\/927001"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/927000"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=926998"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=926998"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=926998"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}