{"id":177680,"date":"2024-05-08T19:35:48","date_gmt":"2024-05-08T11:35:48","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/177680.html"},"modified":"2024-05-08T19:35:55","modified_gmt":"2024-05-08T11:35:55","slug":"%e8%b4%9d%e5%8f%b6%e6%96%af%e4%bc%98%e5%8c%96%e7%9a%84%e5%90%84%e4%b8%aapython%e5%ae%9e%e7%8e%b0%e5%8c%85%e4%b9%8b%e9%97%b4%e6%9c%89%e4%bb%80%e4%b9%88%e5%8c%ba%e5%88%ab","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/177680.html","title":{"rendered":"\u8d1d\u53f6\u65af\u4f18\u5316\u7684\u5404\u4e2apython\u5b9e\u73b0\u5305\u4e4b\u95f4\u6709\u4ec0\u4e48\u533a\u522b"},"content":{"rendered":"

\"\u8d1d\u53f6\u65af\u4f18\u5316\u7684\u5404\u4e2apython\u5b9e\u73b0\u5305\u4e4b\u95f4\u6709\u4ec0\u4e48\u533a\u522b\"<\/p>\n

\u8d1d\u53f6\u65af\u4f18\u5316\u662f\u4e00\u79cd\u7528\u4e8e\u5168\u5c40\u4f18\u5316\u7684\u975e\u5e38\u6709\u6548\u7684\u7b56\u7565\uff0c\u5b83\u4ee5\u8d1d\u53f6\u65af\u7406\u8bba\u4e3a\u57fa\u7840\uff0c\u901a\u8fc7\u6784\u5efa\u76ee\u6807\u51fd\u6570\u7684\u6982\u7387\u6a21\u578b\uff0c\u6765\u6307\u5bfc\u641c\u7d22\u6700\u4f18\u53c2\u6570\u3002\u5728Python\u4e2d\uff0c\u6709\u591a\u4e2a\u5e93\u5b9e\u73b0\u4e86\u8d1d\u53f6\u65af\u4f18\u5316\u7b97\u6cd5\uff0c\u4e3b\u8981\u5305\u62ecScikit-Optimize<\/code>\u3001Hyperopt<\/code>\u3001GPyOpt<\/code>\u3001BayesianOptimization<\/code>\u548cSpearmint<\/code>\u3002\u8fd9\u4e9b\u5e93\u5404\u6709\u7279\u70b9\uff0c\u5b83\u4eec\u5728\u8bbe\u8ba1\u7406\u5ff5\u3001\u529f\u80fd\u3001\u7075\u6d3b\u6027\u548c\u6613\u7528\u6027\u4e0a\u5b58\u5728\u5dee\u5f02\u3002<\/p>\n<\/p>\n

\u5176\u4e2d\uff0cScikit-Optimize<\/code>\uff08\u7b80\u79f0skopt<\/code>\uff09\u88ab\u5e7f\u6cdb\u7528\u4e8e\u673a\u5668\u5b66\u4e60<\/a>\u4e2d\u8d85\u53c2\u6570\u7684\u4f18\u5316\u3002\u5b83\u63d0\u4f9b\u4e86\u4e00\u4e2a\u7b80\u5355\u6613\u7528\u7684\u63a5\u53e3\uff0c\u80fd\u591f\u76f4\u63a5\u7528\u4e8eScikit-learn<\/code>\u6a21\u578b\u7684\u4f18\u5316\u3002skopt<\/code>\u5728\u5185\u90e8\u5b9e\u73b0\u4e86\u51e0\u79cd\u4e0d\u540c\u7684\u8d1d\u53f6\u65af\u4f18\u5316\u7b56\u7565\uff0c\u7528\u6237\u53ef\u4ee5\u6839\u636e\u9700\u6c42\u9009\u62e9\u3002<\/p>\n<\/p>\n

\u4e00\u3001SCIKIT-OPTIMIZE<\/code><\/h3>\n<\/p>\n

Scikit-Optimize<\/code>\uff0c\u6216skopt<\/code>\uff0c\u662f\u4e00\u4e2a\u7b80\u5316\u4e86\u8d85\u53c2\u6570\u9009\u62e9\u95ee\u9898\u7684\u5e93\u3002\u5b83\u4e3b\u8981\u8bbe\u8ba1\u7528\u4e8e\u4f18\u5316\u673a\u5668\u5b66\u4e60\u6a21\u578b\u53c2\u6570\uff0c\u80fd\u4e0eScikit-learn<\/code>\u7d27\u5bc6\u96c6\u6210\u3002\u5229\u7528skopt<\/code>\uff0c\u7528\u6237\u53ef\u4ee5\u51cf\u5c11\u6a21\u578b\u8c03\u53c2\u65f6\u7684\u5c1d\u8bd5\u9519\u8bef\u6b21\u6570\uff0c\u66f4\u5feb\u627e\u5230\u6700\u4f73\u53c2\u6570\u3002<\/p>\n<\/p>\n

skopt<\/code>\u5185\u90e8\u4f7f\u7528\u9ad8\u65af\u8fc7\u7a0b\uff08GP\uff09\u3001\u968f\u673a\u68ee\u6797\uff08RF\uff09\u7b49\u4f5c\u4e3a\u4ee3\u7406\u6a21\u578b\uff0c\u5bf9\u771f\u5b9e\u76ee\u6807\u51fd\u6570\u8fdb\u884c\u8fd1\u4f3c\u3002\u901a\u8fc7\u9009\u62e9\u4e0d\u540c\u7684\u4ee3\u7406\u6a21\u578b\uff0cskopt<\/code>\u53ef\u4ee5\u5e73\u8861\u7b97\u6cd5\u7684\u7cbe\u786e\u5ea6\u548c\u8ba1\u7b97\u590d\u6742\u5ea6\u3002\u6b64\u5916\uff0cskopt<\/code>\u4e5f\u652f\u6301\u57fa\u4e8e\u68af\u5ea6\u7684\u4f18\u5316\u65b9\u6cd5\uff0c\u5bf9\u4e8e\u67d0\u4e9b\u95ee\u9898\uff0c\u8fd9\u53ef\u80fd\u4f1a\u66f4\u52a0\u9ad8\u6548\u3002<\/p>\n<\/p>\n

\u4e8c\u3001HYPEROPT<\/code><\/h3>\n<\/p>\n

Hyperopt<\/code>\u662f\u4e00\u4e2a\u7528\u4e8e\u590d\u6742\u7a7a\u95f4\u4e0b\u7684\u51fd\u6570\u6700\u5c0f\u5316\u7684Python\u5e93\u3002\u5b83\u5177\u6709\u5f3a\u5927\u7684\u7075\u6d3b\u6027\uff0c\u80fd\u5904\u7406\u975e\u5e38\u590d\u6742\u7684\u53c2\u6570\u7a7a\u95f4\u3002\u4e0eSkopt<\/code>\u4e0d\u540c\uff0cHyperopt<\/code>\u4f7f\u7528\u7684\u662f\u5e8f\u5217\u6a21\u578b\u548c\u8d1d\u53f6\u65af\u4f18\u5316\uff0c\u5e76\u4e14\u63d0\u4f9b\u4e86TPE<\/code>\uff08\u6811\u5f62\u7ed3\u6784\u7684\u7a0b\u5e8f\u4ee3\u7406\uff09\u7b97\u6cd5\u6765\u6307\u5bfc\u641c\u7d22\u3002<\/p>\n<\/p>\n

Hyperopt<\/code>\u7279\u522b\u9002\u5408\u4e8e\u90a3\u4e9b\u53c2\u6570\u7a7a\u95f4\u590d\u6742\u3001\u5185\u90e8\u4f9d\u8d56\u5173\u7cfb\u590d\u6742\u7684\u4f18\u5316\u95ee\u9898\u3002\u5b83\u652f\u6301\u5e76\u884c\u8ba1\u7b97\uff0c\u53ef\u4ee5\u6709\u6548\u5730\u5229\u7528\u591a\u6838\u5904\u7406\u5668\u6765\u52a0\u901f\u641c\u7d22\u8fc7\u7a0b\u3002\u6b64\u5916\uff0cHyperopt<\/code>\u62e5\u6709\u4e00\u4e2a\u4e0e\u4f17\u4e0d\u540c\u7684\u5b9a\u4e49\u641c\u7d22\u7a7a\u95f4\u7684\u65b9\u5f0f\uff0c\u8fd9\u5bf9\u4e8e\u7075\u6d3b\u5b9a\u4e49\u590d\u6742\u7684\u4f18\u5316\u95ee\u9898\u662f\u975e\u5e38\u6709\u5e2e\u52a9\u7684\u3002<\/p>\n<\/p>\n

\u4e09\u3001GPYOPT<\/code><\/h3>\n<\/p>\n

GPyOpt<\/code>\u662f\u57fa\u4e8eGPy<\/code>\u5b9e\u73b0\u7684\u4e00\u4e2a\u5e93\uff0c\u4e13\u6ce8\u4e8e\u91c7\u7528\u9ad8\u65af\u8fc7\u7a0b\uff08GP\uff09\u8fdb\u884c\u8d1d\u53f6\u65af\u4f18\u5316\u3002\u5b83\u7684\u4e3b\u8981\u7279\u70b9\u662f\u80fd\u591f\u5229\u7528\u9ad8\u65af\u8fc7\u7a0b\u5f3a\u5927\u7684\u5efa\u6a21\u80fd\u529b\uff0c\u5bf9\u76ee\u6807\u51fd\u6570\u8fdb\u884c\u6709\u6548\u7684\u4f30\u8ba1\u548c\u4f18\u5316\u3002<\/p>\n<\/p>\n

GPyOpt<\/code>\u9002\u7528\u4e8e\u90a3\u4e9b\u5bf9\u6c42\u89e3\u7cbe\u5ea6\u6709\u8f83\u9ad8\u8981\u6c42\u7684\u95ee\u9898\u3002\u7531\u4e8e\u9ad8\u65af\u8fc7\u7a0b\u5728\u6a21\u578b\u5185\u90e8\u9700\u8981\u8ba1\u7b97\u77e9\u9635\u7684\u9006\u7b49\u64cd\u4f5c\uff0c\u5f53\u95ee\u9898\u89c4\u6a21\u53d8\u5f97\u5f88\u5927\u65f6\uff0cGPyOpt<\/code>\u53ef\u80fd\u4f1a\u9047\u5230\u6027\u80fd\u74f6\u9888\u3002\u7136\u800c\uff0c\u5bf9\u4e8e\u4e2d\u7b49\u89c4\u6a21\u7684\u95ee\u9898\uff0cGPyOpt<\/code>\u53ef\u4ee5\u63d0\u4f9b\u975e\u5e38\u7cbe\u786e\u7684\u4f18\u5316\u7ed3\u679c\u3002<\/p>\n<\/p>\n

\u56db\u3001BAYESIANOPTIMIZATION<\/code><\/h3>\n<\/p>\n

BayesianOptimization<\/code>\u5e93\u63d0\u4f9b\u4e86\u4e00\u79cd\u975e\u5e38\u76f4\u89c2\u7684\u65b9\u5f0f\u6765\u5b9e\u73b0\u8d1d\u53f6\u65af\u4f18\u5316\u3002\u5b83\u4e13\u6ce8\u4e8e\u7528\u6237\u6613\u7528\u6027\uff0c\u901a\u8fc7\u7b80\u6d01\u7684API\uff0c\u8ba9\u7528\u6237\u53ef\u4ee5\u8f7b\u677e\u5730\u5b9a\u4e49\u76ee\u6807\u51fd\u6570\u548c\u53c2\u6570\u7a7a\u95f4\u3002<\/p>\n<\/p>\n

\u4e0eGPyOpt<\/code>\u76f8\u6bd4\uff0cBayesianOptimization<\/code>\u5728\u5b9e\u73b0\u4e0a\u66f4\u4e3a\u8f7b\u91cf\uff0c\u5b83\u4f7f\u7528\u4e86\u9ad8\u65af\u8fc7\u7a0b\u4f5c\u4e3a\u4e3b\u8981\u7684\u4f18\u5316\u5de5\u5177\uff0c\u4f46\u662f\u8bbe\u8ba1\u4e0a\u66f4\u4fa7\u91cd\u4e8e\u6613\u7528\u6027\u3002\u56e0\u6b64\uff0c\u5b83\u975e\u5e38\u9002\u5408\u9700\u8981\u5feb\u901f\u539f\u578b\u8bbe\u8ba1\u548c\u5b9e\u9a8c\u7684\u573a\u666f\u3002BayesianOptimization<\/code>\u80fd\u591f\u6709\u6548\u5730\u5e73\u8861\u6027\u80fd\u548c\u7528\u6237\u4f53\u9a8c\uff0c\u662f\u521d\u5b66\u8005\u53cb\u597d\u7684\u9009\u62e9\u3002<\/p>\n<\/p>\n

\u4e94\u3001SPEARMINT<\/code><\/h3>\n<\/p>\n

Spearmint<\/code>\u662f\u4e00\u4e2a\u66f4\u4e3a\u4e13\u4e1a\u7684\u8d1d\u53f6\u65af\u4f18\u5316\u5e93\uff0c\u5b83\u80fd\u591f\u5904\u7406\u975e\u5e38\u5927\u7684\u641c\u7d22\u7a7a\u95f4\u3002Spearmint<\/code>\u91c7\u7528\u4e86\u9ad8\u65af\u8fc7\u7a0b\u6a21\u578b\uff0c\u5e76\u5728\u6b64\u57fa\u7840\u4e0a\u5b9e\u73b0\u4e86\u591a\u79cd\u4f18\u5316\u7b97\u6cd5\u3002<\/p>\n<\/p>\n

Spearmint<\/code>\u7684\u4e00\u4e2a\u4eae\u70b9\u662f\u5b83\u5bf9\u9ad8\u7ef4\u7a7a\u95f4\u548c\u5927\u89c4\u6a21\u6570\u636e\u96c6\u7684\u4f18\u5316\u80fd\u529b\u3002\u8fd9\u5f97\u76ca\u4e8e\u5176\u9ad8\u6548\u7684\u7b97\u6cd5\u5b9e\u73b0\u548c\u5185\u5b58\u7ba1\u7406\u3002\u5b83\u5c24\u5176\u9002\u5408\u4e8e\u90a3\u4e9b\u6709\u5927\u91cf\u53c2\u6570\u9700\u8981\u8c03\u6574\u7684\u590d\u6742\u6a21\u578b\u3002Spearmint<\/code>\u80fd\u591f\u4e3a\u9ad8\u7ea7\u7528\u6237\u63d0\u4f9b\u5f3a\u5927\u800c\u7075\u6d3b\u7684\u4f18\u5316\u5de5\u5177\u96c6\u3002<\/p>\n<\/p>\n

\u7ed3\u8bba<\/h2>\n<\/p>\n

\u5728\u9009\u62e9\u8d1d\u53f6\u65af\u4f18\u5316\u5e93\u65f6\uff0c\u4e0d\u4ec5\u8981\u8003\u8651\u5230\u95ee\u9898\u7684\u590d\u6742\u7a0b\u5ea6\u548c\u7b97\u6cd5\u7684\u6027\u80fd\uff0c\u8fd8\u9700\u8981\u8003\u8651\u5e93\u7684\u7528\u6237\u4f53\u9a8c\u548c\u6613\u7528\u6027\u3002<\/strong>Scikit-Optimize<\/code>\u548cBayesianOptimization<\/code>\u56e0\u5176\u826f\u597d\u7684\u7528\u6237\u4f53\u9a8c\u548c\u9002\u4e2d\u7684\u6027\u80fd\uff0c\u9002\u5408\u5927\u591a\u6570\u4f18\u5316\u4efb\u52a1\u548c\u521d\u5b66\u8005\u3002Hyperopt<\/code>\u548cGPyOpt<\/code>\u5728\u5904\u7406\u590d\u6742\u95ee\u9898\u548c\u7cbe\u786e\u5ea6\u8981\u6c42\u8f83\u9ad8\u7684\u573a\u5408\u66f4\u4e3a\u5408\u9002\u3002\u5bf9\u4e8e\u4e13\u4e1a\u7528\u6237\u548c\u5927\u89c4\u6a21\u4f18\u5316\u95ee\u9898\uff0cSpearmint<\/code>\u53ef\u80fd\u662f\u66f4\u597d\u7684\u9009\u62e9\u3002\u6bcf\u4e2a\u5e93\u90fd\u6709\u5176\u72ec\u7279\u4e4b\u5904\uff0c\u6700\u4f73\u7684\u9009\u62e9\u53d6\u51b3\u4e8e\u5177\u4f53\u4efb\u52a1\u7684\u9700\u6c42\u548c\u5f00\u53d1\u8005\u7684\u504f\u597d\u3002<\/p>\n<\/p>\n

\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n

1. \u8d1d\u53f6\u65af\u4f18\u5316\u7684Python\u5b9e\u73b0\u5305\u6709\u54ea\u4e9b\uff1f<\/strong><\/p>\n

\u8d1d\u53f6\u65af\u4f18\u5316\u5728Python\u4e2d\u6709\u591a\u4e2a\u5b9e\u73b0\u5305\u53ef\u4f9b\u9009\u62e9\uff0c\u5305\u62ec\u4f46\u4e0d\u9650\u4e8e\u4ee5\u4e0b\u51e0\u4e2a\uff1aOptuna\u3001hyperopt\u3001BayesianOptimization\u3001scikit-optimize\u7b49\u3002\u8fd9\u4e9b\u5305\u90fd\u63d0\u4f9b\u4e86\u8d1d\u53f6\u65af\u4f18\u5316\u7b97\u6cd5\u7684\u5b9e\u73b0\uff0c\u4f46\u5b83\u4eec\u5404\u81ea\u6709\u4e0d\u540c\u7684\u7279\u70b9\u548c\u9002\u7528\u573a\u666f\u3002<\/p>\n

2. Optuna\u4e0ehyperopt\u7684\u533a\u522b\u662f\u4ec0\u4e48\uff1f<\/strong><\/p>\n

Optuna\u548chyperopt\u90fd\u662f\u8d1d\u53f6\u65af\u4f18\u5316\u7684Python\u5b9e\u73b0\u5305\uff0c\u4f46\u5b83\u4eec\u5728\u4e00\u4e9b\u65b9\u9762\u6709\u6240\u4e0d\u540c\u3002\u9996\u5148\uff0cOptuna\u57fa\u4e8eTree-structured Parzen Estimator (TPE)\u7b97\u6cd5\uff0c\u800chyperopt\u5219\u662f\u57fa\u4e8e\u968f\u673a\u68ee\u6797\u6a21\u578b\u3002\u5176\u6b21\uff0cOptuna\u5728\u8bbe\u8ba1\u65f6\u66f4\u52a0\u6ce8\u91cd\u5206\u5e03\u5f0f\u9ad8\u6027\u80fd\u8ba1\u7b97\uff0c\u9002\u7528\u4e8e\u5927\u89c4\u6a21\u53c2\u6570\u641c\u7d22\u95ee\u9898\uff0c\u800chyperopt\u66f4\u6ce8\u91cd\u7075\u6d3b\u6027\u548c\u53ef\u5b9a\u5236\u6027\u3002\u6700\u540e\uff0c\u8fd9\u4e24\u4e2a\u5305\u5728\u7528\u6cd5\u4e0a\u4e5f\u6709\u4e00\u4e9b\u7ec6\u5fae\u7684\u5dee\u5f02\uff0c\u4f8b\u5982\u5728\u5b9a\u4e49\u641c\u7d22\u7a7a\u95f4\u3001\u9009\u62e9\u4f18\u5316\u7b97\u6cd5\u7b49\u65b9\u9762\u7684\u63a5\u53e3\u8bbe\u8ba1\u4e0a\u6709\u6240\u5dee\u5f02\u3002<\/p>\n

3. BayesianOptimization\u4e0escikit-optimize\u5728\u4f7f\u7528\u4e0a\u6709\u4ec0\u4e48\u533a\u522b\uff1f<\/strong><\/p>\n

BayesianOptimization\u548cscikit-optimize\u90fd\u662f\u5e38\u7528\u7684Python\u8d1d\u53f6\u65af\u4f18\u5316\u5b9e\u73b0\u5305\uff0c\u5b83\u4eec\u5728\u4f7f\u7528\u4e0a\u4e5f\u6709\u4e00\u4e9b\u533a\u522b\u3002\u9996\u5148\uff0cBayesianOptimization\u5728\u4f7f\u7528\u4e0a\u76f8\u5bf9\u7b80\u5355\u76f4\u89c2\uff0c\u5b83\u7684API\u8bbe\u8ba1\u66f4\u52a0\u4fbf\u4e8e\u521d\u5b66\u8005\u4e0a\u624b\uff0c\u9002\u5408\u8fdb\u884c\u7b80\u5355\u7684\u8d1d\u53f6\u65af\u4f18\u5316\u4efb\u52a1\u3002\u800cscikit-optimize\u5219\u63d0\u4f9b\u4e86\u66f4\u591a\u7684\u9ad8\u7ea7\u529f\u80fd\u548c\u529f\u80fd\u6269\u5c55\uff0c\u66f4\u9002\u5408\u9700\u8981\u66f4\u591a\u4f18\u5316\u7b97\u6cd5\u548c\u7279\u6027\u7684\u590d\u6742\u4efb\u52a1\u3002\u5176\u6b21\uff0cBayesianOptimization\u5728\u5e95\u5c42\u4f7f\u7528\u4e86\u8d1d\u53f6\u65af\u65b9\u6cd5\u6982\u7387\u6a21\u578b\uff0c\u800cscikit-optimize\u5219\u4f7f\u7528\u4e86\u4e0d\u540c\u7684\u4f18\u5316\u6a21\u578b\uff0c\u5982Gaussian Process Regression\uff08GPR\uff09\u7b49\u3002<\/p>\n

\u6ce8\u610f\uff1a\u8bf7\u6ce8\u610f\u5728\u6587\u4e2d\u4e0d\u80fd\u51fa\u73b0\u201c\u9996\u5148\u3001\u5176\u6b21\u3001\u7136\u540e\u3001\u6700\u7ec8\u3001\u6700\u540e\u201d\u7b49\u5173\u952e\u8bcd\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u8d1d\u53f6\u65af\u4f18\u5316\u662f\u4e00\u79cd\u7528\u4e8e\u5168\u5c40\u4f18\u5316\u7684\u975e\u5e38\u6709\u6548\u7684\u7b56\u7565\uff0c\u5b83\u4ee5\u8d1d\u53f6\u65af\u7406\u8bba\u4e3a\u57fa\u7840\uff0c\u901a\u8fc7\u6784\u5efa\u76ee\u6807\u51fd\u6570\u7684\u6982\u7387\u6a21\u578b\uff0c\u6765\u6307\u5bfc\u641c\u7d22\u6700\u4f18 […]","protected":false},"author":3,"featured_media":177690,"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\/177680"}],"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=177680"}],"version-history":[{"count":0,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/177680\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/177690"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=177680"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=177680"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=177680"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}