{"id":8639,"date":"2024-06-01T19:01:01","date_gmt":"2024-06-01T11:01:01","guid":{"rendered":""},"modified":"2024-06-01T19:01:01","modified_gmt":"2024-06-01T11:01:01","slug":"\u7279\u5f81\u5de5\u7a0b\u4e0e\u6570\u636e\u9884\u5904\u7406_Data Structure Visualization","status":"publish","type":"post","link":"https:\/\/mushiming.com\/8639.html","title":{"rendered":"\u7279\u5f81\u5de5\u7a0b\u4e0e\u6570\u636e\u9884\u5904\u7406_Data Structure Visualization"},"content":{"rendered":"
<\/p>\n
\u524d\u8a00<\/strong><\/p>\n \u8fd9\u4e24\u5e74\u6211\u4eec\u89c2\u5bdf\u5230\u8d8a\u6765\u8d8a\u591a\u7684\u7b97\u6cd5\u5de5\u7a0b\u5e08\u91cd\u89c6\u6570\u636e\u7684\u7279\u5f81\u5de5\u7a0b\uff0cAI\u4e1a\u754c\u5927\u4f6c\u5434\u6069\u8fbe\u6559\u6388\u57282021\u5e74\u63d0\u51fa\u4e86\u4ecemodel-centric AI\u5207\u6362\u5230data-centric AI\u7684\u8bba\u8c03\uff0c\u6211\u4e2a\u4eba\u8ba4\u4e3adata-centric AI\u7684\u4e09\u4e2a\u6838\u5fc3\u5c31\u662f\u7279\u5f81\u5de5\u7a0b\uff0c\u6837\u672c\u5de5\u7a0b\u548c\u6570\u636e\u96c6\u8d28\u91cf\uff08\u672c\u7cfb\u5217\u6587\u7ae0\u5c06\u56f4\u7ed5\u8fd9\u4e09\u4e2a\u6838\u5fc3\u6765\u4ecb\u7ecd\uff09\uff0c\u5373\u4f7f\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\uff0c\u5bf9\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\uff0c\u7279\u5f81\u5de5\u7a0b\u4ecd\u7136\u662f\u91cd\u4e2d\u4e4b\u91cd\uff0c\u800c\u5bf9\u4e8e\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\uff0c\u968f\u7740\u673a\u5668\u5b66\u4e60\u5e73\u53f0\u7684\u6210\u719f\u6bd4\u5982\u4e9a\u9a6c\u900a\u4e91\u79d1\u6280\u7684\u62f3\u5934\u4ea7\u54c1 Amazon SageMaker <\/strong>https:\/\/aws.amazon.com\/cn\/sagemaker\/\uff0c\u8ba9\u4eba\u4eec\u4ece\u5e95\u5c42\u57fa\u7840\u8bbe\u65bd\u4ee5\u53ca\u7e41\u6742\u7684\u8bad\u7ec3\u548c\u63a8\u7406\u7684\u5404\u79cd\u73af\u5883\u642d\u5efa\u4e2d\u89e3\u653e\u51fa\u6765\uff0c\u4ece\u800c\u6709\u66f4\u591a\u7684\u65f6\u95f4\u548c\u7cbe\u529b\u5173\u6ce8\u6570\u636e\u548c\u7279\u5f81\u672c\u8eab\uff0c\u771f\u6b63\u8fc8\u5165\u201cAI\u5e73\u6c11\u5316\u201d\uff0c\u53ea\u8981\u638c\u63e1\u4e86data-centric AI\u4e09\u4e2a\u6838\u5fc3\uff0c\u5927\u5bb6\u90fd\u53ef\u4ee5\u5229\u7528\u6210\u719f\u7684\u6a21\u578b\u6765\u5efa\u6a21\u3002\u5f88\u591a\u573a\u666f\u4e0b\uff0c\u751a\u81f3\u4f60\u53ea\u8981\u628a\u9ad8\u8d28\u91cf\u7684\u6570\u636e\u96c6\u51c6\u5907\u597d\u4ee5\u540e\uff08\u6bd4\u5982\u53ef\u4ee5\u501f\u52a9Amazon SageMaker Data Wrangler<\/strong>https:\/\/aws.amazon.com\/cn\/sagemaker\/data-wrangler\/\u6765\u505a\u6570\u636e\u96c6\u7684\u51c6\u5907\uff09\uff0c\u5c31\u53ef\u4ee5\u5229\u7528 SageMaker Canvas https:\/\/aws.amazon.com\/cn\/sagemaker\/canvas\/\u6765\u57fa\u4e8eUI\u7684code-free\u5efa\u6a21\uff0c\u4e0d\u9700\u8981\u5173\u5fc3\u7b97\u6cd5\u672c\u8eab\u7684\u540c\u65f6\u8fd8\u80fd\u5f97\u5230\u4e0d\u9519\u7684\u6548\u679c\u3002<\/p>\n \u4e3a\u4e86\u884c\u6587\u65b9\u4fbf\uff0c\u5728\u4e4b\u540e\u7684\u8ba8\u8bba\u4e2d\u6211\u4e0d\u4e13\u95e8\u63d0\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\u3002\u5bf9\u4e8e\u6d89\u53ca\u5230\u975e\u7ed3\u6784\u5316\u6570\u636e\u7684CV\uff0cNLP\uff0c\u8bed\u97f3\u8bc6\u522b\u8fd9\u4e09\u5927\u9886\u57df\u6765\u8bf4\uff0c\u73b0\u5728\uff082022\u5e74\uff09\u4e3b\u6d41\u90fd\u662f\u7528\u6df1\u5ea6\u5b66\u4e60\u6765\u5efa\u6a21\u800c\u4e0d\u8003\u8651\u590d\u6742\u7684\u4eba\u5de5\u63d0\u53d6\u7684\u7279\u5f81\u3002\u56e0\u6b64\u76ee\u524d\u6765\u8bf4\uff0c\u6211\u4eec\u8ba8\u8bba\u7684\u7279\u5f81\u5de5\u7a0b\u9488\u5bf9\u7684\u5c31\u662f\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\u3002\u66f4\u8be6\u7ec6\u7684data-centric AI\u76f8\u5173\u7684\u5185\u5bb9\u8bf7\u53c2\u8003\u8be5github repo https:\/\/github.com\/yuhuiaws\/ML-study\/tree\/main\/Data-centric%20AI\u3002<\/p>\n \u66f4<\/strong>\u591a\u6570\u636e\u5de5\u7a0b\u7cfb\u5217\u7cbe\u8bb2\u5185\u5bb9\uff0c\u656c\u8bf7\u671f\u5f85\u7cfb\u5217\u63a8\u9001<\/strong><\/p>\n \u672c\u6587\u5927\u7eb2<\/strong><\/p>\n \u4ec0\u4e48\u662f\u7279\u5f81\uff1f<\/p>\n<\/li>\n \u8fde\u7eed\u7279\u5f81\u548ccategory\u7279\u5f81<\/p>\n<\/li>\n \u4ec0\u4e48\u662f\u597d\u7684\u7279\u5f81\uff1f<\/p>\n<\/li>\n \u7279\u5f81\u5de5\u7a0b\u6982\u89c8<\/p>\n<\/li>\n \u7279\u5f81\u5de5\u7a0b\u8be6\u7ec6\u6b65\u9aa4<\/p>\n<\/li>\n \u7279\u5f81\u9884\u5904\u7406<\/p>\n<\/li>\n \u7279\u5f81\u751f\u6210<\/p>\n<\/li>\n \u7279\u5f81\u9009\u62e9<\/p>\n<\/li>\n \u7279\u5f81\u964d\u7ef4<\/p>\n<\/li>\n \u603b\u7ed3<\/p>\n<\/li>\n<\/ul>\n \u4ec0\u4e48\u662f\u7279\u5f81\uff1f<\/strong><\/p>\n \u7279\u5f81\u8fd9\u4e2a\u672f\u8bed\u6709\u5f88\u591a\u522b\u540d\uff0c\u6bd4\u5982\u5b57\u6bb5\uff0c\u5c5e\u6027\uff0c\u81ea\u53d8\u91cf\uff08\u5728\u4eca\u540e\u7684\u5185\u5bb9\u4e2d\u53ef\u80fd\u4f1a\u6df7\u7528\u8fd9\u4e9b\u522b\u540d\uff09\u7b49\u7b49\u3002\u6211\u4eec\u770b\u4e00\u4e2a\u4f8b\u5b50\uff0c\u4e0b\u8868\u4e2dPlay Cricket\u662f\u76ee\u6807\u53d8\u91cf\u4e5f\u5c31\u662f\u6211\u4eec\u5e38\u8bf4\u7684label\/ \u6807\u6ce8\uff0c\u5176\u4ed6\u7684\u53d8\u91cf\u6bd4\u5982Height\uff0cGender\u7b49\u90fd\u662f\u7279\u5f81\u3002<\/p>\n <\/p>\n \u7279\u5f81\u7684\u5927\u81f4\u7c7b\u578b\u5982\u4e0b\u8868\uff1a<\/strong><\/p>\n \u7c7b\u578b<\/strong><\/p>\n<\/td>\n \u4e3e\u4f8b<\/strong><\/p>\n<\/td>\n<\/tr>\n \u57fa\u672c\u7279\u5f81<\/p>\n<\/td>\n IP\u5730\u5740\uff0c\u6027\u522b\uff0c\u5e74\u9f84\uff0c\u5e38\u9a7b\u5730\uff0c\u624b\u673a\u5382\u5546<\/p>\n<\/td>\n<\/tr>\n \u7edf\u8ba1\u7279\u5f81<\/p>\n<\/td>\n \u6e38\u620f\u73a9\u5bb6\u6700\u8fd1\u4e00\u5468\u5185\u7684\u7d2f\u79ef\u6d88\u8d39\u91d1\u989d<\/p>\n<\/td>\n<\/tr>\n \u7ec4\u5408\u7279\u5f81<\/p>\n<\/td>\n \u7528\u6237\u6700\u8fd1\u70b9\u51fb\u8fc7\u768410\u4e2a\u5546\u54c1\u7ec4\u6210\u7684item id\u5e8f\u5217\u7279\u5f81\uff1b\u7537\u6027\u7528\u6237\u4e14\u4f7f\u7528\u82f9\u679c\u624b\u673a\u8fd9\u6837\u7684\u4ea4\u53c9\u7279\u5f81<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n \u7279\u5f81\u67092\u4e2a\u7edf\u8ba1\u7279\u70b9\uff1a<\/strong><\/p>\n \u4e00\u662f\u7279\u5f81\u662f\u5426\u53d1\u6563\uff0c\u5982\u679c\u4e00\u4e2a\u7279\u5f81\u7684\u65b9\u5dee\u63a5\u8fd1\u4e8e0\uff0c\u6211\u4eec\u79f0\u8fd9\u4e2a\u7279\u5f81\u4e0d\u53d1\u6563\u3002\u4e5f\u5c31\u662f\u8bf4\u6240\u6709\u6837\u672c\u5728\u8fd9\u4e2a\u7279\u5f81\u4e0a\u57fa\u672c\u6ca1\u6709\u5dee\u5f02\uff0c\u8fd9\u4e2a\u7279\u5f81\u5bf9\u4e8e\u6837\u672c\u7684\u533a\u5206\u6ca1\u6709\u4ec0\u4e48\u7528\uff0c\u53ef\u4ee5\u4e22\u5f03\u5b83\uff08\u6bd4\u59821\u4e07\u6761\u6837\u672c\u4e2d\u7684\u201c\u6027\u522b\u201d\u8fd9\u4e2a\u7279\u5f81\u7684\u53d6\u503c\u90fd\u662f\u201c\u7537\u6027\u201d\uff0c\u8fd9\u65f6\u5019\u6211\u4eec\u8bf4\u201c\u6027\u522b\u201d\u8fd9\u4e2a\u7279\u5f81\u7684\u65b9\u5dee\u5c31\u662f0. \u6bd4\u59821\u4e07\u6761\u6837\u672c\u4e2d\u7684\u201c\u56fd\u5bb6\u201d\u8fd9\u4e2a\u7279\u5f81\u7684\u53d6\u503c\u67099900\u90fd\u662f\u201c\u4e2d\u56fd\u201d\uff0c\u5176\u4ed6\u56fd\u5bb6\u4e00\u5171\u7684\u6837\u672c\u6570\u624d100\u6761\uff0c\u8fd9\u4e2a\u65f6\u5019\u6211\u4eec\u8bf4\u201c\u56fd\u5bb6\u201d\u8fd9\u4e2a\u7279\u5f81\u7684\u65b9\u5dee\u63a5\u8fd1\u4e8e0\uff09<\/p>\n \u4e8c\u662f\u7279\u5f81\u4e0e\u76ee\u6807\u7684\u76f8\u5173\u6027\uff0c\u4e0e\u76ee\u6807\u76f8\u5173\u6027\u9ad8\u7684\u7279\u5f81\uff08\u5305\u62ec\u6b63\u76f8\u5173\u548c\u8d1f\u76f8\u5173\uff09\uff0c\u5e94\u5f53\u4f18\u9009\u9009\u62e9\u3002\uff08\u6bd4\u5982\u9884\u6d4b\u8bd5\u7528\u4ea7\u54c1\u7684\u7528\u6237\u662f\u5426\u5728\u4e00\u4e2a\u6708\u5185\u4ed8\u8d39\u7684\u76ee\u6807\u4efb\u52a1\uff0c\u201c\u7528\u6237\u7684\u7d2f\u79ef\u767b\u5f55\u6b21\u6570\u201d\u8fd9\u6837\u7684\u7279\u5f81\u53ef\u80fd\u5c31\u662f\u4e0e\u76ee\u6807\u76f8\u5173\u6027\u9ad8\u7684\u7279\u5f81\uff09<\/p>\n \u8fde\u7eed\u7279\u5f81\u548ccategory\u7279\u5f81<\/strong><\/p>\n \u5728\u505a\u7279\u5f81\u5de5\u7a0b\u7684\u65f6\u5019\uff0c\u6211\u4eec\u8981\u65f6\u523b\u6ce8\u610f\u6bcf\u4e2a\u7279\u5f81\u5230\u5e95\u662f\u8fde\u7eed\u7279\u5f81\u8fd8\u662fcategory\u7279\u5f81\uff0c\u8fd9\u4e2a\u975e\u5e38\u91cd\u8981\u3002Category\u7279\u5f81\u6307\u8bed\u4e49\u4e0a\u6709\u679a\u4e3e\u7279\u70b9\u7684\u7279\u5f81\uff0c\u5305\u62ec\u5b57\u7b26\u4e32\u7c7b\u578b\u548c\u6570\u503c\u7c7b\u578b\u3002\u6bd4\u5982[\u201c\u5317\u4eac\u201d\uff0c\u201c\u4e0a\u6d77\u201d\uff0c\u201c\u6df1\u5733\u201d]\u8fd9\u6837\u7684\u201c\u57ce\u5e02\u201d\u7279\u5f81\u5c31\u662f\u5b57\u7b26\u4e32\u7c7b\u578bcategory\u7279\u5f81\uff1b\u800c[\u201c\u4e2d\u56fd\u201d\uff0c\u201c\u7f8e\u56fd\u201d\uff0c\u201c\u65e5\u672c\u201d]\u8fd9\u6837\u7684\u7279\u5f81\u7ecf\u8fc7\u6570\u5b57\u7f16\u7801\u4e3a[0\uff0c1\uff0c2]\u540e\u5f97\u5230\u7684\u201c\u56fd\u5bb6\u201d\u7279\u5f81\u5c31\u662f\u6570\u503c\u7c7b\u578b\u7684category\u7279\u5f81\u3002\u79bb\u6563\u7279\u5f81\u662f\u6307\u5177\u6709\u6570\u503c\u7c7b\u578b\u7684Category\u7279\u5f81\u3002Category\u7279\u5f81\u7684\u679a\u4e3e\u503c\u6570\u91cf\u8d8a\u591a\uff0c\u6211\u4eec\u79f0\u5b83\u7684\u57fa\u6570\u8d8a\u9ad8\uff08\u4ee5\u540e\u4e3a\u4e86\u884c\u6587\u65b9\u4fbf\uff0c\u4e0d\u518d\u533a\u5206\u79bb\u6563\u7279\u5f81\u548cCategory\u7279\u5f81\uff09\uff1b\u8fde\u7eed\u7279\u5f81\u6307\u8bed\u4e49\u4e0a\u8fde\u7eed\u7684\u6570\u503c\u578b\u7279\u5f81\uff0c\u5305\u62ec\u6574\u6570\u7c7b\u578b\u548c\u6d6e\u70b9\u7c7b\u578b\u3002\u9700\u8981\u6ce8\u610f\u7684\u662f\uff0c\u6570\u503c\u578b\u7279\u5f81\u4e0d\u4e00\u5b9a\u5c31\u662f\u8fde\u7eed\u7279\u5f81\u3002\uff08\u6bd4\u5982\u201c\u8eab\u9ad8\u201d\uff0c\u201c\u4f53\u91cd\u201d\u8fd9\u6837\u7684\u7279\u5f81\u90fd\u662f\u8fde\u7eed\u7279\u5f81\uff1b\u800c\u5982\u4e0a\u9762\u63d0\u5230\u7684\u7528\u6570\u5b57\u7f16\u7801\u8fc7\u7684\u201c\u56fd\u5bb6\u201d\u8fd9\u6837\u7684\u7279\u5f81\u5c31\u4e0d\u662f\u8fde\u7eed\u7279\u5f81\uff09<\/p>\n \u5bf9\u4e8eCategory\u7279\u5f81\u6765\u8bf4\uff0c\u6211\u4eec\u8981\u5224\u65ad\u5b83\u662f\u65e0\u5e8f\u8fd8\u662f\u6709\u5e8f\u3002\u6709\u5e8f\u548c\u65e0\u5e8f\u53ea\u662f\u76f8\u5bf9\u4e8e\u5f53\u524d\u7684\u76ee\u6807\u4efb\u52a1\u3002\u6bd4\u5982\u4ece\u5c0f\u5230\u5927\u6392\u5217\u7684\u201c\u5b66\u5386\u7b49\u7ea7\u201d\u7279\u5f81[\u201c\u5c0f\u5b66\u201d\uff0c\u201c\u521d\u4e2d\u201d\uff0c\u201c\u9ad8\u4e2d\u201d\uff0c\u201c\u5b66\u58eb\u201d\uff0c\u201c\u7855\u58eb\u201d\uff0c\u201c\u535a\u58eb\u201d]\uff0c\u5728\u9884\u6d4b\u6536\u5165\u8fd9\u6837\u7684\u4efb\u52a1\u7684\u65f6\u5019\u53ef\u80fd\u662f\u6709\u5e8f\u7279\u5f81\uff08\u6709\u5e8f\u7684\u542b\u4e49\u5728\u8fd9\u91cc\u7684\u4f53\u73b0\u662f\uff1a\u53ef\u80fd\u5b66\u5386\u8d8a\u9ad8\uff0c\u6536\u5165\u8d8a\u9ad8\uff09\uff0c\u800c\u5728\u9884\u6d4b\u8eab\u4f53\u5065\u5eb7\u7a0b\u5ea6\u8fd9\u6837\u7684\u4efb\u52a1\u7684\u65f6\u5019\u53ef\u80fd\u5c31\u662f\u65e0\u5e8f\u7279\u5f81\uff08\u65e0\u5e8f\u7684\u542b\u4e49\u5728\u8fd9\u91cc\u7684\u4f53\u73b0\u662f\uff1a\u662f\u5426\u5065\u5eb7\u4e0e\u5b66\u5386\u9ad8\u4f4e\u6ca1\u6709\u4ec0\u4e48\u5173\u7cfb\uff09\u3002<\/p>\n \u7279\u5f81\u5de5\u7a0b\u9636\u6bb5\u4e0d\u540c\u5b50\u6b65\u9aa4\u7684\u4e0d\u540c\u65b9\u6cd5\u5bf9category\u7279\u5f81\u7684\u9002\u7528\u6027\u662f\u4e0d\u540c\u7684\uff08\u6bd4\u5982\u7279\u5f81\u7f29\u653e\u4e3b\u8981\u662f\u9488\u5bf9\u8fde\u7eed\u6027\u7279\u5f81\uff1b\u6bd4\u5982\u7279\u5f81\u964d\u7ef4\u7684PCA\u4e3b\u6210\u5206\u5206\u6790\u7b97\u6cd5\u53ea\u662f\u9488\u5bf9\u8fde\u7eed\u6027\u7279\u5f81\uff09\u3002\u4e0d\u540c\u7684\u673a\u5668\u5b66\u4e60\u6a21\u578b\u5bf9\u8fde\u7eed\u7279\u5f81\u548ccategory\u7279\u5f81\u7684\u9002\u7528\u6027\u662f\u4e0d\u540c\u7684\uff08\u6bd4\u5982GMM\u9ad8\u65af\u6df7\u5408\u6a21\u578b\u53ea\u80fd\u4f7f\u7528\u8fde\u7eed\u7279\u5f81\uff0c\u800c\u591a\u9879\u5f0f\u5206\u5e03\u6734\u7d20\u8d1d\u53f6\u65af\u6a21\u578b\u53ea\u80fd\u4f7f\u7528category\u7279\u5f81\uff1b\u6bd4\u5982XGBoost https:\/\/xgboost.readthedocs.io\/en\/stable\/\u6a21\u578b\u628a\u6240\u6709\u7279\u5f81\u90fd\u5f53\u6210\u8fde\u7eed\u7279\u5f81\uff09\u3002<\/p>\n \u4e0d\u540c\u7684\u673a\u5668\u5b66\u4e60\u6a21\u578b\u5bf9category\u7279\u5f81\u7684\u65e0\u5e8f\u548c\u6709\u5e8f\u7684\u9002\u7528\u6027\u4e5f\u662f\u4e0d\u540c\u7684\u3002\uff08\u6bd4\u5982\u903b\u8f91\u56de\u5f52\u6a21\u578bLR\uff0c\u5982\u679ccategory\u7279\u5f81\u7684\u57fa\u6570\u4e0d\u662f\u5f88\u9ad8\uff08\u6bd4\u5982\u5c0f\u4e8e10\u7684\u60c5\u51b5\uff09\uff0c\u4e00\u4e2a\u53ef\u884c\u7684\u505a\u6cd5\u662f\u628a\u5355\u4e2acategory\u7279\u5f81\u53d8\u6210\u7531\u591a\u4e2a\u4e8c\u503c\u7279\u5f81\u7ec4\u6210\u7684One-hot\u5411\u91cf\uff08\u5982\u4e0b\u56fe\u6240\u793a\uff09\u3002<\/p>\n <\/p>\n \u6bd4\u5982LightGBM https:\/\/lightgbm.readthedocs.io\/en\/latest\/\u53ef\u4ee5\u5904\u7406category\u7279\u5f81\u5e76\u4e14\u5b83\u4e0d\u8003\u8651\u8be5category\u7279\u5f81\u662f\u6709\u5e8f\u8fd8\u662f\u65e0\u5e8f\uff0c\u56e0\u4e3aLightGBM\u5728\u505a\u6811\u7684\u8282\u70b9\u5206\u88c2\u7684\u65f6\u5019\uff0c\u5bf9\u4e8ecategory\u7279\u5f81\u7684\u5019\u9009\u5206\u88c2\u70b9\u8ba1\u7b97\u8fc7\u7a0b\u4e0d\u6d89\u53ca\u6392\u5e8f\uff09<\/p>\n \u4ec0\u4e48\u662f\u597d\u7684\u7279\u5f81\uff1f<\/strong><\/p>\n \u6211\u4eec\u7684\u7f8e\u597d\u613f\u671b\u662f\u9001\u5230\u6a21\u578b\u4e2d\u7684\u7279\u5f81\u662f\u6211\u4eec\u201c\u60f3\u8981\u201d\u7684\u7279\u5f81\uff0c\u8fd9\u6837\u7684\u7279\u5f81\u7684\u7279\u8d28\u5982\u4e0b\uff1a<\/p>\n \u7279\u8d28<\/strong><\/p>\n<\/td>\n \u4e3e\u4f8b<\/strong><\/p>\n<\/td>\n \u5907\u6ce8<\/strong><\/p>\n<\/td>\n<\/tr>\n \u5bf9\u5f53\u524d\u4efb\u52a1\u201c\u6709\u7528\u201d\u7684\u7279\u5f81<\/p>\n<\/td>\n \u201capp\u7248\u672c\u201d\u8fd9\u4e2a\u7279\u5f81\u5bf9\u4e8e\u9884\u6d4b\u7528\u6237\u662f\u5426\u4ed8\u8d39\u7684\u4efb\u52a1\u5c31\u5f88\u6709\u7528\uff1b\u201c\u7528\u6237\u662f\u5426\u559c\u6b22\u8d2d\u7269\u201d\u8fd9\u4e2a\u7279\u5f81\u5bf9\u9884\u6d4b\u7528\u6237\u662f\u5426\u4f1a\u62a5\u540d\u9a6c\u62c9\u677e\u8fd9\u6837\u7684\u4efb\u52a1\u53ef\u80fd\u5c31\u6ca1\u6709\u7528\u3002<\/p>\n<\/td>\n \u7279\u5f81\u7684\u597d\u574f\u4e00\u5b9a\u8981\u653e\u5728\u5177\u4f53\u7684\u4e0a\u4e0b\u6587\u4e2d\u6765\u770b\u3002<\/p>\n<\/td>\n<\/tr>\n \u597d\u7684category\u7279\u5f81\u7684\u679a\u4e3e\u503c\u5e94\u5728\u6570\u636e\u96c6\u4e2d\u51fa\u73b0\u591a\u6b21<\/p>\n<\/td>\n \u6bd4\u59828\u6b21\u4ee5\u4e0a<\/p>\n<\/td>\n \u67d0\u4e2a\u7279\u5f81\u503c\u51fa\u73b0\u6b21\u6570\u592a\u5c11\u7684\u8bdd\uff0c\u6a21\u578b\u5bf9\u8be5\u7279\u5f81\u503c\u5c31\u51e0\u4e4e\u5b66\u4e60\u4e0d\u5230\u4ec0\u4e48\u77e5\u8bc6\u3002<\/p>\n<\/td>\n<\/tr>\n \u6700\u597d\u5177\u6709\u6e05\u6670\u660e\u786e\u7684\u542b\u4e49<\/p>\n<\/td>\n \u201c\u5546\u54c1\u6700\u8fd1\u4e00\u5929\u7684\u70b9\u51fb\u7387\u201d\u8fd9\u6837\u7684\u7279\u5f81\u542b\u4e49\u5c31\u5f88\u6e05\u6670\uff1b\u800c\u201c\u5546\u54c1day1\u70b9\u51fb\u7387\u201d\u8fd9\u6837\u7684\u7279\u5f81\u5c31\u6709\u6b67\u4e49\u4e86\uff0c\u662f\u8868\u793a\u7b2c\u4e00\u5929\u7684\u70b9\u51fb\u7387\u8fd8\u662f\u6700\u8fd1\u4e00\u5929\u7684\u70b9\u51fb\u7387\u8fd8\u662f\u5f53\u5929\u7684\u70b9\u51fb\u7387\uff1f<\/p>\n<\/td>\n \u6700\u597d\u80fd\u89c1\u6587\u8bc6\u610f<\/p>\n<\/td>\n<\/tr>\n \u7279\u5f81\u7684\u503c\u4e0d\u5e94\u8be5\u8d85\u51fa\u8bed\u4e49\u8303\u56f4\u6216\u8005\u4e1a\u52a1\u77e5\u8bc6\u8303\u56f4<\/p>\n<\/td>\n \u201c\u5b63\u8282\u201d\u7279\u5f81\u4e0d\u5e94\u8be5\u51fa\u73b0\u9664\u4e86\u6625\u590f\u79cb\u51ac\u5916\u7684\u5176\u4ed6\u5b63\u8282\uff1b\u201c\u5929\u6c14\u6c14\u6e29\u201d\u7279\u5f81\u4e0d\u5e94\u8be5\u6709\u592a\u79bb\u8c31\u7684\u503c\u6bd4\u5982100\u6444\u6c0f\u5ea6\u51fa\u73b0\u3002<\/p>\n<\/td>\n \u4e3a\u4e86\u66f4\u597d\u7684\u6a21\u578b\u6548\u679c\uff0c\u8fd9\u4e2a\u4e00\u5b9a\u8981\u5728\u8bad\u7ec3\u65f6\u548c\u63a8\u7406\u65f6\u90fd\u8981\u68c0\u67e5\u5e76\u5904\u7406\u7684<\/p>\n<\/td>\n<\/tr>\n \u4e0e\u76ee\u6807\u53d8\u91cf\u76f8\u5173\u6027\u9ad8\u7684\u7279\u5f81\uff08\u5305\u62ec\u6b63\u76f8\u5173\u548c\u8d1f\u76f8\u5173\uff09<\/p>\n<\/td>\n \u201c\u7528\u6237\u6d88\u8d39\u7b49\u7ea7\u201d\u7279\u5f81\u5bf9\u4e8e\u7528\u6237\u6d41\u5931\u9884\u6d4b\u4efb\u52a1\u53ef\u80fd\u5c31\u662f\u5f3a\u7279\u5f81<\/p>\n<\/td>\n \u7ecf\u5e38\u53eb\u8fd9\u6837\u7684\u7279\u5f81\u662f\u201c \u5f3a\u7279\u5f81\u201c<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n \u7279\u5f81\u5de5\u7a0b\u6982\u89c8<\/strong><\/p>\n \u7279\u5f81\u5de5\u7a0b\u672c\u8d28\u662f\u4e00\u9879\u5de5\u7a0b\u6d3b\u52a8\uff0c\u5b83\u76ee\u7684\u662f\u6700\u5927\u9650\u5ea6\u5730\u4ece\u539f\u59cb\u6570\u636e\u4e2d\u63d0\u53d6\u5e76\u52a0\u5de5\u7279\u5f81\u4ee5\u4f9b\u6a21\u578b\u6216\u8005\u7b97\u6cd5\u4f7f\u7528\u3002\u5728\u4f20\u7edf\u673a\u5668\u5b66\u4e60\u9886\u57df\u6d41\u4f20\u7740\u8fd9\u6837\u4e00\u53e5\u8bdd\uff1a\u201c\u6570\u636e\u548c\u7279\u5f81\u51b3\u5b9a\u4e86\u673a\u5668\u5b66\u4e60\u7684\u4e0a\u9650\uff0c\u800c\u6a21\u578b\u548c\u7b97\u6cd5\u53ea\u662f\u903c\u8fd1\u8fd9\u4e2a\u4e0a\u9650\u800c\u5df2\u201d\uff0c\u4ece\u800c\u53ef\u89c1\u7279\u5f81\u5de5\u7a0b\u7684\u91cd\u8981\u6027\u3002\u5176\u5b9e\u5bf9\u4e8e\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\uff0c\u5373\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u7279\u5f81\u5de5\u7a0b\u4e5f\u662f\u6bd4\u6a21\u578b\u672c\u8eab\u8981\u91cd\u8981\u7684\u3002<\/p>\n \u7279\u5f81\u5de5\u7a0b\u76ee\u524d\u4e3b\u8981\u662f\u7eaf\u624b\u5de5\u6253\u9020\uff08\u6307\u7684\u662f\u7ed3\u6784\u5316\u6570\u636e\u5efa\u6a21\uff09\uff0c\u56e0\u4e3a\u7279\u5f81\u5de5\u7a0b\u662f\u827a\u672f\uff0c\u51e0\u4e4e\u4e0d\u53ef\u80fd\u5168\u90e8\u81ea\u52a8\u5316\u3002\u4f46\u662f\u7279\u5f81\u5de5\u7a0b\u7684\u67d0\u4e2a\u90e8\u5206\u53ef\u4ee5\u81ea\u52a8\u5316\u6bd4\u5982\u81ea\u52a8\u751f\u6210\u7279\u5f81\uff1a\u76ee\u524d\uff08\u52302022\u5e745\u6708\uff09\u7279\u5f81\u751f\u6210\u81ea\u52a8\u5316\u4e0d\u662f\u4e3b\u6d41\uff1b\u7b2c\u56db\u8303\u5f0f\u5728\u5c1d\u8bd5\u4e00\u4e9b\u7279\u5f81\u5de5\u7a0b\u81ea\u52a8\u5316\uff0c\u5305\u62ec\uff1a\u8fde\u7eed\u7279\u5f81\u81ea\u52a8\u79bb\u6563\u5316\uff0c \u65f6\u5e8f\u7279\u5f81\u81ea\u52a8\u63d0\u53d6\uff0c\u7279\u5f81\u4ea4\u53c9\u81ea\u52a8\u751f\u6210\u3002<\/p>\n \u4e0b\u9762\u6211\u4eec\u6765\u4e86\u89e3\u7279\u5f81\u5de5\u7a0b\u7684\u8be6\u7ec6\u6b65\u9aa4\u3002<\/p>\n 1<\/strong><\/p>\n \u6b65\u9aa4\u4e00: \u7279\u5f81\u9884\u5904\u7406\u4e4b\u5b57\u7b26\u4e32\u7279\u5f81\u8f6c\u6362\u4e3a\u6570\u5b57\u7279\u5f81<\/strong><\/p>\n \u6211\u4eec\u62ff\u5230\u7684\u539f\u59cb\u6570\u636e\u4e2d\u7ecf\u5e38\u4f1a\u6709\u4e00\u4e9b\u5b57\u6bb5\u662f\u5b57\u7b26\u4e32\u7c7b\u578b\u7684\uff0c\u8fd9\u4e2a\u65f6\u5019\u6211\u4eec\u9700\u8981\u9996\u5148\u6839\u636e\u5177\u4f53\u673a\u5668\u5b66\u4e60\u4efb\u52a1\u6765\u5224\u5b9a\u8fd9\u4e2a\u5b57\u7b26\u4e32\u7279\u5f81\u672c\u8d28\u4e0a\u662fcategory\u7279\u5f81\u8fd8\u662f\u6587\u672c\u8bed\u4e49\u7279\u5f81\uff1f\u6bd4\u5982\u5bf9\u4e8e\u7535\u5f71\u63a8\u8350\u573a\u666f\u7684CTR\u70b9\u51fb\u7387\u9884\u4f30\u4efb\u52a1\uff0c\u7535\u5f71\u5185\u5bb9\u7684\u6587\u672c\u63cf\u8ff0\u7279\u5f81\u662f\u5f62\u5982\u201c\u8fd9\u4e2a\u5f71\u7247\u5448\u73b0\u4e86\u4e2d\u56fd60\u5e74\u4ee3\u77e5\u9752\u4e0b\u4e61\u7684\u5404\u79cd\u666f\u8c61\u201d\uff0c\u8be5\u7279\u5f81\u662f\u5bf9\u7535\u5f71\u5185\u5bb9\u672c\u8eab\u7684\u523b\u753b\uff0c\u5c5e\u4e8e\u6587\u672c\u8bed\u4e49\u7279\u5f81\uff1b\u7535\u5f71\u51fa\u4ea7\u5730\u7684\u6587\u672c\u63cf\u8ff0\u7279\u5f81\u5f62\u5982\u201c\u4e2d\u56fd\u5317\u5f71\u201d\uff0c\u201c\u7f8e\u56fd\u597d\u83b1\u575e\u201d\u7b49\u8fd9\u4e9b\u662f\u5177\u6709\u679a\u4e3e\u7279\u70b9\u7684\uff0c\u5c5e\u4e8ecategory\u7279\u5f81\u3002<\/p>\n \u5982\u679c\u5224\u5b9a\u4e3a\u6587\u672c\u8bed\u4e49\u7279\u5f81\u5e76\u4e14\u8be5\u7279\u5f81\u5bf9\u5f53\u524d\u76ee\u6807\u4efb\u52a1\u6709\u7528\u7684\u8bdd\uff0c\u4e00\u79cd\u505a\u6cd5\u662f\u7528\u53e6\u4e00\u4e2a\u9884\u8bad\u7ec3\u7684\u8bed\u8a00\u6a21\u578b\u6765\u83b7\u5f97\u8fd9\u4e2a\u6587\u672c\u7279\u5f81\u7684sentence embedding\u5411\u91cf\uff08embedding\u5411\u91cf\u53ef\u4ee5\u7b80\u5355\u7406\u89e3\u4e3a\u539f\u59cb\u7279\u5f81\u6620\u5c04\u5230\u53e6\u4e00\u4e2a\u7a7a\u95f4\u7684\u4e0d\u53ef\u89e3\u91ca\u7684\u67d0\u79cd\u8868\u793a\uff09\uff0c\u7136\u540e\u548c\u5176\u4ed6\u7279\u5f81\u62fc\u63a5\u4f5c\u4e3a\u4e0b\u6e38\u6a21\u578b\u7684\u7279\u5f81\u5411\u91cf\u3002\u5982\u679c\u5224\u5b9a\u4e3acategory\u7279\u5f81\uff0c\u9700\u8981\u628a\u8fd9\u4e2a\u5b57\u7b26\u4e32category\u7279\u5f81\u8f6c\u4e3a\u6570\u503c\u578bcategory\u7279\u5f81\u3002<\/p>\n \u53c2\u8003\u5982\u4e0b\u7684\u6d41\u7a0b\u5bf9\u5b57\u7b26\u4e32category\u7279\u5f81\u505a\u8f6c\u6362\uff1a\u9996\u5148\u7ed3\u5408\u4e1a\u52a1\u77e5\u8bc6\u8003\u8651\u662f\u5426\u4e22\u5f03\u5bf9\u76ee\u6807\u4efb\u52a1\u6ca1\u6709\u610f\u4e49\u7684\u5b57\u7b26\u4e32category\u7279\u5f81\uff08\u6bd4\u5982\u5bf9\u4e8e\u4fe1\u7528\u5361\u6b3a\u8bc8\u68c0\u6d4b\u4efb\u52a1\u6765\u8bf4\uff0c\u201c\u59d3\u540d\u201dcategory\u7279\u5f81\u5bf9\u8fd9\u4e2a\u4efb\u52a1\u6ca1\u6709\u4ec0\u4e48\u610f\u4e49\uff0c\u53ef\u4ee5\u76f4\u63a5\u4e22\u5f03\u3002\u6bd4\u5982\u5bf9\u4e8e\u97f3\u4e50\u6b4c\u66f2\u63a8\u8350\u573a\u666f\u6765\u8bf4\uff0c\u201c\u6b4c\u66f2\u7684\u6f14\u5531\u8005\u201dcategory\u7279\u5f81\u5c31\u4e0d\u5e94\u8be5\u4e22\u5f03\uff0c\u56e0\u4e3a\u7528\u6237\u53ef\u80fd\u5c31\u662f\u5bf9\u6f14\u5531\u8005\u611f\u5174\u8da3\u624d\u64ad\u653e\u4ed6\u5531\u7684\u6b4c\u66f2\uff09\uff1b\u7136\u540e\u6216\u8005\u628a\u5b83\u8f6c\u6362\u4e3a\u5177\u6709\u7269\u7406\u610f\u4e49\u7684\u6570\u5b57\uff08\u6bd4\u5982\u7c7b\u4f3c\u5c0f\u533a\u95e8\u724c\u53f7\u4fe1\u606f\u8fd9\u6837\u7684\u5730\u5740\u7279\u5f81\u8f6c\u6362\u4e3a\u7ecf\u7eac\u5ea62\u4e2a\u7279\u5f81\uff09\uff1b\u6216\u8005\u5c06\u5355\u4e2a\u5b57\u7b26\u4e32\u7279\u5f81\u62c6\u4e3a\u591a\u4e2a\u5b57\u7b26\u4e32\u7279\u5f81\u5e76\u7f16\u7801\u4e3a\u6570\u5b57\u503c\uff08\u6bd4\u5982\u628a\u901a\u4fe1\u5730\u5740\u7279\u5f81\u6216\u8005IP\u7279\u5f81\u53ef\u4ee5\u62c6\u5206\/\u8f6c\u6362\u4e3a\u56fd\u5bb6\uff0c\u7701\uff0c\u5e02\u4e09\u4e2a\u7279\u5f81\uff09\uff1b\u6216\u8005\u5bf9\u67d0\u4e2a\u5b57\u7b26\u4e32category\u7279\u5f81\u7684\u8bad\u7ec3\u96c6\u4e2d\u51fa\u73b0\u7684\u6240\u6709\u5b57\u7b26\u4e32\u5efa\u7acb\u4e00\u4e2a\u6620\u5c04\u8868\u5e76\u7d22\u5f15\u5bf9\u5e94\u7684\u5b57\u7b26\u4e32\uff08\u4e3a\u4e86\u5b8c\u6574\uff0c\u9700\u8981\u6709\u4e00\u4e2a\u5360\u4f4d\u7d22\u5f15\u6765\u8868\u793a\u4e4b\u540e\u7684\u6570\u636e\u4e2d\u770b\u5230\u7684\u4e0d\u5728\u6620\u5c04\u8868\u4e2d\u7684\u5b57\u7b26\u4e32\uff09<\/p>\n 2<\/strong><\/p>\n \u6b65\u9aa4\u4e8c: \u7279\u5f81\u9884\u5904\u7406\u4e4b\u5f02\u5e38\u503c\u5904\u7406<\/strong><\/p>\n \u6211\u4eec\u62ff\u5230\u7684\u539f\u59cb\u6570\u636e\u5f88\u53ef\u80fd\u6216\u591a\u6216\u5c11\u90fd\u4f1a\u6709\u4e00\u4e9b\u7279\u5f81\u7684\u53d6\u503c\u4e0d\u592a\u5408\u7406\uff0c\u4e5f\u5c31\u662f\u8bf4\u8fd9\u4e9b\u7279\u5f81\u503c\u53ef\u80fd\u662f\u5f02\u5e38\u503c\u3002\u5f02\u5e38\u70b9\uff08\u53c8\u53eb\u79bb\u7fa4\u70b9\uff09\u4ea7\u751f\u539f\u56e0\uff1a\u4e00\u79cd\u662f\u4eba\u4e3a\u9519\u8bef\uff08\u4eba\u5de5\u5f55\u5165\u6570\u636e\u6216\u8005\u4eba\u5de5\u6574\u7406\u6570\u636e\u51fa\u73b0\u4e86\u95ee\u9898\uff09\uff1b\u4e00\u79cd\u662f\u81ea\u7136\u539f\u56e0\uff08\u67d0\u4e9b\u6570\u636e\u6837\u672c\u672c\u8eab\u5c31\u548c\u5927\u591a\u6570\u6570\u636e\u6837\u672c\u4e0d\u540c\u6216\u8005\u91c7\u96c6\u6570\u636e\u7684\u8bbe\u5907\/\u670d\u52a1\u7684\u95ee\u9898\u5bfc\u81f4\uff09<\/p>\n \u5bf9\u4e8e\u975e\u5f02\u5e38\u68c0\u6d4b\u4efb\u52a1\uff0c\u673a\u5668\u5b66\u4e60\u6a21\u578b\u5e0c\u671b\u6570\u636e\u96c6\uff08\u5305\u62ec\u8bad\u7ec3\u96c6\uff0c\u6d4b\u8bd5\u96c6\uff0c\u9a8c\u8bc1\u96c6\uff09\u90fd\u662f\u6b63\u5e38\u7684\u6837\u672c\u70b9\u3002\u5904\u7406\u7684\u6d41\u7a0b\u5982\u4e0b\uff1a\u9996\u5148\u9700\u8981\u628a\u5f02\u5e38\u68c0\u6d4b\u4efb\u52a1\u4f5c\u4e3a\u975e\u5f02\u5e38\u68c0\u6d4b\u4efb\u52a1\u7684\u524d\u7f6e\u4efb\u52a1\u6765\u505a\u6570\u636e\u9884\u5904\u7406\uff0c\u627e\u5230\u5e76\u4eba\u5de5\u786e\u8ba4\u5f02\u5e38\u70b9\u3002\uff08\u5f02\u5e38\u68c0\u6d4b\u7684\u65b9\u6cd5\u6709\u5f88\u591a\u6bd4\u5982\u57fa\u4e8e\u89c4\u5219\u7684\uff0c\u57fa\u4e8e\u7edf\u8ba1\u7684\uff0c\u57fa\u4e8e\u6a21\u578b\u7684\u7b49\u7b49\uff0c\u53ef\u4ee5\u53c2\u8003\u6211\u603b\u7ed3\u7684\u53e6\u4e00\u4e2a\u6587\u6863 https:\/\/github.com\/yuhuiaws\/ML-study\/blob\/main\/ML%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E5%85%A5%E9%97%A8\/%E5%BC%82%E5%B8%B8%E6%A3%80%E6%B5%8B%E4%BB%8B%E7%BB%8D.pptx\uff09\uff1b\u7136\u540e\u5982\u679c\u5f02\u5e38\u70b9\u7684\u539f\u56e0\u80fd\u5f52\u7ed3\u5230\u7279\u5f81\u7c92\u5ea6\uff1a\u5982\u679c\u4e4b\u540e\u8981\u9009\u62e9\u5bf9\u5f02\u5e38\u503c\u76f8\u5bf9\u4e0d\u654f\u611f\u7684\u6a21\u578b\u6bd4\u5982\u968f\u673a\u68ee\u6797\u6216XGBoost\uff0c\u8fd9\u91cc\u53ef\u4ee5\u5148\u8df3\u8fc7\u5f02\u5e38\u503c\u5904\u7406\u6b65\u9aa4\u3002\u7136\u540e\u628a\u5f02\u5e38\u503c\u5f53\u6210\u7f3a\u5931\u503c\u6765\u5904\u7406\uff1b\u6216\u8005\u7528\u53d8\u6362\u65b9\u6cd5\uff0c\u6bd4\u5982\u5bf9\u5f02\u5e38\u7684\u8fde\u7eed\u6027\u7279\u5f81\u53d6\u81ea\u7136\u5bf9\u6570\uff0c\u5bf9\u5f02\u5e38\u7684\u8fde\u7eed\u6027\u7279\u5f81\u5206\u7bb1\u5904\u7406\uff0c\u5bf9\u5f02\u5e38\u7684\u8fde\u7eed\u7279\u5f81\u505a3-sigma\u622a\u65ad\u5904\u7406\uff0c\u5206\u4f4d\u6570\u4e0a\u4e0b\u9650\u622a\u65ad\u5904\u7406\u6216\u8005\u6700\u5927\u6700\u5c0f\u503c\u622a\u65ad\u5904\u7406\uff08\u53c2\u8003Amazon SageMaker Data Wrangler\u4e2d\u5173\u4e8e\u5f02\u5e38\u503c\u7684\u5904\u7406https:\/\/docs.aws.amazon.com\/sagemaker\/latest\/dg\/data-wrangler-transform.html\uff1a\uff09\uff1b\u5426\u5219\u4ece\u6570\u636e\u96c6\u4e2d\u5254\u9664\u5f02\u5e38\u6837\u672c\uff0c\u628a\u5f02\u5e38\u6837\u672c\u5355\u72ec\u5b58\u50a8\u4ee5\u4fbf\u4e4b\u540e\u505a\u5f02\u5e38\u68c0\u6d4b\u3002<\/p>\n 3<\/strong><\/p>\n \u6b65\u9aa4\u4e09: \u7279\u5f81\u9884\u5904\u7406\u4e4b\u7f3a\u5931\u503c\u5904\u7406<\/strong><\/p>\n \u6211\u4eec\u62ff\u5230\u7684\u539f\u59cb\u6570\u636e\u4e2d\u4e00\u822c\u90fd\u4f1a\u6709\u4e00\u4e9b\u7279\u5f81\u6709\u7f3a\u5931\u7684\u73b0\u8c61\uff08\u89c1\u4e0b\u56fe\uff09\u3002\u7f3a\u5931\u503c\u7684\u539f\u56e0\u6709\u51e0\u4e0b\u51e0\u79cd\uff1a\u6682\u65f6\u65e0\u6cd5\u83b7\u53d6\u67d0\u4e9b\u7279\u5f81\u503c\uff1b\u51fa\u73b0\u4eba\u4e3a\u9519\u8bef\uff1b\u6570\u636e\u91c7\u96c6\u9636\u6bb5\u51fa\u73b0\u7684\u7a0b\u5e8f\/\u8bbe\u5907\u6545\u969c\u5bfc\u81f4\uff1b\u5386\u53f2\u5c40\u9650\u6027\uff1b\u6709\u610f\u9690\u7792\u7684\u7279\u5f81\u503c\uff1b\u6709\u4e9b\u6837\u672c\u7684\u7279\u5f81\u503c\u6839\u672c\u5c31\u662f\u4e0d\u5b58\u5728\u7684\uff1b\u4ece\u5f02\u5e38\u503c\u8f6c\u6362\u800c\u6765\uff1b\u7ebf\u4e0a\u9884\u6d4b\u65f6\u4ecefeature store\u83b7\u53d6\u7279\u5f81\u8d85\u65f6\u3002<\/p>\n <\/p>\n \u901a\u5e38\u7528\u6765\u5904\u7406\u7f3a\u5931\u503c\u7684\u65b9\u5f0f\uff1a\u5728\u9884\u6d4b\u65f6\u5904\u7406\u7f3a\u5931\u503c\u5e76\u5199\u5b8c\u6574\u7684\u7279\u5f81\u5411\u91cf\u5230\u65e5\u5fd7\uff0c\u8bad\u7ec3\u65f6\u76f4\u63a5\u7528\u8be5\u65e5\u5fd7\uff08\u6700\u5efa\u8bae\u7684\u65b9\u5f0f\uff0c\u80fd\u6700\u5927\u7a0b\u5ea6\u7684\u4fdd\u8bc1\u7279\u5f81\u7684\u7ebf\u4e0a\u7ebf\u4e0b\u4e00\u81f4\u6027\uff09\uff1b\u6216\u8005\u8ba9\u8bad\u7ec3\u7b97\u6cd5\u5728\u8bad\u7ec3\u9636\u6bb5\u81ea\u884c\u5904\u7406\u7f3a\u5931\u503c\uff08XGBoost\u548cLightGBM\u90fd\u652f\u6301\u81ea\u884c\u5904\u7406\u7f3a\u5931\u503c\uff09\uff1b\u6216\u8005\u5728\u8bad\u7ec3\u5f00\u59cb\u524d\u5904\u7406\u6389\u7f3a\u5931\u503c\u3002\uff08\u5efa\u8bae\uff1a\u5728\u505a\u7f3a\u5931\u503c\u5904\u7406\u7684\u65f6\u5019\uff0c\u5c1d\u8bd5\u65b0\u589e\u4e00\u4e2a\u7279\u5f81\u6765\u6807\u8bc6\u6709\u7f3a\u5931\u503c\u7684\u7279\u5f81\u5bf9\u5e94\u7684\u8be5\u6761\u6837\u672c\u8fd9\u4e2a\u7279\u5f81\u662f\u5426\u505a\u4e86\u7f3a\u5931\u503c\u5904\u7406\uff0c\u53ef\u80fd\u6700\u540e\u7684\u6a21\u578b\u6548\u679c\u66f4\u597d\uff08\u76f8\u5f53\u4e8e\u7ed9\u6a21\u578b\u66f4\u591a\u7684\u7684\u4fe1\u53f7\uff09\u3002\u5177\u4f53\u53ef\u4ee5\u53c2\u8003Amazon SageMaker Data Wrangler\u7f3a\u5931\u503c\u5904\u7406\u4e2d\u63d0\u5230\u7684\u8fd9\u4e2aindicator https:\/\/docs.aws.amazon.com\/sagemaker\/latest\/dg\/data-wrangler-transform.html:\uff09<\/p>\n \u5bf9\u8bad\u7ec3\u96c6\u5904\u7406\u7f3a\u5931\u503c\uff08\u6709\u53ef\u80fd\u591a\u79cd\u7279\u5f81\u90fd\u6709\u7f3a\u5931\u503c\uff09\uff0c\u6d41\u7a0b\u5982\u4e0b\uff08\u5904\u7406\u8fc7\u7a0b\u4e2d\u4e3a\u4e86\u4e0d\u53d1\u751f\u4fe1\u606f\u6cc4\u9732\uff0c\u4e0d\u8981\u53bb\u770b\u9a8c\u8bc1\u96c6\u548c\u6d4b\u8bd5\u96c6\u4e2d\u7684\u8be5\u7279\u5f81\u7684\u4efb\u4f55\u53d6\u503c\uff09\uff1a<\/p>\n \u9996\u5148\u9010\u4e2a\u68c0\u67e5\u6bcf\u4e2a\u7279\u5f81\u7684\u7f3a\u5931\u60c5\u51b5\u3002\u5bf9\u4e8e\u67d0\u4e2a\u7279\u5f81\uff0c\u5982\u679c\u6709\u592a\u591a\u7684\u8bad\u7ec3\u6837\u672c\u6709\u7f3a\u5931\u503c\u6bd4\u5982\u8d85\u8fc71\u534a\u7684\u6837\u672c\u91cf\uff0c\u8003\u8651\u662f\u5426\u53ef\u4ee5\u4ece\u6570\u636e\u96c6\u4e2d\u53bb\u9664\u8fd9\u4e2a\u7279\u5f81\u3002\u5982\u679c\u9700\u8981\u53bb\u9664\u7684\u7279\u5f81\u6709\u5f88\u591a\u4e2a\uff0c\u8bf4\u660e\u6570\u636e\u96c6\u8d28\u91cf\u592a\u5dee\uff0c\u6700\u597d\u8003\u8651\u91cd\u65b0\u6536\u96c6\u6570\u636e\uff1b<\/p>\n<\/li>\n \u7136\u540e\u9010\u4e2a\u68c0\u67e5\u6bcf\u4e2a\u6837\u672c\u7684\u7f3a\u5931\u60c5\u51b5\u3002\u5bf9\u4e8e\u67d0\u4e2a\u6837\u672c\uff0c\u5982\u679c\u7f3a\u5931\u7684\u7279\u5f81\u6570\u91cf\u592a\u591a\u6bd4\u5982\u8d85\u8fc71\u534a\u6570\u91cf\u7684\u7279\u5f81\u90fd\u7f3a\u5931\uff0c\u90a3\u4e48\u53ef\u4ee5\u8003\u8651\u662f\u5426\u53ef\u4ee5\u628a\u8fd9\u4e2a\u6837\u672c\u7ed9\u53bb\u9664\uff1b<\/p>\n<\/li>\n \u63a5\u7740\u8003\u8651\u5982\u679c\u4e4b\u540e\u7684\u6a21\u578b\u9009\u62e9\u662f\u80fd\u76f4\u63a5\u5904\u7406\u542b\u7f3a\u5931\u503c\u6837\u672c\u7684\u6a21\u578b\u6bd4\u5982XGBoost\uff0c\u90a3\u4e48\u5148\u8df3\u8fc7\u7f3a\u5931\u503c\u5904\u7406\u6b65\u9aa4\uff1b\u5426\u5219\u8fdb\u884c\u7f3a\u5931\u503c\u8865\u5168\u3002<\/p>\n<\/li>\n<\/ul>\n \u5728\u8bad\u7ec3\u96c6\u505a\u5b8c\u7f3a\u5931\u503c\u5904\u7406\u540e\uff0c\u5c31\u53ef\u4ee5\u5904\u7406\u6d4b\u8bd5\u96c6\u548c\u9a8c\u8bc1\u96c6\u7684\u7f3a\u5931\u503c\u4e86\uff1a\u5982\u679c\u662f\u8fde\u7eed\u6027\u7279\u5f81\uff0c\u7528\u8bad\u7ec3\u96c6\u8be5\u7279\u5f81\u7684\u5e73\u5747\u503c\u6216\u8005\u4e2d\u4f4d\u6570\u8865\u5168\uff1b\u5982\u679c\u662fcategory\u7279\u5f81\uff0c\u7528\u8bad\u7ec3\u96c6\u8be5\u7279\u5f81\u7684\u4f17\u6570\uff08\u51fa\u73b0\u9891\u7387\u6700\u9ad8\u7684\u90a3\u4e2a\u503c\uff09\u6765\u8865\u5168\u3002\u5728\u9884\u6d4b\u65f6\u5904\u7406\u7f3a\u5931\u503c\u5e38\u89c1\u7684\u65b9\u6cd5\u5c31\u662f\u4f7f\u7528\u9884\u5148\u8ba1\u7b97\u597d\u7684\u8fde\u7eed\u7279\u5f81\u7684\u5e73\u5747\u503c\/\u4e2d\u4f4d\u6570\uff0c\u6216\u8005category\u7279\u5f81\u7684\u4f17\u6570\u6765\u8865\u5168\u3002<\/p>\n \u9488\u5bf9\u8bad\u7ec3\u96c6\uff0c\u5e38\u89c1\u7684\u7f3a\u5931\u503c\u8865\u5168\u7684\u65b9\u6cd5\uff08\u5747\u503c\u8865\u5168\u548c\u540c\u7c7b\u5747\u503c\u8865\u5168\u53ef\u80fd\u662f\u7528\u7684\u6700\u591a\u7684\u65b9\u6cd5\uff09\uff1a<\/p>\n \u65b9\u6cd5<\/strong><\/p>\n<\/td>\n \u539f\u7406<\/strong><\/p>\n<\/td>\n \u5907\u6ce8<\/strong><\/p>\n<\/td>\n<\/tr>\n \u5747\u503c\u8865\u5168<\/p>\n<\/td>\n \u5982\u679c\u7279\u5f81\u662f\u8fde\u7eed\u578b\uff0c \u7528\u5e73\u5747\u503c\u8865\u5168\uff1b\u5982\u679c\u7279\u5f81\u662f\u79bb\u6563\u578b\uff0c\u7528\u4f17\u6570\u8865\u5168<\/p>\n<\/td>\n \u5bf9\u4e8e\u8fde\u7eed\u7279\u5f81\uff0c\u9664\u4e86\u5747\u503c\u8865\u5168\uff0c\u4e2d\u4f4d\u6570\u8865\u5168\u4e5f\u6bd4\u8f83\u5e38\u7528\u3002<\/p>\n<\/td>\n<\/tr>\n \u540c\u7c7b\u5747\u503c\u8865\u5168<\/p>\n<\/td>\n \u9996\u5148\u6309\u7167\u76ee\u6807\u53d8\u91cf\u7684 label\u5bf9\u6837\u672c\u8fdb\u884c\u5206\u7c7b\uff0c\u7136\u540e\u4ee5\u8be5\u7c7b\u4e2d\u7684\u6837\u672c\u7684\u5747\u503c\u6216\u8005\u4f17\u6570\u6765\u8865\u5168\u7f3a\u5931\u503c<\/p>\n<\/td>\n \u9002\u5408\u5206\u7c7b\u95ee\u9898<\/p>\n<\/td>\n<\/tr>\n \u57fa\u4e8e\u805a\u7c7b\u7684\u8865\u5168<\/p>\n<\/td>\n \u628a\u6ca1\u6709\u7f3a\u5931\u503c\u7684\u7279\u5f81\u90fd\u62bd\u53d6\u51fa\u6765\u7ec4\u6210\u65b0\u7684\u6570\u636e\u96c6\u8fdb\u884c\u805a\u7c7b\uff0c\u7136\u540e\u628a\u6bcf\u4e2a\u6837\u672c\u7684\u7f3a\u5931\u503c\u8fdb\u884c\u540c\u7c07\u5747\u503c\u6216\u8005\u4f17\u6570\u8865\u5168<\/p>\n<\/td>\n \u8fd9\u79cd\u65b9\u6cd5\u7c7b\u4f3c\u540c\u7c7b\u5747\u503c\u8865\u5168\uff0c\u4e0d\u8fc7\u5b83\u5bf9\u5206\u7c7b\u548c\u56de\u5f52\u95ee\u9898\u90fd\u9002\u7528\uff1b\u5b83\u8fd8\u53ef\u4ee5\u505a\u7c07\u95f4\u52a0\u6743\u5747\u503c\u8865\u5168\uff0c\u52a0\u6743\u65b9\u6cd5\u7528\u6837\u672c\u5bf9\u6bcf\u4e2a\u7c07\u7684\u5f52\u5c5e\u5ea6\u3002<\/p>\n<\/td>\n<\/tr>\n \u77e9\u9635\u8865\u5168<\/p>\n<\/td>\n \u4ece\u6700\u5c0f\u5316\u77e9\u9635\u7684\u79e9\u51fa\u53d1\uff0c\u901a\u8fc7\u5229\u7528\u6570\u503c\u4f18\u5316\u6765\u89e3\u51fa\u77e9\u9635\u4e2d\u7684\u7f3a\u5931\u503c\u3002<\/p>\n<\/td>\n -<\/p>\n<\/td>\n<\/tr>\n \u9ad8\u7ef4\u6620\u5c04<\/p>\n<\/td>\n \u628a\u5355\u4e2a\u79bb\u6563category\u53d8\u91cf\u53d8\u4e3aone-hot\u5411\u91cf\u3002one-hot\u5411\u91cf\u957f\u5ea6\u4e3acategory\u679a\u4e3e\u503c\u6570\u91cf+1,\u7f3a\u5931\u503c\u5c31\u8868\u793a\u4e3a\u6700\u540e\u7684other\u7279\u5f81<\/p>\n<\/td>\n \u5bf9\u4e8e\u8fde\u7eed\u7279\u5f81\uff0c\u9ad8\u7ef4\u6620\u5c04\u65e0\u6cd5\u76f4\u63a5\u5904\u7406\u3002\u53ef\u4ee5\u5bf9\u8fde\u7eed\u7279\u5f81\u79bb\u6563\u5316\u4e4b\u540e\uff0c\u518d\u8fdb\u884c\u9ad8\u7ef4\u6620\u5c04\u3002<\/p>\n<\/td>\n<\/tr>\n \u57fa\u4e8e\u9884\u6d4b\u7684\u8865\u5168<\/p>\n<\/td>\n \u5c06\u542b\u7f3a\u5931\u503c\u7684\u7279\u5f81\u4f5c\u4e3a\u9884\u6d4b\u76ee\u6807\uff0c\u5e76\u6392\u9664\u6389\u539f\u6765\u7684\u76ee\u6807\u53d8\u91cf\uff0c\u5efa\u7acb\u6a21\u578b\u6765\u9884\u6d4b\u3002<\/p>\n<\/td>\n \u8fd9\u79cd\u65b9\u6cd5\u7684\u6548\u679c\u76f8\u5bf9\u8f83\u597d\u3002<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n \u5bf9\u4e8e\u6709\u7f3a\u5931\u503c\u7684\u6570\u636e\u96c6\u5728\u7ecf\u8fc7\u7f3a\u5931\u503c\u5904\u7406\u540e\u5982\u4f55\u9009\u578b\uff0c\u53ef\u4ee5\u53c2\u8003\u5982\u4e0b\uff1a\u6570\u636e\u91cf\u5f88\u5c0f\uff0c\u7528\u8d1d\u53f6\u65af\u6a21\u578b\uff1b\u6570\u636e\u91cf\u9002\u4e2d\u6216\u8005\u8f83\u5927\uff0c\u7528XGBoost\u6216\u8005LightGBM\u8fd9\u6837\u7684\u6a21\u578b\uff1b\u6570\u636e\u91cf\u8f83\u5927\uff0c\u4e5f\u53ef\u4ee5\u7528\u795e\u7ecf\u7f51\u7edc\uff1b\u907f\u514d\u4f7f\u7528\u8ddd\u79bb\u5ea6\u91cf\u76f8\u5173\u7684\u6a21\u578b\uff0c\u6bd4\u5982SVM\uff0c\u7ebf\u6027\u56de\u5f52\uff0c\u903b\u8f91\u56de\u5f52\uff08\u53ef\u80fd\u7684\u539f\u56e0\u662f\uff0c\u7f3a\u5931\u503c\u5904\u7406\u4f1a\u5e26\u5165\u4e00\u5b9a\u7684\u566a\u97f3\uff0c\u800c\u57fa\u4e8e\u8ddd\u79bb\u7684\u6a21\u578b\u672c\u8eab\u5bf9\u566a\u97f3\u5c31\u6bd4\u8f83\u654f\u611f\uff09<\/p>\n \u603b\u7ed3<\/strong><\/p>\n Data-centric AI\u4e4b\u7279\u5f81\u5de5\u7a0b\u7b2c\u4e00\u8bb2\u5230\u6b64\u7ed3\u675f\u3002\u672c\u6587\u4ecb\u7ecd\u4e86\u7279\u5f81\u76f8\u5173\u7684\u6982\u5ff5\u4ee5\u53ca\u8fde\u7eed\u7279\u5f81\u548ccategory\u7279\u5f81\u7684\u7279\u70b9\uff0c\u5f52\u7eb3\u4e86\u201c\u597d\u7684\u7279\u5f81\u201d\u5e94\u6709\u7684\u7279\u8d28\u3002\u5728\u6b64\u57fa\u7840\u4e0a\uff0c\u4ecb\u7ecd\u4e86\u7279\u5f81\u9884\u5904\u7406\u7684\u4e09\u4e2a\u5b50\u6b65\u9aa4\u5373\u5b57\u7b26\u4e32\u7279\u5f81\u8f6c\u6362\u4e3a\u6570\u503c\u7279\u5f81\uff0c\u5f02\u5e38\u503c\u5904\u7406\u548c\u7f3a\u5931\u503c\u5904\u7406\u3002\u76f8\u4fe1\u5927\u5bb6\u73b0\u5728\u5bf9\u7279\u5f81\u4ee5\u53ca\u7279\u5f81\u5de5\u7a0b\u6709\u4e86\u66f4\u6df1\u523b\u7684\u7406\u89e3\uff0c\u63a5\u4e0b\u6765\u7684\u7b2c\u4e8c\u8bb2\u6211\u4f1a\u7ee7\u7eed\u8bb2\u7279\u5f81\u9884\u5904\u7406\u3002\u518d\u6b21\u611f\u8c22\u5927\u5bb6\u7684\u9605\u8bfb\u3002<\/p>\n \u672c\u7bc7\u4f5c\u8005<\/strong><\/p>\n <\/p>\n \u6881\u5b87\u8f89<\/strong><\/p>\n \u4e9a\u9a6c\u900a\u4e91\u79d1\u6280\u673a\u5668\u5b66\u4e60\u4ea7\u54c1\u6280\u672f\u4e13\u5bb6<\/em><\/p>\n \u8d1f\u8d23\u57fa\u4e8e\u4e9a\u9a6c\u900a\u4e91\u79d1\u6280\u7684\u673a\u5668\u5b66\u4e60\u65b9\u6848\u7684\u54a8\u8be2\u4e0e\u8bbe\u8ba1\uff0c\u4e13\u6ce8\u4e8e\u673a\u5668\u5b66\u4e60\u7684\u63a8\u5e7f\u4e0e\u5e94\u7528\uff0c\u6df1\u5ea6\u53c2\u4e0e\u4e86\u5f88\u591a\u771f\u5b9e\u5ba2\u6237\u7684\u673a\u5668\u5b66\u4e60\u9879\u76ee\u7684\u6784\u5efa\u4ee5\u53ca\u4f18\u5316\u3002\u5bf9\u4e8e\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u5206\u5e03\u5f0f\u8bad\u7ec3\uff0c\u63a8\u8350\u7cfb\u7edf\u548c\u8ba1\u7b97\u5e7f\u544a\u7b49\u9886\u57df\u5177\u6709\u4e30\u5bcc\u7ecf\u9a8c\u3002<\/p>\n <\/p>\n <\/p>\n \u542c\u8bf4\uff0c\u70b9\u5b8c\u4e0b\u97624\u4e2a\u6309\u94ae<\/strong><\/p>\n \u5c31\u4e0d\u4f1a\u78b0\u5230bug\u4e86\uff01<\/strong><\/p>\n <\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"\u7279\u5f81\u5de5\u7a0b\u4e0e\u6570\u636e\u9884\u5904\u7406_Data Structure Visualization\u524d\u8a00\u8fd9\u4e24\u5e74\u6211\u4eec\u89c2\u5bdf\u5230\u8d8a\u6765\u8d8a\u591a\u7684\u7b97\u6cd5\u5de5\u7a0b\u5e08\u91cd\u89c6\u6570\u636e\u7684\u7279\u5f81\u5de5\u7a0b\uff0cAI\u4e1a\u754c\u5927...","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[],"tags":[],"_links":{"self":[{"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/posts\/8639"}],"collection":[{"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/comments?post=8639"}],"version-history":[{"count":0,"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/posts\/8639\/revisions"}],"wp:attachment":[{"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/media?parent=8639"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/categories?post=8639"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mushiming.com\/wp-json\/wp\/v2\/tags?post=8639"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\n
\n\n
\n \n \n \n \n \n \n \n \n \n \n \n \n\n
\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
\n\n
\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n