[{"data":1,"prerenderedAt":263},["ShallowReactive",2],{"repo-dmlc-xgboost-zh-CN":3,"i18n-dmlc-xgboost-zh-CN":260},{"mirrorPool":4,"detail":13,"relatedProjects":138},[5,9],{"key":6,"label":7,"base":8,"type":6},"github","mirror.github","https:\u002F\u002Fgithub.com\u002F",{"key":10,"label":11,"base":12,"type":10},"auto-proxy","mirror.autoProxy","https:\u002F\u002Fgithub.xxlab.tech\u002F",{"owner":14,"repo":15,"repoId":16,"repoUrl":17,"starCount":18,"language":19,"description":20,"languages":21,"screenshots":32,"homepage":33,"pushedAt":34,"ownerAvatar":35,"ownerName":36,"ownerBio":37,"lastSyncTime":38,"lastReleaseTime":39,"tags":40,"license":46,"category":47,"subgroup":48,"releases":49},"dmlc","xgboost",16587283,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost",28892,"C++","Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library,  for Python, R, Java, Scala, C and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow",{"C++":22,"Python":23,"Cuda":24,"R":25,"Scala":26,"Java":27,"CMake":28,"Shell":29,"C":30,"PowerShell":31},45.8,20.4,16.7,7.2,4.8,3.3,0.8,0.6,0.4,0.1,[],"https:\u002F\u002Fxgboost.readthedocs.io\u002F","2026-10-19","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F11508361?v=4","Distributed (Deep) Machine Learning Community","A Community of Awesome Machine Learning Projects","2026-10-20T10:23:46.638009+00:00","2026-09-15T08:48:08Z",[41,42,43,44,45,15],"distributed-systems","gbdt","gbm","gbrt","machine-learning","Apache-2.0","development","机器学习框架",[50,63,76,87,98,108,118,128],{"tag":51,"publishAt":39,"semVer":52,"prerelease":55,"assets":56},"v3.3.1",[53,53,54],3,1,false,[57],{"fileName":58,"sizeMb":59,"sha256":60,"downloadCount":61,"originUrl":62},"xgboost-src-3.3.1.tar.gz",2.61,"871d1a05779bc99bcb7f5fe1ff826e7d2e6a8a9c30d65a1d2fa0d47da32df07a",113,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.3.1\u002Fxgboost-src-3.3.1.tar.gz",{"tag":64,"publishAt":65,"semVer":66,"prerelease":55,"assets":69},"v3.4.2","2026-09-15T08:31:52Z",[53,67,68],4,2,[70],{"fileName":71,"sizeMb":72,"sha256":73,"downloadCount":74,"originUrl":75},"xgboost-src-3.4.2.tar.gz",2.62,"d7de76bdaf48e9e0bc84b27c0c42f15f3d831e624d102d6c019bab7027b3cabd",1364,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.4.2\u002Fxgboost-src-3.4.2.tar.gz",{"tag":77,"publishAt":78,"semVer":79,"prerelease":55,"assets":80},"v3.4.1","2026-08-15T08:30:07Z",[53,67,54],[81],{"fileName":82,"sizeMb":83,"sha256":84,"downloadCount":85,"originUrl":86},"xgboost-src-3.4.1.tar.gz",2.63,"34a5cb99a67bb98b44f204767eeeae642b65a86b2ecfca60082e4d74fd4d169a",658,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.4.1\u002Fxgboost-src-3.4.1.tar.gz",{"tag":88,"publishAt":89,"semVer":90,"prerelease":55,"assets":92},"v3.4.0","2026-08-04T10:46:20Z",[53,67,91],0,[93],{"fileName":94,"sizeMb":72,"sha256":95,"downloadCount":96,"originUrl":97},"xgboost-src-3.4.0.tar.gz","af4588b34c7fa1bfde258006beebbe181454f1fb74266f81883b23059af3b9fb",282,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.4.0\u002Fxgboost-src-3.4.0.tar.gz",{"tag":99,"publishAt":100,"semVer":101,"prerelease":55,"assets":102},"v3.3.0","2026-06-17T20:56:43Z",[53,53,91],[103],{"fileName":104,"sizeMb":59,"sha256":105,"downloadCount":106,"originUrl":107},"xgboost-src-3.3.0.tar.gz","22d4fba822fba5cd02299bf0c63ec68ff72606bc1b1bd910423d4b83c2f108ff",1574,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.3.0\u002Fxgboost-src-3.3.0.tar.gz",{"tag":109,"publishAt":110,"semVer":111,"prerelease":55,"assets":112},"v3.2.0","2026-02-10T08:14:33Z",[53,68,91],[113],{"fileName":114,"sizeMb":83,"sha256":115,"downloadCount":116,"originUrl":117},"xgboost-src-3.2.0.tar.gz","16a31dfbc0c54544c9c36ab5f696fa7b646c125f161c52c814d757a58241a404",14585,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.2.0\u002Fxgboost-src-3.2.0.tar.gz",{"tag":119,"publishAt":120,"semVer":121,"prerelease":55,"assets":122},"v3.1.3","2026-01-09T09:57:40Z",[53,54,53],[123],{"fileName":124,"sizeMb":83,"sha256":125,"downloadCount":126,"originUrl":127},"xgboost-src-3.1.3.tar.gz","67800a7c1c0455c22c9be73dbf3c39bfd9ac9627b2cb617eb2795fd675a9d49e",492,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.1.3\u002Fxgboost-src-3.1.3.tar.gz",{"tag":129,"publishAt":130,"semVer":131,"prerelease":55,"assets":132},"v3.1.2","2025-11-20T13:33:54Z",[53,54,68],[133],{"fileName":134,"sizeMb":83,"sha256":135,"downloadCount":136,"originUrl":137},"xgboost-src-3.1.2.tar.gz","12f2d6f735fa71e007c40171fd926c12306276dd299dc48f6c923e4f3891c33e",693,"https:\u002F\u002Fgithub.com\u002Fdmlc\u002Fxgboost\u002Freleases\u002Fdownload\u002Fv3.1.2\u002Fxgboost-src-3.1.2.tar.gz",[139,159,174,195,215,229],{"owner":140,"repo":140,"repoId":141,"category":47,"subgroup":48,"language":142,"tags":143,"starCount":151,"ownerUid":152,"latestTag":153,"platforms":154,"lastReleaseTime":156,"downloadCount":157,"description":158},"pytorch",65600975,"Python",[144,145,146,45,147,148,149,150],"autograd","deep-learning","gpu","neural-network","numpy","python","tensor",104936,21003710,"v2.14.1",[155],"other","2026-09-30T19:14:48Z",2205,"PyTorch 是一个 Python 深度学习框架，提供带 GPU 加速的张量计算与基于动态图的自动求导，是科研与生产中训练神经网络的主流选择之一。",{"owner":160,"repo":161,"repoId":162,"category":47,"subgroup":48,"language":142,"tags":163,"starCount":167,"ownerUid":168,"latestTag":169,"platforms":170,"lastReleaseTime":171,"downloadCount":172,"description":173},"Lightning-AI","pytorch-lightning",178626720,[164,165,166,145,45,149,140],"ai","artificial-intelligence","data-science",31418,58386951,"2.6.6",[155],"2026-09-10T09:39:52Z",11025,"PyTorch Lightning 是一个深度学习训练框架（Python），把反向传播、混合精度、多 GPU 与分布式训练等工程代码自动化，让同一份模型代码从单卡扩展到上万卡而无需改动。",{"owner":175,"repo":176,"repoId":177,"category":164,"subgroup":48,"language":19,"tags":178,"starCount":185,"ownerUid":186,"latestTag":187,"platforms":188,"lastReleaseTime":192,"downloadCount":193,"description":194},"microsoft","onnxruntime",156939672,[179,145,180,45,181,182,140,183,184],"ai-framework","hardware-acceleration","neural-networks","onnx","scikit-learn","tensorflow",22180,6154722,"v1.31.1",[189,190,191],"windows","mac","linux","2026-10-26T19:29:51Z",563,"微软开源的跨平台机器学习推理和训练加速器，支持PyTorch、TensorFlow\u002FKeras等深度学习框架及scikit-learn、LightGBM等经典机器学习库。",{"owner":196,"repo":197,"repoId":198,"category":164,"subgroup":48,"language":19,"tags":199,"starCount":208,"ownerUid":209,"latestTag":210,"platforms":211,"lastReleaseTime":212,"downloadCount":213,"description":214},"lightgbm-org","LightGBM",64991887,[200,201,202,42,43,44,203,204,205,45,206,149,207],"data-mining","decision-trees","distributed","gradient-boosting","kaggle","lightgbm","parallel","r",18831,266222920,"v4.7.0",[189,190,191,155],"2026-07-18T20:20:12Z",760,"一款基于决策树算法的高效、分布式梯度提升（GBDT）框架，用于排序、分类等机器学习任务。",{"owner":216,"repo":217,"repoId":218,"category":164,"subgroup":48,"language":19,"tags":219,"starCount":222,"ownerUid":223,"latestTag":224,"platforms":225,"lastReleaseTime":226,"downloadCount":227,"description":228},"google-deepmind","mujoco",400501136,[217,220,221],"physics","robotics",15835,8596759,"3.17.0",[189,190,191],"2026-10-22T21:45:20Z",320,"一个由Google DeepMind维护的通用物理引擎，用于机器人、生物力学、图形和机器学习等领域的快速、精确仿真。",{"owner":230,"repo":231,"repoId":232,"category":47,"subgroup":48,"language":142,"tags":233,"starCount":253,"ownerUid":254,"latestTag":255,"platforms":256,"lastReleaseTime":257,"downloadCount":258,"description":259},"quantumlib","Cirq",114306758,[234,235,236,237,238,239,149,240,241,242,243,244,245,246,247,248,249,250,251,252],"algorithms","api","cirq","google","google-quantum","nisq","quantum","quantum-algorithms","quantum-circuit","quantum-circuit-simulator","quantum-computer-simulator","quantum-computing","quantum-development-kit","quantum-information","quantum-programming","quantum-programming-language","quantum-simulation","sdk","simulation",5086,31279789,"v1.7.0",[155],"2026-06-30T18:44:40Z",339,"Cirq 是 Google 开源的 Python 量子电路框架，用于创建、编辑与运行 NISQ 量子电路，可在内置模拟器或真实量子硬件上执行。",{"description":261,"readmeHtml":262},"XGBoost 是一个优化的分布式梯度提升库（C++ 核心，提供 Python、R、Java 等多语言绑定），以高效并行的树提升算法解决表格数据上的分类与回归问题，可运行到数十亿样本规模。","\u003Ch2 id=\"项目概述\">项目概述\u003C\u002Fh2>\u003Cp>XGBoost（eXtreme Gradient Boosting）是一个高效、灵活、可移植的分布式梯度提升库，实现 GBDT 等树提升算法，是表格数据竞赛与工业建模中使用最广泛的机器学习库之一。同一份代码可以跑在单机、Kubernetes、Hadoop、Spark、Dask 等分布式环境上，支持数十亿样本的训练。提供 Python、R、Java、Scala、C++ 等多语言接口。\u003C\u002Fp>\n\u003Cblockquote>\n\u003Cp>表格数据建模的常青树，单机到集群用同一套代码。\u003C\u002Fp>\n\u003C\u002Fblockquote>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>项目\u003C\u002Fth>\n\u003Cth>内容\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>平台\u003C\u002Ftd>\n\u003Ctd>Linux、macOS、Windows\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>语言\u003C\u002Ftd>\n\u003Ctd>C++ 核心，Python\u002FR\u002FJava\u002FScala\u002FC++ 绑定\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>许可\u003C\u002Ftd>\n\u003Ctd>Apache-2.0\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>运行环境\u003C\u002Ftd>\n\u003Ctd>单机、Hadoop、Spark、Dask、Kubernetes 等\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2 id=\"主要功能\">主要功能\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>梯度提升\u003C\u002Fstrong>：并行化树提升（GBDT\u002FGBM），分类、回归、排序一体覆盖\u003C\u002Fli>\n\u003Cli>\u003Cstrong>分布式训练\u003C\u002Fstrong>：同一份代码跑在 Spark、Dask、Hadoop 等集群环境\u003C\u002Fli>\n\u003Cli>\u003Cstrong>GPU 加速\u003C\u002Fstrong>：NVIDIA GPU 上训练与推理显著提速\u003C\u002Fli>\n\u003Cli>\u003Cstrong>稀疏感知\u003C\u002Fstrong>：自动处理缺失值与稀疏特征\u003C\u002Fli>\n\u003Cli>\u003Cstrong>正则化\u003C\u002Fstrong>：内置 L1\u002FL2 正则与剪枝，抑制过拟合\u003C\u002Fli>\n\u003Cli>\u003Cstrong>交叉验证与调参\u003C\u002Fstrong>：内置 CV 工具，与 Optuna 等调参库集成\u003C\u002Fli>\n\u003Cli>\u003Cstrong>多语言接口\u003C\u002Fstrong>：Python、R、JVM、C++ 等绑定共享同一核心\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"适用人群\">适用人群\u003C\u002Fh2>\u003Cul>\n\u003Cli>做表格数据建模的数据科学家\u003C\u002Fli>\n\u003Cli>参加机器学习竞赛的选手\u003C\u002Fli>\n\u003Cli>需要在大规模数据上训练树模型的工程师\u003C\u002Fli>\n\u003Cli>生产环境部署模型的团队\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"快速上手\">快速上手\u003C\u002Fh2>\u003Col>\n\u003Cli>安装 Python 包：\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cdiv class=\"code-block\">\u003Cpre>\u003Ccode class=\"hljs language-bash\">pip install xgboost\u003C\u002Fcode>\u003C\u002Fpre>\u003C\u002Fdiv>\u003Col start=\"2\">\n\u003Cli>用 scikit-learn 风格接口训练：\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cdiv class=\"code-block\">\u003Cpre>\u003Ccode class=\"hljs language-python\">from xgboost import XGBClassifier\n\nmodel = XGBClassifier(n_estimators=300, max_depth=6)\nmodel.fit(X_train, y_train)\u003C\u002Fcode>\u003C\u002Fpre>\u003C\u002Fdiv>\u003Col start=\"3\">\n\u003Cli>超大规模数据改用 Dask 或 Spark 接口，集群训练代码几乎不变\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Ch2 id=\"核心优势\">核心优势\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>精度高\u003C\u002Fstrong>：树提升在表格任务上长期是最强基线之一\u003C\u002Fli>\n\u003Cli>\u003Cstrong>扩展性好\u003C\u002Fstrong>：从笔记本到数千节点集群同一套 API\u003C\u002Fli>\n\u003Cli>\u003Cstrong>工程成熟\u003C\u002Fstrong>：十年打磨，文档、教程与生产案例齐全\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"系统要求\">系统要求\u003C\u002Fh2>\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>项目\u003C\u002Fth>\n\u003Cth>要求\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>操作系统\u003C\u002Ftd>\n\u003Ctd>Linux、macOS、Windows\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>硬件\u003C\u002Ftd>\n\u003Ctd>CPU 为主，GPU 可选加速\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>语言环境\u003C\u002Ftd>\n\u003Ctd>Python 3.x \u002F R \u002F JVM 等\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2 id=\"注意事项\">注意事项\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>数据类型\u003C\u002Fstrong>：输入需转为 DMatrix 或数值矩阵，类别特征需先编码\u003C\u002Fli>\n\u003Cli>\u003Cstrong>GPU 环境\u003C\u002Fstrong>：GPU 版本需要匹配的 CUDA 环境\u003C\u002Fli>\n\u003C\u002Ful>\n",1793197846616]