[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-78934":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":8,"htmlUrl":8,"language":9,"languages":8,"totalLinesOfCode":8,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":14,"forks30d":14,"starsTrendScore":18,"compositeScore":19,"rankGlobal":8,"rankLanguage":8,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":8,"pushedAt":8,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":14,"starSnapshotCount":14,"syncStatus":15,"lastSyncTime":28,"discoverSource":29},78934,"ml-dl-formula-cheatsheet","Jerry-0821\u002Fml-dl-formula-cheatsheet","Jerry-0821",null,"TeX",337,20,4,1,0,2,8,223,6,3.97,"MIT License",false,"main",true,[],"2026-06-12 02:03:49","# ML & Deep Learning Formula Cheat Sheets\n\nCompact LaTeX formula references for theory-focused machine learning and deep\nlearning review.\n\nThis repository contains two standalone PDF review sheets designed for quick\nmathematical lookup. They emphasize formulas, notation, objectives, update\nrules, evaluation quantities, model decisions, tensor shapes where relevant,\nand compact derivations without requiring readers to search across full lecture\nnotes or textbooks.\n\nThe content follows the study-material scope covered by each sheet. It is not\nintended to be an exhaustive textbook-level reference or a universal catalog of\nevery machine learning and deep learning topic. Instead, it is a focused\nmathematical review resource for beginners and learners who want to strengthen\ntheir theoretical foundation.\n\nThese sheets are not tutorials and do not replace complete courses or\ntextbooks. They are compact references for revision and formula checking.\n\n## Download\n\n| Sheet | Focus | PDF |\n| --- | --- | --- |\n| Deep Learning Formula Cheat Sheet | Neural networks, optimization, CNNs, sequence models, attention, Transformers, and tensor shapes | [Download PDF](.\u002Fmain.pdf) |\n| Machine Learning Formula & Decision Sheet | Regression, classification, evaluation, diagnosis, tree ensembles, clustering, recommenders, and reinforcement learning | [Download PDF](.\u002Fmachine-learning-formula-decision-sheet.pdf) |\n\n## Preview\n\n### Machine Learning Formula & Decision Sheet\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fmachine-learning-contents.png\" width=\"47%\" alt=\"Machine learning preview: contents\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fmachine-learning-lasso.png\" width=\"47%\" alt=\"Machine learning preview: Lasso and Elastic Net formulas\">\n\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fmachine-learning-roc-auc.png\" width=\"47%\" alt=\"Machine learning preview: PR-AUC and model diagnosis formulas\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fmachine-learning-deep-q.png\" width=\"47%\" alt=\"Machine learning preview: Q-learning and Deep Q Networks\">\n\u003C\u002Fp>\n\n### Deep Learning Formula Cheat Sheet\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fdeep-learning-contents.png\" width=\"47%\" alt=\"Deep learning preview: contents\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fdeep-learning-optimization.png\" width=\"47%\" alt=\"Deep learning preview: optimization formulas\">\n\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fdeep-learning-attention.png\" width=\"47%\" alt=\"Deep learning preview: attention decoder formulas\">\n  \u003Cimg src=\"assets\u002Fpreview\u002Fdeep-learning-transformer.png\" width=\"47%\" alt=\"Deep learning preview: Transformer formulas\">\n\u003C\u002Fp>\n\n## About the Sheets\n\n### Machine Learning Formula & Decision Sheet\n\nA mathematical reference for machine learning theory and model-based\ndecisions. It combines objectives, update rules, compact derivations, metric\ninterpretation, and judgment-style checkpoints for questions where selecting\nthe correct method matters as much as recalling the formula.\n\n### Deep Learning Formula Cheat Sheet\n\nA compact formula and tensor-shape reference for deep learning. It focuses on\nforward computations, losses, gradient flows, optimizer updates,\narchitecture-specific objectives, and shape rules.\n\n## Coverage\n\n| Machine Learning Formula & Decision Sheet | Deep Learning Formula Cheat Sheet |\n| --- | --- |\n| Linear, polynomial, and logistic regression | Neural-network notation and forward propagation |\n| Cost functions, gradient descent, regularization, Lasso, and Elastic Net | Loss functions, backpropagation, initialization, and optimization |\n| Evaluation, ROC-AUC, PR-AUC, bias\u002Fvariance, and error analysis | Regularization and batch normalization |\n| Neural-network foundations, decision trees, bagging, and boosting | CNNs, classic architectures, object detection, and YOLO |\n| K-means, anomaly detection, and recommender systems | Face recognition and neural style transfer |\n| Reinforcement learning, Bellman equations, Q-learning, and Deep Q Networks | RNN\u002FGRU\u002FLSTM, embeddings, Seq2Seq, attention, and Transformers |\n\n## Design Principles\n\n- Keep entries compact and formula-focused.\n- Prefer display mathematics for central objectives and updates.\n- Define notation close to the formulas that use it.\n- Include shapes where dimensions clarify the computation.\n- Include short derivations where they explain an update, metric, or decision rule.\n- Prefer concise tables and notes over textbook-length prose.\n\n## Artifacts\n\n| Artifact | Status |\n| --- | --- |\n| `machine-learning-formula-decision-sheet.pdf` | Available as the current Machine Learning review sheet. |\n| `main.pdf` | Available as the current Deep Learning formula sheet. |\n| Deep Learning LaTeX source in `main.tex` and `sections\u002F` | Included in this repository and built by GitHub Actions. |\n\nThe current tagged Deep Learning draft is\n[v0.2.0 - Formula Hierarchy and Core Extensions](https:\u002F\u002Fgithub.com\u002FJerry-0821\u002Fml-dl-formula-cheatsheet\u002Freleases\u002Ftag\u002Fv0.2.0).\n\n## Building From Source\n\nThe currently included LaTeX source builds the Deep Learning sheet:\n\n```bash\nmake pdf\n```\n\nManual fallback:\n\n```bash\nlatexmk -pdf main.tex\n```\n\nGitHub Actions builds `main.pdf` and uploads it as the\n`deep-learning-formula-cheatsheet-pdf` artifact.\n\n## Source and Scope Policy\n\nEach sheet follows the scope of the study materials used to prepare it.\nExternal references may be used to verify standard formulas, notation, shapes,\nor mathematical correctness, but the sheets are not intended to silently\nexpand into complete textbooks. The goal is a reliable, compact mathematical\nreview resource within the covered topic range.\n","该项目提供机器学习和深度学习的公式速查表，旨在帮助用户快速回顾相关理论知识。核心功能包括两份独立的PDF文档，一份专注于深度学习领域如神经网络、优化算法、卷积神经网络等，另一份则涵盖了机器学习中的回归、分类、聚类等基本概念与技术。这些文档以LaTeX编写，确保了数学公式的清晰展示，并且特别强调了符号、目标函数、更新规则等内容，方便读者理解和记忆。适合于正在学习或复习机器学习及深度学习基础知识的学生、研究人员以及从业人员使用，在准备考试、论文写作或是日常工作中需要快速查找相关公式时尤为有用。","2026-06-11 03:57:20","CREATED_QUERY"]