必要條件和事前準備
透過集合功能整理內容
你可以依據偏好儲存及分類內容。
機器學習密集課程是否適合您？
我對機器學習有一定程度的瞭解，但想進一步瞭解最新的相關資訊。
機器學習密集課程是重溫重點的絕佳選擇。依序完成所有單元，或只選取您感興趣的單元。
我有實際經驗，能將機器學習概念應用於資料處理和建構模型。
雖然「機器學習速成課程」可讓您複習機器學習的基本概念，但您也許想進一步瞭解一些進階機器學習課程，這些課程會介紹各種領域中解決機器學習問題的工具和技巧。
我想找教學課程，瞭解如何使用 Keras 等機器學習 API。
雖然「機器學習速成課程」包含多項程式設計練習，使用 numpy、pandas 和 Keras 等 ML 程式庫，但主要著重於教導 ML 概念，並未深入講解 ML API。如需其他 Keras 資源，請參閱
Keras 開發人員指南。
請先詳閱以下預習和必備條件部分，再開始修習機器學習衝刺課程，確保您已準備好完成所有單元。
預備作業
開始機器學習密集課程前，請先完成下列步驟：
-
如果您是機器學習的新手，請參閱「機器學習簡介」一文。本短程自修課程將介紹機器學習的基本概念。
-
如果您是 NumPy 新手，請完成 NumPy 超快速教學課程 Colab 練習，這會提供您這門課程所需的所有 NumPy 資訊。
-
如果您是 pandas 新手，請完成 pandas 超快速教學課程 Colab 練習，這會提供您這門課程所需的所有 pandas 資訊。
必要條件
機器學習密集課程不會假設您具備任何機器學習相關知識，也不會要求您具備這類知識。不過，為了瞭解課程內容和完成練習，建議學生具備下列先決條件：
以下各節會提供其他背景資料的連結，這些資料可能會對您有所助益。
代數學
線性代數
三角學
統計資料
微積分 (選修，適用於進階主題)
Python 程式設計
Python 教學課程將介紹下列 Python 基礎知識：
部分程式設計練習會使用下列較進階的 Python 概念：
Bash 終端機和 Cloud 控制台
如要在本機電腦或雲端控制台中執行程式設計練習，您必須熟悉如何使用指令列：
除非另有註明，否則本頁面中的內容是採用創用 CC 姓名標示 4.0 授權，程式碼範例則為阿帕契 2.0 授權。詳情請參閱《Google Developers 網站政策》。Java 是 Oracle 和/或其關聯企業的註冊商標。
上次更新時間：2025-01-28 (世界標準時間)。
[null,null,["上次更新時間：2025-01-28 (世界標準時間)。"],[[["\u003cp\u003eGoogle's Machine Learning Crash Course offers a flexible learning experience for users with varying levels of machine learning expertise, including beginners, those seeking a refresher, and experienced practitioners.\u003c/p\u003e\n"],["\u003cp\u003eThe course requires prework, such as familiarity with Python, NumPy, and pandas, and has prerequisites in algebra, linear algebra, statistics, and optionally, calculus, to fully grasp the concepts.\u003c/p\u003e\n"],["\u003cp\u003eWhile focusing on core ML concepts, the course incorporates practical programming exercises using libraries like NumPy, pandas, and Keras but doesn't delve deep into specific ML APIs.\u003c/p\u003e\n"],["\u003cp\u003eLearners are encouraged to complete the prework, including an introductory machine learning course and tutorials for NumPy and pandas, to ensure preparedness.\u003c/p\u003e\n"],["\u003cp\u003eThe course leverages the Colaboratory platform, offering browser-based programming exercises that require no setup and are best experienced on Chrome or Firefox desktops.\u003c/p\u003e\n"]]],[],null,["# Prerequisites and prework\n\n\u003cbr /\u003e\n\n### Is Machine Learning Crash Course right for you?\n\nI have little or no machine learning background. \nWe recommend going through all the material in order. \n[START LEARNING](/machine-learning/crash-course/linear-regression) \nI have some background in machine learning, but I'd like a more current and complete understanding. \nMachine Learning Crash Course will be a great refresher. Go through all the modules in order, or select only those modules that interest you. \n[START LEARNING](/machine-learning/crash-course/linear-regression) \nI have practical experience applying machine learning concepts to work with data and build models. \nWhile Machine Learning Crash Course may be useful to you as a refresher of fundamental machine learning concepts, you may also want to explore some of our advanced machine learning courses, which cover tools and techniques for solving machine learning problems in a variety of domains. \n[START LEARNING](/machine-learning/advanced-courses) \nI am looking for tutorials on how to use ML APIs like Keras. \nWhile Machine Learning Crash Course includes several programming exercises that use ML libraries such as numpy, pandas, and Keras, it is primarily focused on teaching ML concepts, and does not teach ML APIs in depth. For additional Keras resources, see the [Keras Developer guides](https://keras.io/guides/).\n\nPlease read through the following [Prework](#prework) and\n[Prerequisites](#prerequisites) sections before beginning Machine Learning\nCrash Course, to ensure you are prepared to complete all the modules.\n\nPrework\n-------\n\nBefore beginning Machine Learning Crash Course, do the following:\n\n1. If you're new to machine learning, take [Introduction to Machine\n Learning](/machine-learning/intro-to-ml). This short self-study course introduces fundamental machine learning concepts.\n2. If you are new to [NumPy](https://numpy.org), do the [NumPy\n Ultraquick Tutorial](https://colab.research.google.com/github/google/eng-edu/blob/main/ml/cc/exercises/numpy_ultraquick_tutorial.ipynb?utm_source=mlcc&utm_campaign=colab-external&utm_medium=referral&utm_content=mlcc-prework&hl=en) Colab exercise, which provides all the NumPy information you need for this course.\n3. If you are new to [pandas](https://pandas.pydata.org/), do the [pandas\n UltraQuick Tutorial](https://colab.research.google.com/github/google/eng-edu/blob/main/ml/cc/exercises/pandas_dataframe_ultraquick_tutorial.ipynb?utm_source=mlcc&utm_campaign=colab-external&utm_medium=referral&utm_content=mlcc-prework&hl=en) Colab exercise, which provides all the pandas information you need for this course.\n\nProgramming exercises run directly in your browser (no setup\nrequired!) using the [Colaboratory](https://colab.research.google.com)\nplatform. Colaboratory is supported on most major browsers, and is most\nthoroughly tested on desktop versions of Chrome and Firefox.\n\nPrerequisites\n-------------\n\nMachine Learning Crash Course does not presume or require any prior knowledge in\nmachine learning. However, to understand the concepts presented\nand complete the exercises, we recommend that students meet the\nfollowing prerequisites:\n\n- You must be comfortable with variables, linear equations,\n graphs of functions, histograms, and statistical means.\n\n- You should be a good programmer. Ideally, you should have some\n experience programming in [Python](https://www.python.org/) because\n the programming exercises are in Python. However, experienced\n programmers without Python experience can usually complete the programming\n exercises anyway.\n\nThe following sections provide links to additional background material\nthat is helpful.\n\n### Algebra\n\n- [variables](https://www.khanacademy.org/math/algebra/x2f8bb11595b61c86:foundation-algebra/x2f8bb11595b61c86:intro-variables/v/what-is-a-variable), [coefficients](https://www.khanacademy.org/math/cc-sixth-grade-math/cc-6th-equivalent-exp/cc-6th-parts-of-expressions/v/expression-terms-factors-and-coefficients), and [functions](https://www.khanacademy.org/math/algebra-home/alg-functions)\n- [linear equations](https://wikipedia.org/wiki/Linear_equation) such as \\\\(y = b + w_1x_1 + w_2x_2\\\\)\n- [logarithms](https://wikipedia.org/wiki/Logarithm), and logarithmic equations such as \\\\(y = ln(1+ e\\^z)\\\\)\n- [sigmoid function](https://wikipedia.org/wiki/Sigmoid_function)\n\n### Linear algebra\n\n- [tensor and tensor rank](https://www.tensorflow.org/guide/tensor)\n- [matrix multiplication](https://wikipedia.org/wiki/Matrix_multiplication)\n\n### Trigonometry\n\n- [tanh](https://reference.wolfram.com/language/ref/Tanh.html) (discussed as an [activation function](https://developers.google.com/machine-learning/glossary#activation_function); no prior knowledge needed)\n\n### Statistics\n\n- [mean, median, outliers](https://www.khanacademy.org/math/probability/data-distributions-a1/summarizing-center-distributions/v/mean-median-and-mode), and [standard deviation](https://wikipedia.org/wiki/Standard_deviation)\n- ability to read a [histogram](https://wikipedia.org/wiki/Histogram)\n\n### Calculus (*optional, for advanced topics*)\n\n- concept of a [derivative](https://wikipedia.org/wiki/Derivative) (you won't have to actually calculate derivatives)\n- [gradient](https://www.khanacademy.org/math/multivariable-calculus/multivariable-derivatives/gradient-and-directional-derivatives/v/gradient) or slope\n- [partial derivatives](https://wikipedia.org/wiki/Partial_derivative) (which are closely related to gradients)\n- [chain rule](https://wikipedia.org/wiki/Chain_rule) (for a full understanding of the [backpropagation algorithm](https://developers.google.com/machine-learning/crash-course/backprop-scroll/) for training neural networks)\n\n### Python Programming\n\nThe following Python basics are covered in [The Python Tutorial](https://docs.python.org/3/tutorial/):\n\n- [defining and calling functions](https://docs.python.org/3/tutorial/controlflow.html#defining-functions),\n using positional and [keyword](https://docs.python.org/3/tutorial/controlflow.html#keyword-arguments) parameters\n\n- [dictionaries](https://docs.python.org/3/tutorial/datastructures.html#dictionaries),\n [lists](https://docs.python.org/3/tutorial/introduction.html#lists),\n [sets](https://docs.python.org/3/tutorial/datastructures.html#sets) (creating, accessing, and iterating)\n\n- [`for` loops](https://docs.python.org/3/tutorial/controlflow.html#for-statements),\n `for` loops with multiple iterator variables (e.g., `for a, b in [(1,2), (3,4)]`)\n\n- [`if/else` conditional blocks](https://docs.python.org/3/tutorial/controlflow.html#if-statements) and\n [conditional expressions](https://docs.python.org/2.5/whatsnew/pep-308.html)\n\n- [string formatting](https://docs.python.org/3/tutorial/inputoutput.html#old-string-formatting)\n (e.g., `'%.2f' % 3.14`)\n\n- variables, assignment, [basic data types](https://docs.python.org/3/tutorial/introduction.html#using-python-as-a-calculator)\n (`int`, `float`, `bool`, `str`)\n\nA few of the programming exercises use the following more advanced\nPython concept:\n\n- [list comprehensions](https://docs.python.org/3/tutorial/datastructures.html#list-comprehensions)\n\n### Bash Terminal and Cloud Console\n\nTo run the programming exercises on your local machine or in a cloud console,\nyou should be comfortable working on the command line:\n\n- [Bash Reference Manual](https://tiswww.case.edu/php/chet/bash/bashref.html)\n- [Bash Cheatsheet](https://github.com/LeCoupa/awesome-cheatsheets/blob/master/languages/bash.sh)\n- [Learn Shell](http://www.learnshell.org/)\n\n[Help Center](https://support.google.com/machinelearningeducation)"]]
