[null,null,["上次更新時間:2025-07-27 (世界標準時間)。"],[[["\u003cp\u003eAutoML automates tasks in the machine learning workflow, like feature engineering, algorithm selection, and hyperparameter tuning, making model building faster and easier.\u003c/p\u003e\n"],["\u003cp\u003eWhile manual training involves writing code and iteratively adjusting it, AutoML reduces repetitive work and the need for specialized skills.\u003c/p\u003e\n"],["\u003cp\u003eAutoML empowers users to focus on the core machine learning problem and data instead of getting bogged down in manual tasks within the model development cycle.\u003c/p\u003e\n"],["\u003cp\u003eThis module explores the benefits and limitations of using AutoML, common patterns, and how to apply them to machine learning projects, assuming prior knowledge of basic machine learning concepts.\u003c/p\u003e\n"]]],[],null,["| **Estimated module length:** 30 minutes\n| **Learning objectives**\n|\n| - Automate tasks in a machine learning workflow.\n| - Determine the benefits and limitations of using AutoML with your machine learning model.\n| - Enumerate the common AutoML patterns and apply them to your ML projects.\n| **Prerequisites:**\n|\n| This module assumes you are familiar with the concepts covered in the\n| following modules:\n|\n| - [Introduction to Machine Learning](/machine-learning/intro-to-ml)\n| - [Linear regression](/machine-learning/crash-course/linear-regression)\n| - [Working with numerical data](/machine-learning/crash-course/numerical-data)\n| - [Working with categorical data](/machine-learning/crash-course/categorical-data)\n| - [Datasets, generalization, and overfitting](/machine-learning/crash-course/overfitting)\n\nIf you are starting a new machine learning (ML) project, you may be wondering\nif manual training is your only option to build a machine learning model. With\nmanual training, you write code using an ML framework to create a model. During\nthis process, you choose which algorithms to explore and iteratively tune\nhyperparameters to find the right model.\n\nOf course, model training is not the only thing you need to think about. In\npractice, building a machine learning model from prototype to production\ninvolves repetitive tasks and specialized skills. A simple exploratory ML\nworkflow looks something like this:\n**Figure 1.** Simple machine learning exploration workflow.\n\n**Repetitive tasks** - The ML workflow can involve repetitive work and\nexperimentation. For example, during model development you typically need to\nexplore different combinations of algorithms and hyperparameters to identify the\nmost appropriate model. With manual training, you write specialized code to\ntrain the model and then adjust the code to run experiments with different ML\nalgorithms and hyperparameters to find the best model. For small or exploratory\nprojects this manual process may not be a problem, but for larger projects these\nrepetitive tasks can be time consuming.\n\n**Specialized Skills** - Manually developing an ML model involves specialized\nskills. In practice, not every team planning to develop a machine learning model\nmay have these skills. If a team does not have a dedicated data scientist, doing\nthis work manually might not even be feasible.\n\nLuckily, certain steps in model development can be automated to reduce the\nburden of repetitive work and the need for specialized skills. Automating these\ntasks is the subject of this module on automated machine learning (AutoML).\n\nWhat is AutoML?\n\n[**AutoML**](/machine-learning/glossary#automl) is a process of automating\ncertain tasks in a machine learning workflow.\nYou can think of AutoML as a set of tools and technologies that make building\nmachine learning models faster and more accessible to a wider group of users.\nThough automation can help throughout the ML workflow, the tasks that are often\nassociated with AutoML are the ones included in the model development cycle\nshown in Figure 1. These repetitive tasks include:\n\n- **Data Engineering**\n - Feature engineering.\n - Feature selection.\n- **Training**\n - Identifying an appropriate ML algorithm.\n - Selecting the best hyperparameters.\n- **Analysis**\n - Evaluating metrics generated during training based on test and validation datasets.\n\nWith AutoML, you can focus on your ML problem and data rather than on feature\nselection, tuning hyperparameters, and choosing the right algorithm.\n| **Key terms:**\n|\n- [AutoML](/machine-learning/glossary#automl) \n[Help Center](https://support.google.com/machinelearningeducation)"]]