[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-07。"],[],[],null,["# Vertex AI tutorials\n\nThe following tutorials step\nthrough the various stages of a Vertex AI Neural Architecture Search run. To help\nyou integrate your docker with the Neural Architecture Search service,\nwe encourage you to *run* these tutorials, not only read through\nthem. This process\nwill then help you to integrate your own docker with the\nNeural Architecture Search service.\n\nThese tutorials are lightweight and only require a few cloud CPUs\nto run, which your project should have by default without\nany additional quota needs.\n\n1. [Run baseline training on Google Cloud using your docker](https://github.com/google/vertex-ai-nas/blob/main/third_party/tutorial/vertex_tutorial1.md).\n2. [Create search spaces](https://github.com/google/vertex-ai-nas/blob/main/third_party/tutorial/vertex_tutorial2.md).\n3. [Run an architecture search on Google Cloud](https://github.com/google/vertex-ai-nas/blob/main/third_party/tutorial/vertex_tutorial3.md).\n4. [Add latency constraint to search: FLOPS or device based](https://github.com/google/vertex-ai-nas/blob/main/third_party/tutorial/vertex_tutorial4.md).\n\nAfter running the tutorials, you should read the\n[best practices and suggested workflow](/vertex-ai/docs/training/neural-architecture-search/suggested-workflow)\nbefore running the first Neural Architecture Search search."]]