Skip to content

Latest commit

 

History

History
43 lines (24 loc) · 4.2 KB

README.md

File metadata and controls

43 lines (24 loc) · 4.2 KB

更多最新的AI开发学习交流信息,请访问华为云AI开发者社区:AI Gallery

如果您访问github较慢,可以访问gitee代码库地址,内容完全一致:https://gitee.com/ModelArts/ModelArts-Lab


ModelArts-Lab

基于华为云ModelArts平台的示例代码库。

此库提供的案例包括代码以及数据集均只能用于学习交流,案例数据集来自于开源社区,所有案例及数据集请勿用于除学习之外的其他商业用途,如果违反所产生的法律纠纷与提供方无关。

Online Documents

案例内容介绍 Introduction of Cases

自动学习案例 ExeML Cases

AI要规模化走进各行各业,必须要降低AI模型开发难度和门槛。当前仅少数算法工程师和研究员掌握AI的开发和调优能力,并且大多数算法工程师仅掌握算法原型开发能力,缺少相关的原型到真正产品化、工程化的能力。而对于大多数业务开发者来说,更是不具备AI算法的开发和参数调优能力。这导致大多数企业都不具备AI开发能力。

ModelArts自动学习让零AI基础的业务开发者可快速完成模型的训练和部署。依据开发者提供的标注数据及选择的场景,无需任何代码开发,自动生成满足用户精度要求的模型。可支持图片分类、物体检测、预测分析、声音分类场景。可根据最终部署环境和开发者需求的推理速度,自动调优并生成满足要求的模型。

In order to improve the Industry AI process , AI must reduce the difficulty and threshold of AI model development. At present, only a few algorithm engineers and researchers master the development and tuning capabilities of AI, and most algorithm engineers only master the algorithm prototyping capabilities, lacking the relevant prototype to real product and engineering capabilities. For most developers, the development and parameter tuning capabilities of the AI algorithm are not available. This has resulted in most companies not having AI development capabilities.

ModelArts ExeML to enable zero AI-based developers to quickly complete training and deployment of models. According to the annotation data provided by the developer and the selected scenario, the model that meets the user's accuracy requirements is automatically generated without any code development. Supports image classification, object detection, predictive analysis, and sound classification scenes. The model can be automatically tuned and generated to meet the requirements based on the speed of the final deployment environment and the speed of the developer's needs.

Notebook案例 Notebook Cases

ModelArts集成了Jupyter Notebook,可为AI开发者提供在线的交互式开发调试工具。开发者通过创建开发环境,可以自行编写和调测模型训练代码,然后基于该代码进行模型的训练。

ModelArts Notebook开发环境非常适合作为人工智能教学和学习的工具,当前已有多个知名教育机构基于ModelArts开设人工智能专业课程。

ModelArts integrates Jupyter Notebook(an open source tools) which provide an interactive development tools for AI developers on browser. After creating a development environment, developers can write and run model training code by themselves, and then train the model based on the code.

The ModelArts Notebook development environment is ideally suited as a tool for teaching and learning artificial intelligence. There are many well-known educational institutions that offer artificial intelligence courses based on ModelArts.

综合AI开发案例 Train_inference Cases

在ModelArts平台完成端到端从数据准备、模型开发、模型训练、模型部署发布、模型共享(AI市场)等全流程的人工智能模型开发,及应用实践。 End-to-end data development, model development, model training, model deployment and release, model sharing (AI market) and other full-process artificial intelligence model development, and application practices In ModelArts.