用于提升产品智能的自动机器学习工具

利用NanoEdge AI Studio找到最适合您的嵌入式项目的AI库并进行配置。
以最低的投入,将具有器件上学习功能的机器学习模型集成到您的STM32微控制器或智能传感器中。

New in version 3.4


Sampling Finder tool
Feature extraction and selection
Addition of a validation stage

简单易学

NanoEdge AI Studio通过简单的步骤指导您创建异常检测、分类或回归库。 按照分步说明来收集、验证数据,并生成要集成到您项目中的C代码。

1
Project settings
2
Signals
3
Benchmark
4
Emulator
5
Validation
6
Deployment
Select your hardware, the type of signal, and the memory you want to allocate to your project. Optimize the settings to generate your machine learning libraries.
Upload from files, use the embedded datalogger or stream the signal you need for your project through USB.
Launch the automatic benchmark to find the right combinations of signal preprocessing, Machine Leaning algorithms and hyper-parameterization for your project. Then choose the library among a pre-selection based on the most suitable level of accuracy, confidence, and memory footprints.
Evaluate the libraries you generated directly on the studio with data streamed in real time.
Generate detailed reports of your AI libraries to compare them with further tests, and select the one you wish to embed in your final product.
Generate optimized C-code for STM32 MCUs and embed the AI library directly in your project in a few clicks. You’re ready to go!

从想法到数据记录,短短数分钟即可完成

捕获数据对于启动项目非常重要。 这是复杂、耗时的一步,因此NanoEdge AI Studio中加入了嵌入式数据记录功能。
NanoEdge AI Studio的数据记录功能可帮助您采集和管理STWIN开发板上的高速数据。 您无需编写任何代码即可处理您的工业级传感器! 只需将您的板件连接到串行端口,启动NanoEdge AI Studio,即可开始。

开始使用NanoEdge AI Studio

探索如何通过点击几下鼠标即可创建数据集以及创建您的第一个机器学习算法