Upgrades to AI software ease machine learning development
Upgrades to NanoEdge AI Studio and STM32Cube.AI software tools by STMicroelectronics accelerate embedded AI and machine learning (ML) development projects, said the company.
Both NanoEdge AI Studio and STM32Cube.AI facilitate moving AI and ML to the edge of an application. At the edge, AI / ML delivers advantages such as privacy by design, deterministic and real time response, greater reliability and lower power consumption.
NanoEdge AI Studio is an automated ML tool for applications that does not require the development of neural networks. It is used with STM32 microcontrollers and MEMS sensors that include ST’s embedded intelligent sensor processing unit (ISPU). For neural networks, STM32Cube.AI is an AI model optimiser and compiler for the STM32.
NanoEdge AI Studio version 3.2 now contains an automatic data logger generator that increases development productivity. Its inputs include the ST development board and developer-defined sensor parameters, such as data rate, range, sample size, and number of axes. NanoEdge AI Studio uses these to produce the binary for the development board without the developer having to write any code.
Data manipulation features in NanoEdge AI Studio allow the user to clean and optimise the captured data in the NanoEdge AI Studio in a few clicks, said ST. A new validation stage has been added, which helps users assess their algorithms by showing inference time, memory usage and common performance metrics, such as accuracy, and F1-Score. It also highlights information about the pre-processing and ML model involved in the selected library.
In addition to adding more pre-processing techniques and ML models for anomaly detection and regression algorithms to boost performance, the tool supports creation of smart libraries that can predict future system states using multi-order regression models.
STM32Cube.AI version 7.3 is integrated into the STM32 ecosystem and enables conversion of pre-trained neural networks into optimised C code for the STM32 family of 32-bit Arm Cortex-core microcontrollers. The enhanced STM32Cube.AI adds greater flexibility for neural network (NN) optimisations, added ST. It can adapt existing neural networks to achieve performance demands, fit within memory limitations, or balance both. The update also brings support for TensorFlow 2.10 models and new kernel performance improvements.